|
Privacy Policy
You can now take a tour of posts in the Ask Aaron Archives that have been referenced to answer new questions -- let's call them Less Frequently Asked Questions. Click the 'Mystery Post Tour' button above to get started.
Twice Equals None
Q: I'm moving up from my FingerTech Viper antweight kit to a beetle, but I am having some issues controling my new robot with the Flysky FS-CT6B transmitter from the Viper. I have a Scorpion ESC wired to two motors. When I move my joystick up and down it only moves one motor, and then I move my stick left and right it moves the 2nd motor.
I read somewhere that my transmitter might be in "drone" mode and I need to program it into "robot" mode, but everything was fine with my antweight. Is there an alternative solution?
A: [Mark J.] The problem is that your old transmitter and new ESC are both providing the channel mixing needed for single-stick control:
Different models of Scorpion ESCs have different methods of disabling the MIX option:
The Center Isn't Off
Q: I am trying to make a combat robot for the first time, and with the right joystick in the center resting position it reads as half power and turns the motor on. I have looked across the internet on how to make the resting position read as off and if I go down the motor reverses and forward it goes forward. I’m using the fs-i6s transmitter and a TOYTENSI 20A esc. [Social Media]
A: [Mark J.] Your radio is not the problem - the problem is your ESC.
All the TOYTENSI ESCs I can find are for model airplanes. An airplane ESC is uni-directional with the zero-throttle stick position all the way down. Some aircraft ESCs can be switched to forward/reverse response with center off, but the functions available on the TOYTENSI programming card do not include this option. You need a different ESC.
To keep things simple search for an ESC that specifies 'bidirectional' or 'forward/reverse', like the Repeat Robotics Brushless Drive ESC. ESCs that say 'thrust reversing' are different and will not give you the stick response you're seeking.
Powerful Forces at Work
Q: I've seen some people talk about battle hardening outrunner brushless motors (i.e. Robert Cowan's wonderful guide/video on the topic and others) and I was wondering how many teams/bots (not really looking for a discrete value just an approximate percent) you know of that can attribute the hardening to saved matches or motors? Our team is building a heavyweight, and we don't exactly have the resources to replace a set of smoked motors that overheat because of glue prior to competition. The current team thinking is to assume that the motors are protected enough by the frame and armor that we are more likely to burn them up electrically vs break them mechanically. [Warren, Ohio]
A: [Mark J.] I can't give you a meaningful percentage, because... No one knows when their effort at hardening saved a motor, they only know when it didn't.
I understand and share your concerns about heat retention from encasing stators in a thick layer of epoxy and glass 'microballoons' in the fashion demonstrated by Robert Cowan -- one of few areas in which we disagree. See this archived post for my approach to stator hardening, which maintains a large surface area and aids cooling via improved thermal conduction. A similar approach works well to secure rotor magnets. More epoxy ≠ better.
Battle hardening is not done to protect motors them from direct mechanical damage in the way your frame and armor can. Many builders have difficulty wraping their heads around the magnitude of the acceleration forces encountered in robot combat. The forces generated in abruptly stopping a weapon motor spinning at thousands of RPM can run into hundreds of Gs, as can the forces exerted on the entire weapon or drive motor from the acceleration imposed by a spinner weapon hit to the chassis. That is what you harden against. Comment: Also, as a team that decided on using VESCs like a fellow poster who asked about them the other day, I think they got their hypershock blog posts mixed up with the VESC manual about optoisolators. [Warren, Ohio] Reply: That clears up some of the confusion. Thank you! My Gumby Brain Hurts
Q: Hello! You previously answered someone's question about running 3S drive motors using a 4S battery, and you suggested lowering the power of the ESCs by 25%. Would this work for brushless motors? I know they are more complicated, but would the result be the same?
The role of the [brushless] ESC is to deliver power from the battery to the motor in a controlled manner. If you input 50% throttle on the controller, the ESC will deliver 50% 'power' to the motor. What is 'power' depends on the firmware used. Some will use the average voltage sent to the motor, others use the speed target, and some use a mix of both.
References to a 'speed target' have some people believing that the R/C signal is interpreted by the ESC as a specific motor speed, and that the ESC will pass thru whatever current and voltage is necessary to achieve that target. This is incorrect.
The 'speed target' is a rotational velocity measured as a function of the number of ESC polarity reversal cycles per minute (eRPM). Below this target (about 20% of full mechanical RPM) a sensorless ESC restricts the current available to the motor to provide a smooth start-up. When the eRPM target is reached the ESC switches to plain vanilla pulse width modulation (PWM) and responds to the R/C signal just like a brushed speed controller. Contrary to some forum posts I've seen, a 75% transmitter throttle limit will keep brushless motor voltage and current levels at the 3S voltage level with a 4S battery.
Two Warnings
Like Joining a Cult
Q: Hello! I want to start by saying thanks for all the info you have on your site. It's a wonderful resource.
My question for you is: Do you know anything about how this Trampa 6 Mk VI VESC performs in combat robotics? [Kentucky-Ohio Zone] A: [Mark J.] The Trampa 6 VESC controlling a 6374 motor is a common choice for heavyweight robot drive. In general the Trampa products perform well if you can get them properly set up. See additional comments below. Q: I found a blog post from Hypershock where they mention use of one of the high voltage models and was wondering if you knew of any other resources on the tuning of VESC's for combat robots.
That said, I do know of a combat setup guide that can at least get you started. Note that the guide is a few years old and does not use the current firmware:
Q: Also, in their diagrams for how to wire the VESC, Hypershock is very adamant about the use of optoisolators, but I haven't seen much of any other resources suggest that they be used. Is this an overblown concern? A: I am unable to find the HyperShock blog post you reference. Use of opto-isolators may be specific to their set-up as I have not run across other combat robot posts that have even mentioned optos. If in doubt I would suggest contacting the manufacturer. Long Motor Bad
Q: Hi! I remember hearing that brushless outrunners (especially ones that are longer than they are wide) need their ends supported when used in combat robots. However, I can't seem to find any mention of it now.
The explanation was that, when these motors are face mounted, upon impact the inertia of the heavy can will cause it to wiggle/bend the motor, which can break it right where the stator joins the body. (See my diagram). Is this a real issue? [Cambridge, Massachusetts]
A: [Mark J.] Yes, this is a real issue. The rotor can of an outrunner motor attaches to the motor shaft way out at the end of the motor farthest from the 'face'. The longer the motor, the longer the lever arm that the mass of the rotor acts thru to stress the base of the stator. A good weapon impact can generate V-E-R-Y L-A-R-G-E accelerations.
Q: How does one solve this? Do you just never buy long motors? Do you try and extend the shaft so that you can put a bearing on the far end of the motor? (This seems suboptimal to me because if the chassis deforms and that bearing moves out of line with the axis of rotation of the motor, the whole thing will lock up). Do you add a cylindrical wall just barely larger than the motor so that the amount it can wiggle is very limited?
Inrunners are exempt from this because they have a solid casing that surrounds the entire motor, eliminating the stator weak point.
Thanks for your help.
A: That last bit is the key. An inrunner casing is structural and supports the bearing at the far end of the motor. That is what you must emulate if you want to run a long outrunner
Build a stiff frame around the rotor to hold a bearing that supports the shaft at the end of the motor. This motor frame must be independent from the chassis structure, so it is not subjected to impact deformation and it acts just like the can on an inrunner. Stiff in all directions, please.
Alternative Not Needed
Q: A few years ago I used the VEX 29 motor controllers for my brushed drive motors but they seem to be discontinued. Are there any good alternatives for single brushed motor controllers? [Social Media]
A: [Mark J.] There are alternatives, but you don't need them. The VEX 29 ESCs are still available direct from VEX Robotics and they are still a bargain. The excellent article on VEX 29 modification and use on Jamison Go's website covers all the details. Big But No Bog
Q: What is the maximum useful speed I can achieve with Repeat Robotics Mk2 brushed motors? At 3s, these motors claim 1220 rpm free spin, and a maximum output of 3.7 watts. In a 1lb bot (using two total motors), what is the largest wheel I can use before I bog them down? A: [Mark J.] Popping the motor specs into the Tentacle Drivetrain Calculator says that 3" wheels won't bog the motors. On a 3S LiPo they'll spin the tires at 65% of stall torque. Acceleration is still OK, and top speed side-to-side in an 8 foot arena is 6.7 MPH. Q: My ignorance is showing, but how did you do that? I couldn't find voltage or torque constants to plug into the Tentacle Calculator.
A: Piece of cake. The link you provided says that on a 3S battery the motor specs are:
Torque Constant = stall torque divided by stall amps = 16.5 oz-in / 2 amps = 8.25 oz-in/amp
Too Little Money and Too Far Away
Q: Hi, I am currently making a battle bot and I'm trying to use hoverboard wheels for it. It is a 60lb drum with a $700 budget. I need help with motors and motor control. I'm looking to spend between $150 and $200 on this aspect of the bot. I'm open to changes and I'll take any advice. Thank you!!!
A: [Mark J.] Where are you going to compete with a 60 pound robot? I know of only one tournament that hosts that class, and it's nowhere near New York. Do you have a safe arena to test and compete within?
Given your budget (a good 3-pound robot can cost $700) a pair of hoverboard hubmotors would seem to be a reasonable drive system. They are rarely used in combat robots because they are not made for high acceleration or great pushing power, but a pair of 350-watt hoverboard motors can gently push around a 60-pound 'bot. If you can find the 24 volt 350 watt version wheels you can power them with the BotBitz 30 Amp Brushless ESCs and stay pretty close to your budget.
Another option would be a a pair of converted 18v cordless drill motors and 3" wheels. Performance is better with about 450 watts output per motor and gearing better suited to acceleration. If you stick with 3" wheels suitable R/C speed controllers for them are readily available. Here's a drivetrain analysis of a 60-pound 'bot with two drill motors and 3" wheels done with the
Tentacle Drivetrain Calculator:
Front? Back? Jumper?
Q: I bought the Repeat Robotics Budget Ant DESC. Channel mixing is disabled by default, but can be enabled by "cutting the jumper near the i on the back of the board". I'm not sure which side of the DESC is the "back" and I don't see anything that looks like a "jumper". Can you give me a better description of how to enable mixing on this DESC? [Social Media]
A: [Mark J.] I think this is best described with an illustration. I've put a green box around the 'jumper', which in this case is not a wire but just a solder blob across two metal pads on the circuit board. You will need to trim away part of the clear plastic shrink tube around the circuit board to gain access to the 'jumper', but rather than cutting the jumper you may just 'de-solder' the solder bridge so that it looks like the 'O' jumper below it. I wonder what the 'O' jumper does?
Thanks, Peter
A: [Mark J.] You didn't tell me what ESC you have so I'll give you a generic sketch:
A: [Mark J.] AM32 is an open source firmware for ARM-based ESCs. Unlike BLHeli and other aircraft-oriented firmware AM32 has paid special attention to delivering smooth and plentiful low speed torque. This provides better control and response for robot drive, and quicker spin-up times for brushless weaponry. Drivers report linear torque response that feels just like brushed motors and big improvements in weapon performance.
There are several reliable and battle-proven insect-class ESCs with well-written documentation. I strongly suggest you spend a bit more money and pick something from that group.
That said, some builders do use FPVKing ESCs. One FVPKing ESC has the capacity to independently control two FingerTech SilverSpark motors running at 12 volts. If you get a good one you may make it thru a few fights with it. Pack a couple of spares in your parts box.
Note The FPVKing has two receiver plugs of different styles: a standard 3-wire type (black-red-white wires) and a non-standard single-wire type (yellow wire). The standard plug goes into the CH2 slot in your receiver and the single yellow wire connector is slipped over top 'signal' pin in the CH1 slot (the yellow boxes in the photo above).
A: [Mark J.] Where Kt is the motor's Torque Constant:
The voltage constant for a brushless motors are right in the motor description, but how do I calculate values for the torque constant (Kt) and stall current for a sensorless brushless motor? Doesn't the brushless ESC "soft start" make these specs useless? Is there some estimation process that can give me useable numbers for the calculator? [Akron, Ohio]
A: [Mark J.] You're correct about the 'soft start' feature of brushless ESCs -- there really isn't a 'stall current' figure for brushless motors that can directly plug into the Tentacle Drivetrain Calculator and mesh with the formulas it uses to determine the accelleration performance of the drivetrain. Worse, the internal resistance (Ri) of many brushless motors is so low that it becomes necessary to account for the internal resistance of the battery cells and the ESC to get reasonable torque figures for the drivetrain.
Sometimes the esc beeps when this lag happens almost as though it is resetting itself when the lagging occurs. [emphasis addded]
When we start it slowly, even until max throttle, it works perfectly fine. Even when we use both joysticks at the same time. We tested power to the esc and it’s receiving enough. Right now we’re wondering about switching esc’s to see if the problem is with the esc, motor, or a connection between the motor and ESC.
Is this a common problem or do you have any ideas? Please let me know. Thanks!
A: [Mark J.] The only info you have given about your robot is the ESCs you are using and the number of motors. The hamburger is bad.
The symptom you describe is common when some component in the electrical system cannot keep up with the current demand and the system voltage drops. Slowly accelerating the motors does not draw much current, but quick acceleration requires more torque and more torque equals more current.
That beeping ESC is resetting. Under rising current load its input voltage is dropping below the minimum required to keep its firmware operating and it goes thru an off/on cycle. The problem is likely not caused by the ESC itself: although it may be more sensitive to the voltage drop than the other ESC the actual problem is most likely elsewhere. The cause may be related to one or more of these components:
If you'd like me to attempt to further troubleshoot the problem I'll need to know full specs for all of your components: motors, battery, BEC, gearboxes, radio,
There is an on/off switch harness that the last user did not use. He says that he found a way to do whatever it was doing with the transmitter he used. The on/off switch has only two wires, not 3-pin so I guess it does not plug into the 3-pin connector on an ESC. Seller mentioned where it went in quick discussion but my brain did not record it so now I'm lost on where to put it.
The only place I see a possible two-wire connection is the Brake/Coast jumper. Is this where it goes to turn the ESC off? [Franklin, Ohio]
A: [Mark J.] There is no place to wire an on/off switch on the Victor 883 ESC itself. The brake/coast jumper controls how quickly your motors come to a stop when the ESC gets a stop command from the radio -- keep it on 'brake'. Power is controlled by adding a switch to the positive input wire from the battery (see diagram). You will need a high capacity switch or removable link capable of handling the full current demands of the motors -- which may be substantial.
Smaller class robots can use a simple single pole single throw (SPST) switch of suitable capacity, or you can make a removable link out of a small two-pole connector.
Q: A follow up question on the Bargain Bot. The bot has two 24v motors each attached to their own 24v Victor 883 motor controller. Seller of the bot said I just needed to hook up an RC Receiver and matching/bound transmitter and it would be good to go. Well I've bought two transmitter/receiver sets now.... just tried the second one. Same issue with both (beside poorly translated Chinese manuals): 883's flash that no PWM signal is found. I have hooked each 883 to different channels on receivers and turned on the transmitter before powering up the bot: no change. I reversed the cable with signal wire on different pin: no change. The ESCs always flash orange indicating no PWM signal.
Am I missing a part? Do I need to power the receiver separately? I was thinking it got power from a regulated voltage circuit built into the 883, but could be wrong on that.
A: Some ESCs have a built-in voltage regulator called a Battery Eliminator Circuit (BEC) to power the receiver. The Victor 883s do not have a built-in BEC so you do need to power the receiver separately. If your robot is as old as I think it is the builder may have used a separate receiver battery pack for this purpose. I would suggest adding a 'stand alone' BEC to power the receiver: something like this.
The diagram below shows how a stand-alone BEC is wired in parallel to a generic motor controller. You may plug the BEC output into any unused port on the receiver. You want and need only one BEC for your system.
Q: Do I need a converter of some sort from RC type control signals to PWM signals of these larger motor controllers in bots??
A: The Victor ESCs were originally designed to work with a special control system that had a higher voltage magnitude Pulse Width Modulation (PWM) signal than typically comes from an R/C receiver. For best signal reliability they require an IFI PWM Signal Driver to boost signal voltage. I see from a photo included with your question that signal drivers are already installed on your receiver cables. Keep them in place and you'll be fine.
Q: The BEC you linked to says it is a "Brushless Receiver Servo Power Supply". I think my motors are brushed so should I get a brushed version?
A: You do have brushed motors (they look like Bosch GPA 750s) but there is no difference in BECs for brushed and brushless applications. The BEC just provides 5 volts to the receiver power buss. I suspect that the BEC manufacturer added 'brushless' to the description in hopes of getting a few extra search hits.
Q: I see BECs in 5 volt, 6 volt, and adjustable voltages. Are receivers 5 volts these days?
A: Acceptable receiver voltage ranges vary with manufacturer, but I know of no receivers that cannot handle 4.8 to 6.0 volts. The 6 volt BECs may be preferred by R/C aircraft builders to get a little more speed from their control surface servos. For your purpose either 5 or 6 volts will be just fine.
A: [Mark J.] It's difficult to directly compare brushed and brushless motors, but brushed has certain advantages:
Like any brushed motor, it is important to 'run them in' at low-speed/no-load to allow the brushes to wear a bit and match the curve of the commutator. Failure to do this may result in electrical arcing that can damage the brushes and commutator.
What mechanism should I use to prevent the motor from burning?
Thanks in advance.
A: [Mark J.] Thank you for your condolences, Long Island. The passage of years means little; I think of Aaron every day. Keeping 'Ask Aaron' going is the best therapy I have and I appreciate the on-going support of the community.
Your concern about stalling electric motors is well placed -- particularly for high-performance motors pushed close to their physical limits. A motor receiving power but unable to spin can draw enough current to damage itself as well as its controller and battery. The measures required to protect your weapon motor system will vary a great deal with weight class and weapon type:
A: [Mark J.] A servo is a gearmotor plus an ESC and a sensor that tells the ESC the position of the output shaft. Adding a position sensor with an interface to the ESC effectively turns a gearmotor into a servo, and a web search for "convert motor to servo" will supply instructions and videos for this type of conversion.
But you don't need a servo for your purpose...
You're just looking to cut power to the motor when the output shaft reaches an end point in either direction. There are a few brushed motor controllers that have inputs for limit switches that are placed to cut power when the end of desired motion is reached but still allow power in the other direction to let the motor to return to the start point.
But you don't need limit switches either...
Take a look at this previous post from the Ask Aaron Weapons archive:
A: [Mark J.] You may be surprised, Kansas City. I know multiple veteran teams that DO use hard endstops and DO manually cut power at the ends of travel. More parts equals more failure points, and a failed limit switch can disable a weapon just as effectively as having your opponent rip it off. Simple is good.
I asked the teams you mentioned for details on their weapon arms and got these responses:
Zachary Lytle writes: "[Skorpios] is far less complicated than you might expect. We believe the fewer things you have in the robot to break the better. So the arm is just clutched and it's Diana's job to turn the motor off before it fries." Just Don't Be Sloppy
Q: Hi, I was reading the RioBotz combat robotics tutorial and it said that you could not shorten the wires from a sensorless brushless motor. I was recently watching a video where the builder had shortened the wires of a sensorless brushless motor. The RioBotz tutorial was written a while ago so I figured that it was outdated on this issue but I didn't want to ruin a good motor.
A: [Mark J.] The RioBotz combat robotics tutorial is a fine piece of work. Their warning on shortening the leads on sensorless brushless motors is not outdated, but saying that "you should never shorten brushless motor wires" [RBCRT, page 239] is too harsh.
Hobby sensorless brushless motor controllers rely on monitoring the changes in induction between the electrical paths thru the motor windings as the motor spins. If you haphazardly clip and re-crimp motor leads it is theoretically possible to cause an imbalance in the induction between a couple of these paths that might confuse the controller.
Keep the new lengths of the motor leads equal and pay careful attention to all electrical connections you re-do. You'll be fine.
Royale With Cheese
Q: So, my fingertech speed controllers are receiving power and signal but are not giving power to my motors. Please Help!
A: [Mark J.] Eager to help, but many things could be wrong and you have given me very little information. The Hamburger is Bad.
I'll assume for the moment that you have the controllers wired up correctly, but I need details about your components and how they are behaving:
Two Into One
Q: Can I splice the signal and ground wires of two ESC's (brushless in my case) and have them plug into the same channel in my receiver? I want to be able to drive them both simultaneously, but they don't need to be perfectly in sync with each other. [Parker, Colorado]
A: [Mark J.] It would help to know exactly what you're doing but in general, yes. You can drive two unsynched ESC/motors from a single receiver output port signal. Do not try to drive a single brushless motor with two ESCs. You Took It All Away
Q: I am trying to use a Scorpion Mini ESC in conjunction with a servo powered lifter; both the drive & lifter are powered by a 2S battery. I snipped the BEC jumper because it limits power to the servo to 5v, which I don’t want. However, the ESC won’t talk to my HK-T6A RX now. The ESC blinks meaning it isn’t getting a signal, and the RX doesn’t light up. I have tried recalibrating the ESC, re-binding the RX, and replacing the RX, but nothing changes. The bot DOES work when I put in an un-snipped ESC, which narrows it down.
A: [Mark J.] Cuting the BEC jumper on your Scorpion ESC does not allow full battery voltage to flow to the receiver -- it removes ALL power output from the ESC via the three-wire receiver cables. Your receiver is no longer getting any power at all.
You have two options to supply full battery voltage to the servo:
Already Stretched Tight
Q: With brushed motors, you're typically able to overvolt them 25%-50% without issue (in the short term at least). Does a similar rule of thumb apply for brushless motors? [Roseville, California]
A: [Mark J.] No! See Frequently Asked Questions #25.
You can overvolt consumer-grade motors, but hobby motors designed for model aircraft -- brushed or brushless -- are already pushed to the edge of thermal failure on voltage and current. The same applies to larger high performance motors like the combat-rated Magmotors.
Not Current Limiting
Q: I'm running a 2212 980kv brushless weapon motor with a 35 amp ESC on a 4S lipo battery in my beetle spinner. Performance is good, but I burn out one or sometimes two motors at each tournament I've entered. Should I be using a lower amperage ESC to prevent motor failures? [Ventura, California]
A: [Mark J.] Decreasing the amperage rating of your small hobby-grade ESC will not improve the longevity of your weapon motor, but it may cause early ESC failure.
Aircraft Specs Mislead
Q: I've choosen a Brushless Turnigy D2830-11 1000kv for my weapon motor and I'm trying to find how much current it draws so I can choose a battery. There's no stall current given, but they list a Max Motor Current of 21 amps. Is that the same as stall current? If not, what current value do I need to consider when picking my battery?
A: [Mark J.] 'Max current' for a hobby brushless motor is not the same as 'stall current', and neither specification will be much help in battery selection.
See also: Combat robot brushless motor selection.
Big Boy Pants
Q: How does High Frequency Injection (HFI) work? I know it's supposed to help Brushless motors spin at low RPM but that's about it. A: [Mark J.] I'll keep this as user-friendly as possible, but you'd best put on your Big Boy Electrical Engineering Pants because it gets pretty deep pretty fast.
Brushless motor controllers have two tasks to perform:
If you'd like to geek-out on details you can start here.
Unsensored brushless motors from model aircraft have no Hall effect sensors. Their motor controllers must estimate the position of the rotor by monitoring the electrical characteristics of the field coils themselves. The common method for doing this takes advantage of the spinning PMDC motor also acting as a generator, creating a voltage potential opposing the power supplied to the motor. This opposing voltage is called Counter-Electromotive Force or "Back EMF" for short. The brushless controller monitors the polarity and strength of the Back EMF on each of the three motor leads and calculates the rotor position from this data. The problem is at low motor speeds Back EMF is very weak and difficult to accurately read. At zero speed Back EMF completely disappears. The controller must restrict current to the motor when starting rotation and at low speeds while it guesses at rotor position -- and that means poor low-speed torque. High Frequency Injection is a relatively new method of determining rotor position by sending brief high-frequency electric pulses thru the motor leads and comparing the electrical impedance found in each of those leads. The impedance will vary with the magnetic flux in the stator cores, and from that information the controller software can accurately estimate the rotor position -- even at zero speed. With more accurate low-speed position sensing the controller firmware can confidently supply greater current to the motor for better low-speed torque. Current versions of the VESC motor controllers incorporate HFI in their firmware. A Bridge Too Far
Q: I have a problem with my Endbots DESC (that burnt it out). I had to calibrate it so I bridged the two calibration pads (CAL), and it sparked and burnt out. Did I do something wrong? Thanks.
A: [Mark J.] It looks like the soldering job you did on the + power input pad created a solder bridge to one of the surface-mount components next to that pad. I suspect that bypassed the voltage regulator and allowed full battery voltage to leak into the DESC circuits. Too much voltage to the calibration pad: POOF! Small circuit boards require extra care in soldering. Always examine your completed joints carefully to assure that the connection is solid and that you have not created any unwanted electrical paths with excess solder. Rename: Whiny Boy
Q: I accidentally dropped a brushless motor from a height of about 4 feet on the floor. It still runs fine, but it makes a different sound than it used to. It sounds more whiny than it did before, but I don't think it's spinning any slower than it did before I dropped it. My guess is that it's off balance.
What do you think went wrong, and is there anyway to fix it?
A: [Mark J.] I've been getting a lot of questions lately that fit the same template:
So... you dropped an unspecified brushless motor onto the floor and now it's 'whinier' but runs fine? I'd consider this an opportunity! Rename your robot 'Whiny Boy' and embrace the change. Oh, and buy a spare motor to have on hand when the one you dropped fails, cause you probably:
Round and Round
Q: With brushed motors, the output RPM scales roughly proportional to the input voltage. Does this concept apply to brushless motors? If not, then how do they relate? [Sacramento, California]
A: [Mark J.] Both brushed and brushless direct current electric motors have a 'speed constant' represented by the term 'Kv'. This term specifies the maximum no-load speed of the motor in RPM per applied volt.
Hobby brushless motors often include the Kv in the motor name, like
Obey the Law of O
Q: Adding on to my question above on brushless motors and input voltage - does the torque increase with a higher input voltage as well? My basic understanding of electronics tells me that as voltage goes up, current goes down (assuming equal load resistance). But I don't know how that concept applies when dealing with brushless. [Sacramento, California]
Four minutes later... Q: Disregard my previous question. I read your guide to brushless motors - guess I should have started there huh - and I see the answer isn't going to be as straightforward as I thought. Thanks for your help regardless! A: [Mark J.] No problem, Sacramento. You are correct that the answer to your question is a bit complex, but I've wanted to take a shot at an understandable answer to this question for some time. I'm not going to pass up this opportunity!
Motor torque is proportional to current: more current equals more torque. Ohm's law states that the current through a fixed resistive load is directly proportional to the applied voltage:
Current = Voltage ÷ Resistance
Amps = Volts ÷ O 
So for a fixed resistance, current increases with voltage. When stalled (zero RPM) a PMDC motor acts as a simple resistive load, so increasing current by increasing voltage results in increased torque.
When the motor starts to rotate things get more complicated. A spinning motor also acts like a generator that creates a voltage potential acting against the applied voltage. This is called counter-electromotive force or 'Back EMF'. The faster the motor spins, the greater the magnitude of this force -- which has the same effect on current as increased resistance. The result is that the current (and torque) decrease with increasing RPM. 
With the above for background, I can answer your question. Since motor torque is proportional to current and current is proportional to voltage, a PMDC motor (brushed or brushless) operating at increased voltage will produce increased torque at any given RPM. The chart at right is for brushed motors -- brushless motors have non-linear torque curves, but the increase in torque with increased voltage is the same.
Also consider: Current drawn by an electric motor is proportional to the load placed on the motor, irrespective of voltage. For a given torque load a non-stalled motor will draw a specific current regardless of the voltage applied. More voltage applied to a motor with a constant torque load will increase RPM, but the current draw from the motor will remain ~constant. The "Down Under" Solution
Q: The BotBitz website recommends pairing their 30A Brushless ESC to a 4258, a motor that calls for an 80A ESC to operate. Am I missing something here or does the motor not require that much current? [Yuba City, California]
A: [Mark J.] BotBitz recommends an 'off-the-shelf' featherweight combat robot drive solution that includes:
Brushless drive motors in combat robots are oversized to assure their survivability in an environment well outside their intended use; see the 'Drive Motor' section of our Brushless Motor Selection Guide. As a result, the motors are loaded far lighter than they would be in aircraft, and lighter loading equals a lower current requirement.
Q: Hello Mark,
For brushed motors there seems to be some sort of vaguely standard numeric nomenclature to the motor sizes such as n10, n20, n30, 130, 380, 550, 555, 775, etc... What is the meaning behind these numbers? I searched for charts that would relate the motor "number" to length, diameter, or something useful; but was unable to find any. A: [Mark J.] It would be nice if these things made sense, wouldn't it? Unlike the numbers used to describe hobby-grade brushless motors, small brushed motor codes reference only catalog numbers used by the original manufacturer. A '550' motor has specific 'frame' dimensions as a set, but the number '550' does not refer to specific motor dimensions. For example: '550' motors from different manufacturers will have the same 'can' dimensions, shaft diameter, and mounting hole sizes/locations. They will fit the same mounts and space, but performance may vary greatly. Q: Alex Hattori of my fav Team Uppercut mentioned building his own electric motors using automotive grade parts. I've heard of folks adding epoxy for strength but does anyone know of any resources for a beginner to even start learning this process of custom motor building? It seems to have been a mega advantage for them. [Social Media]
What advantage do you hope to gain by building your own motor? The available motors are quite effective and represent hundreds of hours of design, prototyping, testing, and refinement by engineers specializing in the field. Alex Hitori is in the senior year of a mechanical engineering program at MIT with access to the design and construction facilities at the University. Are you prepared to invest the required time and resources?
A few builders take existing motors, strip out the key components, and reassemble them in more robust housings with larger shafts and improved bearing support. This is a more reasonable approach than starting from scratch. You can find an example of such a reconstruction in this archived Ask Aaron post.
Q: Hi Mark. You might remember me from a few months ago when I asked you to diagnose some of the gremlins we were having with the Scorpion Mini Esc in my beetle. We ended up sending it back to Robot Power and they supposedly repaired it, but now we have a different issue. None of the lights on the esc will come on when we have it connected to a battery, but when we connect it to the receiver one of the chips gets very hot, very fast. Is this just a bad board? Or is there something else we might be missing that could help to fix it? I really like the scorpion mini and I'd like to think this is just a bad board among a field of otherwise very robust ESC's.
Thanks!
A: [Mark J.] I remember you, Baton Rouge. Sorry to hear that you're continuing to have trouble with the usually-reliable Scorpion ESC. Maybe you have the exception that proves the rule. It's difficult enough to troubleshoot an electronics problem with the board on the bench in front of me. Diagnosing a problem based on limited info and the board in an unknown circuit is effectively impossible, but I'll take a stab at it.
Q: I saw a claim on [social media] that heavyweight 'Gruff' uses a combination of brushed and brushless motors in its drive train. Is that true? What benefit do they get? What motors are they using? How are the motors linked? [The Hinterlands]
A: [Mark J.] Yes, the 2020 version of 'Gruff' does use one brushed and one brushless motor per side for the drivetrain in a system they call "brushed-more-or-less". The team needed to trim some weight to upgrade their already massive torches, so they swapped out the four 'long can' brushed AmpFlow motors for a pair of 'short can' Magmotors plus a pair of '58113' Leopard Hobby brushless inrunners. Drive power went up from about 13 kW to more than 21 kW and they saved a net 12 pounds for the flame torches.
Why mix brushed and brushless? Unsensored brushless motors like the Leopard often have poor low-speed torque and may not respond crisply in low-speed maneuvers. The small brushed motors have have great torque at low speed for predicatable driving response, and the brushless motors supply huge power in the mid-range. It's a clever way to add brushless power without a complete re-design.
Q: Hi Mark. My name is Jessica and i've just been through your combat guide for the FlySky FS-I6 Radio. I do have a question about an issue i'm experiencing, that i couldn't find anything in the guide about unfortunately. I also couldn't find anything that says: "don't email me your crap or i will kill you!" ^^ so here i am. :-P If you don't mind ... I built a speed tank, from scratch (made from steel, aluminum and 3d printed parts), powered by identical sets of brand new Traxxas XL-5 ESC's and brand new Titan 550 12T motors with 2 identical 2S LiPos. I'm using the FS-I6 with it and got your 'Quick and Dirty Combat Setup' done... but for some weird reason, the left throttle runs way faster than the right throttle. I factory reset the thing, finished the binding and finished the combat setup again and it's still running differently. Motors, ESC's and Batteries are new and haven't been running in anything before. Do you have any ideas on what causes that? I really have no clue what could cause that or how it could be aligned or if it indeed is the transmitter. - Jessie [Direct Email] 
Jessie and I swapped emails to pin down a few details about the problem with her 'Speed Tank'. I didn't know they were a thing! Turns out you can download and print something similar from Thingverse.
I suspected that the Titan 550 12T motors had a timing advance that caused them to spin faster in one direction than the other, but Jessie's drivetrain is set up so that both motors spin the same direction to move the tank forward. Something else was wrong...
Response: Your trouble shooting guide is awesome, i do have a bad motor. I guess i will go to Hobby Town and buy a really new motor today. hrhrhrhr
Thank you so much, you're amazing.
Q: Hello, I noticed you recently updated the spinner calculation spreadsheet, (an extremely useful tool by the way), and I'm trying to find some stored energy values for a weapon driven by multiple BLDC motors. It's fairly simple to figure out how to plug stuff in for one BLDC, but I'm having a hard time understanding what values need to change when you are considering a weapon system with multiple motors. My initial thoughts were that it's based on whether they're wired in series or parallel, but I started getting some strange values that didn't make sense. How do I go about doing this? [Hartford, Connecticut]
A: [Mark J.] Thank you for the compliment on the Run Amok Excel Spinner Spreadsheet. To prevent a scramble from users to get the 'recent update' I will mention that the current 'Version 19' series was introduced in late 2017 with the most recent minor update (19e) in December of 2019. You need to check in more often, Hartford.
Short Answer Assuming that the motors are identical: enter the motor specs for a single motor, then divide the internal resistance 'Ri' by the number of motors used to power the weapon. For the default spreadsheet motor values shown below, if using two motors the 'Ri' value of 22 mΩ would be replaced by
Concerns You mentioned a couple things that are commonly done with brushed motors that can cause serious complications if attempted with brushless motors:
Q: Hello, Hartford guy again. I appreciate your response to my question about the motors in parallel. I figured that it'd be unwise to run more than 1 BLDC per controller (given how picky the system is with just one of the things) but I'm glad I asked- better safe than sorry.
I'm curious why you wouldn't also change the number of magnet poles when you add additional BLDC's in parallel, though?
I'm still taking my first real class on magnetism so a lot of this stuff is relatively new to me, so apologies in advance if this is a silly question. I've been getting a lot of whacky values for power and stored energy, and this confusion might be the root of it- but maybe I'm wrong and this has more to do with the messy nature of trying to calculate spin-up with laughably over specced 'stall current' from hobby-grade motors and ESC's than anything : /
A: I've never had anyone get curious about the relationship between the number of magnetic poles in a brushless motor and the motor's performance until now. You may be sorry you asked. Here we go:
There is an earlier post on eRPM and power switchover in the Ask Aaron Weapons Archive that may add some clarity.
Given your interest in the details of spin-up time calculations, you may be interested in the recently added page: Ask Aaron: Estimating Weapon Spin-up Time. Q: I’m trying to use two ragebridge v2 escs to control two large motors and can’t for life of me remember wiring and how to set the mix and combine options. The manual doesn't explain how to use two. Any help appreciated. [Barnsley, England]
A: [Mark J.] Bridging the 'COMBINE' jumper on a RageBridge ESC converts it into a single-channel speed controller with twice the current handling capacity is has as as two-channel controller: up to 150 amps for 30 seconds.
Q: Thanks. I think im there but id like to use the ragebridge onboard mixing. atm I got each motor fwd/rev but one at a time on each channel???
A: No can do. Turning the RageBridge into a single-channel speed controller negates the possibility of using the native on-board mixing. From the RageBridge V2 Manual:
Q: Apologies in advance if the hamburger is bad here but how would I go about mounting a brushless outrunner to a P61? I've only used 300 series motors before and the installation was pretty straightforward. Im not looking for specific help but more so a jumping off point in where to begin, thanks. [Los Angeles, California]
You might also find Robert Cowan's video on preparing outrunner motors for robot combat duty useful: Battle Hardening Outrunner Motors
Mounting the brushless motor is the easy part. Setting up the brushless ESCs is where it gets tough. I'm sure glad you didn't ask about that...
Q: I am continuing to go through your website and others as part of my research and I'd like to get some input from you on an observation I've made. For ant/beetles that make use of brushless motors, it seems like many people are using ESCs and motors designed for quad copters. I have a background in quad copters as well as 1/10 and 1/8 scale RC racing buggies. From my experience, the brushess motors used in RC cars (and boats) are a lot more durable (i.e typically larger shaft bearings, shafts, etc) and and can take a lot of hard running. My question is why aren't people using brushless motors and ESCs designed for RC cars in their combat robots? These certainly do weigh more, but I'd imagine the durability and power makes up for it no? I feel like I'm missing something here. [Arlington, Virginia]
A: [Mark J.] Have a read thru the Ask Aaron Brushless Motor Selection Guide. Briefly:
Q: Hi! My team is building a 15-lb bot. We are going with a kiwi drive holonomic setup and a drum spinner, all powered by brushless motors. 
We have calculated that we can use Propdrive 2836 2200kv outrunners with 20:1 gearboxes for our drivetrain, and the same motor with a 2:1 belt drive for the spinner. Since all 4 motors are the same, could we run them off one of the 4 in 1 quadcopter ESCs? (With the requisite firmware flashing, control board, etc.) Something like this. These ESCs are designed for quadcopter use where there is plenty of airflow, but we figure we can manage heat by mounting it to our aluminum chassis with a thermal pad.
Thanks for your help. [Cambridge Massachusetts]
A: [Mark J.] Massive wheels... Tiny drum... But you only asked about the ESC, so let's talk about that.
Yes, you can run all the motors from a single 4-in-1 quadcopter ESC, and the motors don't even need to all be alike. I know of several small 'bots running two drive motors and a dissimilar weapon motor from a single compact quad ESC.
NOTE Quadcopter ESCs in general do not use the common 'PWM' receiver output protocol with one three-wire connector per radio channel. The specific ESC you are considering uses the 'DShot' serial protocol, so you will need a receiver with that type of output. Check the requirements of any quad ESC before you proceed.
Update - I asked some builders familiar with Quad ESCs about your choice. It seems the APD f-series ESCs may not be programmable for reverse operation. Another builder suggested the Racerstar ReachUP 100A, but I don't have confirmation on usability.
Q: I give up. How does it work? [Multiple Requests]
A: It's called a Killough Platform -- similar to an omni-wheel in action, but different in structure. The two wheels in each of three cradles are connected by gears to each other and to a drive motor which can rotate the wheels while they remain oriented at 90 degrees to each other. One of the pair of wheels is always in contact with the floor as they rotate and 'walk' the platform along. The wheels remain free to spin on their own axles and roll sideways to comply with motion imparted by the other two cradles. You can see the action clearly in this video.
Q: I'm having an issue with my scorpion esc recently to where it will power a motor off of one terminal, but not the other. We are running it off a 3s lipo with the kitbots motors, and in the past it has worked fine. But recently it has had issues running a motor from one of the terminals. The lights on the scorpion mini esc board will generally light up red or green based on which direction the motor is spinning. For the terminal that's working properly, everything is fine. However, when you connect a motor to the terminal that isn't working, then there are some issues. If you don't connect anything to the problematic terminal and push the sticks back and forth, the lights will turn green and red (as expected, this is normal). However, the second you connect your motor to the problematic terminal, the red and green indicator lights become extremely dim and there's no output at all. We know it isn't a motor issue because we swapped motors and terminals, and the issue is localized to the one terminal on the esc regardless of which motor is used. You can connect the motor to the terminal, push forward on the sticks, and then see the light very dimly come on with no response from the motor. And then the second you disconnect the motor, still pushing forward on the sticks, the light shines at its nominal brightness again. It seems that it's an issue with the ESC. But is this something that can be fixed, or will I have to write off the ESC? [Louisiana State University]
If you can't find anything from your visual examination I'd suggest starting over with a fresh ESC.
P.S. -- I hope you aren't trying to build an 12-pound 'bot for the LSU robot competition around the Scorpion Mini and a pair of KitBots motors.
Reply: Haha, no! This is actually a beetleweight build for the LSU combat robotics club here. But I'm surprised you got word of that competition. We actually are designing a robot for that, too... we have every intention of defeating the senior design students. Maybe you can check it out when the time comes. I'm sure it'll be streamed somewhere. Response: I've heard quite a bit from competitors in the LSU 12-pound competition -- both here at 'Ask Aaron' and on other forums. Judging from the questions being asked, the competition is being taken quite seriously. Best luck to you. Q: Has something about the Tentacle Torque Calculator changed? I made some calculations for an antweight drive train a few weeks ago and now the same setup gives me different performance figures. I didn't make any changes. [Hollywood, Florida]
A: [Mark J.] Something has changed, but it only effects the Fingertech 'Silver Spark' and 'Gold Spark' gearmotors. These gearmotors use the Mabuchi FK-050SH motors, and the Calculator had been using the performance figures from the BaneBots version of that motor. A builder on one of the on-line discussions noticed performance figures for the FK-050SH given on the Fingertech site indicate a different variant of the motor.
I changed the Calculator data numbers for Fingertech gearmotors to mirror the numbers on the Fingertech website. I also adjusted the default voltage up from 6 volts to the commonly used 7.4 volts. The new power figures are about 30% lower than the older BaneBots numbers.
Q: A builder on [social media] recently mentioned using a 2205 2300kv brushless weapon motor in their combat antweight. Was that a typo? Does anyone make an outrunner only 5mm thick? [Akron, Ohio]
A: [Mark J.] I understand your confusion. There are two ways to measure a motor for classification:
Q: Brushless drive motors have me really confused. My beetle has brushed motors but I'd like to build a feather and it seems that everybody is using brushless for their feathers. I don't know where to start. Is there a beginner guide someplace? [Pittsburgh, Pennsylvania]
A: [Mark J.] My usual advice for beginners in any aspect of combat robotics is to look at what successful builders are doing and follow their example. That doesn't work very well with brushless drive where the difference between poor and excellent performance relies so much on the unseen ESC firmware and user-modified settings. Unfortunately, both firmware and currently favored settings change quite frequently.
I've pulled together a collection of posts from the Ask Aaron archives that may be of some use in selecting suitable brushless motors for both weapon and drive use. It also contains a complete 'off-the-shelf' solution for featherwight brushless drive that requires no modification:
Q: I've been watching videos about battle hardening brushless outrunner weapon motors. Doesn't globbing a lot of epoxy on the outside of the stator increase heat build-up? How much epoxy do I really need? [Ames, Iowa]
The epoxy that soaked in between the windings now aids cooling via improved conduction, and the wires are bonded together to prevent shifting.
Q: I'm building a featherweight with RS-775 motors and banebots gearboxes. I'm looking for cheap brushed motor controllers rated for 130 amps and 36 volts. Can you help? 
Q: Hey Mark, I want to try building a sensored brushless motor powered lifter. I'm looking at a worm-drive gearbox with an additional belt-drive stage so I can just "park" the lifter with the motor unpowered and not worry about it slipping back down.
The motor I'm considering gives a "30-second Max Current" rating but does not provide a torque constant (Kt). Is there a way to calculate how much torque I can get from the motor at that 30-second current level?
A: [Mark J.] You're very brave to try a brushless lifter, but your plan sounds workable. You don't have the motor Kt, but I'm sure you have the speed constant Kv, so there is an equation that will give you an approximate torque level at a given current draw. I'd dial back the number it gives by about 20% to adjust for 'real world' conditions:
Q: I kind of need some help here. I have a drum weapon beetle and my scorpion mini esc kind of turns off sometimes when it gets hit. We use two 12v polulu motors for the locomotion and a propdrive 1200kv with a flycolor 36A esc for the weapon. The battery is 3s 1000 mAh. We are sure that the problem is not the connections and its not shortening anything.
Even when the locomotion is not working, the weapon still runs nice. When I turn the weapon off and on during the combat, sometimes the locomotion comes back. Does it have anything to do with the electrical noise or some kind of interference in the receiver?
Note I won my bet -- the Lipo jumper was in place. Rio removed it and the problem is resolved.
Q: Is there a difference between the D-Pack gearbox and the Magmotor gearbox?
A: [Mark J.] The 'D-Pack' was an industrial hydraulic pump motor similar in size to the S28-400 Magmotor. In the early 2000's the D-pack was available from surplus dealers for a very reasonable price and found use as both drive and weapon power in a number of combat robots. The mounting face and hole spacing are different for the two motors, so gearboxes for them reflect those differences.
Q: I have a choice of two versions of the same brushless motor to power my weapon: a 750 Kv and an 1100 Kv. The 750 Kv version will give me more torque, right?
A: [Mark J.] There are a lot of 'half-truths' about brushless motors floating around out on the 'net that are confusing robot builders and leading to poor design choices. You can find this one all over the 'net:
There is a kernal of truth in that statement, but it is incomplete and misleading.
The true part -- the motor speed constant (Kv) and torque constant (Kt) are inversely proportional -- if one goes down the other goes up. Compared to a higher Kv version, a lower Kv motor will produce more torque per amp of current.
Here's a real-world example comparing two versions of the AXI 2208 motor:
The incomplete part -- lowering the motor Kv increases the electrical resistance (Ri), which reduces the current the motor will draw. The misleading part -- if allowed to pull unrestricted current, the lower Kv motor version will produce both less torque and less power.
AXI 2208/34 Gold Line V2 motor
AXI 2208/20 Gold Line V2 motor
The 1820 Kv version of this motor generates greater torque than the 1100 Kv version throughout the RPM range. When bogged down to the '60 second' current maximum it produces 8% more torque, and it does so while spinning 85% faster than the lower Kv motor -- a power increase of:
There are a number of reasons why you might choose a lower Kv version of a motor for a specific application, but Q: I'm investigating using brushless drive on a 3-lb Meltybrain spinner and I was wondering if you had some advice on picking motors. I've only used brushed in the past, which is a pretty straightforward process with a torque calculator. However, I can't seem to find any analogous tools or equations for brushless motor torque.
It seems like people have had success with the DYS BE1806 and AX-2810Q, but simply based on the motor characteristics on HobbyKing I don't know what makes them better than others. I know that a lower Kv means higher torque, but how do you know that the Kv is low enough to drive your bot?
A: [Mark J.] You won't find drive train tools analogous to the
Tentacle Torque Calculator for brushless motors because brushless performance depends as much on the controller firmware as on the motor. All the possible combinations are staggering in number and largely theoretical in practice. The way brushless drive is done is to go for overkill and not worry all that much about the details. See this previous discussion in this archive: Brushed vs. Brushless Drive.
There's a "rule of thumb" for brushless motors circulating out on the builder forums that appears to work well for drive motor selection. The rule states that a set of brushless drive motors together weighing between 2% and 3% of the total weight of the robot will have more than enough power and adequate 'thermal mass' to survive. By that rule a pair of DYS-BE1806s is about right for a beetle and twin AX-2810Qs is overkill (which might be just right for a Melty).
What makes a motor better than others? Somebody used it successfully on their robot and word got out. Brushless meltybrain spinner? You're on your own. Best luck.
Q: Hey its me from Anacortes again. First I want to thank you for diagnosing the BEC problem with my ESC. I never would have figured it out on my own. Anyway, now that I've had some time to reflect on the untimely demise on my ESC, and the fact that I'm way over budget, I'm ready to look for a new ESC. Now, I've seen that some builders seem to be able to save space by using two basic ESCs rather than a single dual ESC. Knowing that I will be running four Kitbots 1000 rpm gearmotors on a 4s lipo, and using a BEC, could you please try to point me in the right direction? [Anacortes, Washington]
A: [Mark J.] Given that you weren't using the on-board mixing or BEC on the Scorpion Mini ESC, your replacement can certainly be more 'basic'. Some of the commonly used single channel ESC options won't handle a 4s LiPo battery so that narrows your choices. Take a look at this post in the Ask Aaron 'Ants, Beetles, and Fairies' archive for a list of available options and a discussion of the cheap Chinese solution.
Q: I like the price and simplicity of the VEX 29 ESCs, but will running two motors apiece draw too much current? I don't know much about how current draw works, and I won't be saving any money if they get destroyed.
A: Current draw is dependent on the load placed on the motors, not the number of motors:
You can run this thru the
Tentacle Torque-Amp Calculator for verification. The distributed version of the calculator does not include the Kitbots motor in its pull-down selections, but I use the following motor settings in my personal copy:
Q: I'm reading thorough The Variable Constant's Guide to Vextrollers and I think I get the gist of how to follow his instructions to hack each individual one. Now I have 3 questions about next steps:
A: I have three answers...
If you examine the receiver connector on the VEX you will find the little barbs that hold the connector pins into the plastic holder. With a small probe you can unclip the barbs for the red and black wire pins (video), pull them backwards out of the holder, clip the pins off, and they're ready to connect to the battery. Leave the white wire and pin in the holder and plug it directly into your tinyMixer.
Q: I'd like to use brushless drive motors in my next robot (a featherweight) but the process of finding a suitable ESC, "flashing" new controller firmware, and tinkering with the various settings to get reversing and decent startup torque is putting me off. Then I need to find suitable motors and gearboxes and bodge the whole thing together. Are their any short cuts for a first time brushless hipster wannabe? [Altadena, California]
A: [Mark J.] The process of selecting and modifying components for brushless drive is daunting, but you're in luck! The Australian combat robot supplier BotBitz has been modifying firmware in brushless ESCs for years. They're best known for converting brushless hobby aircraft brushless ESCs to brushed opperation, but they also modify brushless ESCs for robot drivetrain use. Here's their recommended 'off-the-shelf' setup for novice hipsters:
Q: If I tried to use a 40 amp motor controller with a motor that consumes a continuous 50 amps of current under load, which of the following would happen?
A: [Mark J.] The answer depends on the specific motor controller.
Q: Hi Mark, conceptual questions about brushed vs brushless drive motors.
Many Thanks! [Erie, Pennsylvania]
A: [Mark J.] You're asking me to compare an unspecified brushed motor to an unspecified brushless motor. The hamburger is bad.
The decision over brushed vs. brushless power for a combat robot generally comes down to how much pain and frustration the builder is willing to suffer thru to get a brushless motor operating correctly for the intended purpose. For a brushless spinner weapon the level of frustration is low. The brushless motor and controller are performing a function similar to the hobby aircraft purpose for which they were designed. For a small robot spinner weapon it's possible to match up an off-the-shelf brushless ESC with an outrunner brushless motor and you're in business. Brushed spinner motors are very nearly extinct.
For a brushless drive system the level of frustration is much greater. You'll have to locate a brushless controller that either has firmware that can be modified to provide bi-directional control or which will accept replacement of its firmware with something suitable (not a simple task). Then you need to sort thru the firmware settings to set forward/reverse control and take a guess at pushing the start sequence and brake settings as far as possible toward optimizing low speed torque and driveability. If you guess wrong you remove the melted ESC (and maybe the melted motor) from your 'bot and start over. Brushless drive systems are popular and 'hip', but you can still be entirely competitive with brushed drive.
Efficiency - Brushless motors are more efficient than brushed motors -- by a little. However, in practice the use of 'overkill' brushless motors more than erases that small advantage.
Spinner power - Weapon spin-up time is all about motor power all the way thru the RPM range. A brushless motor has less low-end torque performance and would never 'catch up' to the low-torque advantage of a brushed motor with the same peak power. But again, real-world builders go overkill with brushless motors and make this comparison moot.
Keep it spinning - Mid-range motor torque is not what keeps a spinner weapon spinning after a hit. The kinetic energy stored in the weapon is orders of magnitude greater than any difference in motor torque, and mid-range torque would be very close to equal in the 'equal peak power' motors we're comparing here. No 'conceptual' advantage.
I'd advise that you look to how brushed and brushless motors are being used in successful combat robots and use that as your guide. Any conceptual arguement you might make disappears in the sea of real-world design variables.
Q: on the excellent spinner spreadsheet the formula for getting a rough calculation of the stall torque is given as ((1352/Kv)*(Volts/(mOhms/1000)))/141.69 i think i must be mixed up, because it seems like this formula is producing some crazy numbers - so maybe i'm doing something wrong and you can help me out. for example, if you look at the spec of a brushless motor like the Scorpion HKII-7050-330KV and plug in the numbers:
internal resistance in mOhms = 5 Kv = 330 if you use the spreadsheet formula, you get a stall torque of an incredibly powerful 289 N-m! but if you calculate torque using some other formulas that i've learned about, like Kv/Kt where Kt is 9.549/Max Amps or using the formula PeakPower/w, where w is given by (2*pi*Max RPM)/60 then you find that both of those formulas get you a more reasonable number of 8 N-m of torque, which is way different than 289 N-m. so am i misunderstanding something fundamental here about brushless motors and torque? did i mess up the numbers or formulas somehow? am i being dumb? i think i need some help. -josh in salt lake city, utah
A: [Mark J.] Let's start by applying the formula from the spreadsheet to a motor with known specs. The brushed AmpFlow A28-400 will do nicely:
The Scorpion HKII-7050-330KV can deliver 15,000 watts of output power while sucking down 300 amps of current. If you pay $999.99 for a brushless motor you're entitled to 'incredibly powerful' results, but does it actually deliver the 289 N-m of stall torque the spreadsheet formula says it will? Absolutely not, because brushless motors are different than brushed motors in how they operate at low speeds. The concept of 'stall torque' really doesn't even apply to hobby brushless motors. See this archived post about brushless hammer drive for greater detail on low-speed brushless torque.
So, why do the other formulas you cite give a much different torque number?:
I don't think that you're using the most recent version of the Run Amok Excel Spinner Spreadsheet. The current version (19-D) is optimized for brushless motors and has a more sophisticated method for approximating true low-speed torque that takes motor controller software into account. Download it from the Run Amok tools page.
Q: thanks for the explanation and the pointer to the new spinner spreadsheet. great stuff (!!) but i think the new spreadsheet seems to still have a major limitation related to brushless motors, and i think it might be because there is no ability to plug in the maximum amperage (maybe?) and so instead it goes to some unrealistic theoretical maximum which is so unrealistic as to not be useful to model - or maybe i'm screwing something up.
for example, if you take the stats from that fancy scorpion motor:
and say we use a gear reduction of 5 to 1 and then let's say our horizontal spinner is made of steel, 0.5 meters long, 0.125 meters wide, 25mm thick, which is about 12kg (about 27 lbs).
so the results from the calculator are that this blade achieves 3,000 rpm in less than 0.3 seconds (my gut tells me that this is probably off by a factor of 10 or more) and that it will peak at 8,500 amps, which at 50 volts is almost half a megawatt of power.
so...am i doing something wrong, or is this spreadsheet unable to model a big brushless motors like this very well?
thanks again - josh in salt lake city
A: You're planning on buying a $1000 brushless weapon motor capable of 20+ horsepower output and then strangling it with an inadequate power supply? Well, to each their own...
The spreadsheet assumes that the weapon motor will be provided with all the current the brushless controller chooses to pass to it, without voltage sag. There is no adjustment for a fixed 'current maximum' because brushless controllers generally do not have true current limiting, but the spreadsheet is capable of making specific adjustments for changes in the controller firmware that limit control the motor speed at which the 'soft start' feature cuts out and the maximum power pulse percentage during that soft start period.
In SimonK firmware these features are known as 'Commutation Max' (commutation time in u-sec below which 'soft start' restriction is active) and 'Power Max' (maximum power pulse length during 'soft start') but other firmware will typically have similar features. You can 'play' with these settings in the spreadsheet to see their effect on current consumption and performance, then transfer the settings to the controller firmware to implement those power restrictions. Q:How are robo claw 60A solo ESC for the drive 30+ kg robot [Kathmandu, Nepal] A: [Mark J.] ESC selection depends on much more than the weight of your robot. From Frequently Asked Questions #21:
Q: What Electronic Speed Controller (ESC) should I use?
I can comment that the Robo Claw 60A Solo is not widely used in robot combat. Robots requiring ESCs in that current range generally use the RageBridge 2 ESC.
A: The speed controller amperage capacity requirement for your motors will depend on design factors such as:
Once you have settled on motors and drivetrain details, the Team Tentacle Torque & Amp-Hour Calculator will calculate the peak amperage the drive motors can consume with the robot pushing at full throttle against an immovable object [Amps (per motor) to spin wheels]. Use that number to select an ESC with suitable capacity. Give yourself a little extra capacity (~20%) to allow for unexpected conditions. 
Q: This is me again from Anacortes! The Beetleweight wedge I've been building finally has it's electronics! I took your advice and am not going brushless right away, but I have plans to do so in the future. Upon the arrival of the final part, I excitedly botched together a test chassis using some spare 3d printed parts, cardboard, and a lot of hot glue. Everything works after testing using at 2s lipo, but I have run into two problems.
First, the green wire ports on the Scorpion Mini [ESC] I am using are nothing but trouble. I had a lot of trouble getting the wires to stay in as I positioned the motors and drove around, no matter how tight I screwed in the retaining screws. I really don't want to use any solder, as I do not plan to keep this esc in my robot, and want it to be easy to sell or reuse for another project once I switch to brushless. Is there a way to better secure the wires to the esc in a less permanent fashion? My second problem is that I forgot to solder the capacitors onto the Kitbots 1000 rpm gearmotors that I bought. How exactly should they be soldered on? Is there a wrong and right way? I'll admit that I'm not exactly sure how they work, only that you've said that they "make thing easier for your escs." Anyway thank you so much for your time! [Anacortes, Washington]
A: [Mark J.] I'm always happy to get a progress report and I really love the cardboard test chassis! I think I can help you with your problems.
The Scorpion Mini ESC has screw terminals to accept power input and twin motor output wires. A wire slides into the terminal and tightening a screw clamps the wire in place. The problem you're having comes from the stranded wire shifting and loosening after it is clamped in place. You can prevent the wires from loostening by twisting the wire ends, applying a dab of soldering flux, and 'tinning' the wire with a touch of hot solder. Where you have two wires in a single port, solder them together and insert them as a single wire. This will prevent the wires from compressing and shifting when clamped.
The motor capacitors are important! Solder one leg of the capacitor to each of the motor terminals, as shown in the photo. Polarity does not matter - either leg to either terminal. You may want to slide a length of wire insulation over the bare capacitor leads to prevent accidental shorts. An explanation of the need for the capacitors plus some tips on properly 'breaking in' your motors are found in this archived post. Q: Hey guys! Its a little specific and I'm really sorry if its not the kind of question you want to receive :/
Since the first iteration I've swapped to a bigger motor (650W, 50 something-ish max amps) gotten and programmed a bigger 60A Trackstar ESC, have bought a higher 80C rating battery and I have even changed the radio and receiver and upped the wires to 16AWG and yet I still swing with 0.25 to 0.5 seconds of lag when compared to Dain throughout the entire match! It's maddening! I've never been great with electronics and I'm honestly out of ideas as to what causes this. It was the exact same problem with the 30A ESC as the 60A, the same with 2 entirely different radio and the same with a bigger C rating and lower gauge wiring!! I really feel like I've cycled out all of the electrical components and still have the same issue. I would love your opinion on this. Again genuinely sorry if this isn't the kinda question you're looking for on this forum. I've tried to include as much info to avoid the bad hamburger. Have a good one! [Ontario, Canada]
A:
The bad news is that you've put a lot of time and effort trying to fix the electronics when the electronics aren't causing the delayed response. The good news is that there's a little secret about brushless motors that I'm willing to tip. Follow along...
Brushed and brushless motors are very different:
This start sequence works well if the motor has a light load like an airplane propeller, but if the motor has a heavy load (like an overhead hammer) the small current pulse may not cause the motor to move enough to provide feedback to the controller. The controller will wait a few milliseconds and try again... and again... and again until the motor eventually twitches enough to allow the controller to figure out the correct polarity sequence and provide more current to properly spin the motor.
If there is enough free play ('slop') in the gearbox and/or belt drive between the motor and the hammer so that the motor can spin freely for about one full revolution before it takes up the load of lifting the hammer the whole start-up process takes only an instant. However, if the drive train beween the motor and load is very 'tight' and there is not enough free play to allow the motor to start its spin-up without load you're going to get the type of delay you describe.
You've given me a complete description of your hammer electronics but you have provided no details about the mechanics of the hammer drive train. Twisted Sick Robotics' beetleweight 'Dain' has the weapon motor mated to a 14:1 gearbox that then goes to a 3:1 belt drive for a combined 42:1 reduction. It's a good bet that the combined play in the two-stage gearbox along with a fairly loose timing belt allow the weapon motor enough no-load slop to get thru the startup sequence muy pronto. Given your description of the problem it's also a good bet that your unspecified weapon drive train does not have enough slop. Add some.
Reply: Hi Mark! I appreciate the in-depth tips. I run the exact same gearbox and additional reduction as 'Dain' and - credit where credit is due - WOW it works well! The only difference is that I run the additional reduction using specially cut gears, not belts. I'll be looking into a sensored weapon motor setup or swapping to a belt drive system now. Thanks for all your help! Cheers!
A: Very pretty, Ontario! It would be a pity to remove those bespoke gears; a sensored motor and controller will cure your response delay. May your opponents all have fragile top armor.
Q: I've got a big wheelchair motor with a magnet that popped loose from the motor case. What is the recommended adhesive to stick it back into place? [Pittsburgh, Pennsylvania]
A: [Mark J.] There are specialty adhesives made for this purpose, but they are expensive and difficult to find. I've been very satisfied with 'Original J-B Weld' epoxy which is widely available at hardware and auto parts stores. I prefer it to other 'off the rack' epoxies for its ability to survive high temperatures; not all epoxies are stable when hot, and motor magnets do get hot.
Chip away any residue of the previous adhesive and give the surfaces a quick solvent wipe before applying an even coat of new epoxy. Tip the magnet carefully back into position and let the epoxy cure overnight before completing motor re-assembly.
For tips on 'battle hardening' magnets in smaller motors see this post in the Ask Aaron archives.
Q: I'd like to see if a pair of 36 Volt 1000 Watt 3000 RPM Electric Scooter Motors will work for my 'bot, but the motors have no specs. Is there any way I can use the Tentacle Torque Calculator to see what the performance of my 'bot would be with these motors?
A: You have more motor information than you think. Take a look at Calculating Motor Specifications. 
Q: Hey it's me form the Oak Harbor/Anacortes area again. My robot design has been going through a CAD redesign, but now I'm ready to get back to actually getting the thing to move. A long while back I asked about brushless drive for beetleweights, and you reccomended the Afro 20A ESC. But, they are out of stock at Hobbyking again. Would you recommend any other similar ESC's? [Oak Harbor, Washington]
A: [Mark J.] The Afro ESCs were a stable 'known quantity' produced over a fairly long period of time. Unfortunately, the Atmel chips used in the Afro ESCs have started to dry up and the model aircraft industry that drives brushless ESCs has largely moved away from the robot-friendly SimonK firmware and toward BLHeli firmware. There aren't likely to be any more Afros made.
It's difficult to recommend a replacement ESC for entry-level brushless drive. The main problem is that most small brushless ESCs are produced in relatively low numbers and disappear from the market as fast as robot builders can figure out their quirks and decide if they're useful. Once you've picked out a brushless ESC you'll still need to go in and modify the firmware settings to allow for forward/reverse operation and make a few tweaks to motor brake settings to make them suitable for a robot drivetrain. See if this page makes any sense to you; if it reads like Chinese algebra you're in for long and painful learning experience.
I'm not a good source for answers to questions on brushless ESC firmware settings or for the current 'hot flavor' of ESC. A reasonable place to start would be the Facebook Combat Robotics group. My suggestion is to avoid going 'brushless hipster' and stick to brushed drive motors until you've run out of other aspects of robot design to master.
Q: OK, I've gathered up the hardware needed to set the Simonk firmware on my Afro ESC, identified the leads on the circuit board I need to connect, figured out my release version, and identified the correct 'target' for the SimonK firmware settings tool. Now I have no idea what firmware settings to modify! What settings do I use for robot drive? [Late night phone call from a Midwest area code. Who gave you my number?]
A: [Mark J.] Congratulations on making it that far! As a starting point, use the default values - with these exceptions:
Q: I have a problem my drive Esc are gettin hot. It is a brushed esc witg brake. And I used a 6v motor 280 rpm and 65 mm diameter of the wheels. The chasis are on the attached file which is made from aluminum with a weight of 1kg. How to lessen the temp of esc [Quezon, Philippines]
A: [Mark J.] I'll assume your ESCs are the "RC ESC 20A Brush Motor Speed Controller w/ Brake" widely available on eBay, and from the photo of your chassis you attached it appears that the motors are the "6V DC Small Speed Reduction Gear Motor Metal Gearbox with 65mm Wheel Kit" also found on eBay. These are both fairly standard low-budget components for insect-class robots and should work for you.
It's not unusual for small ESCs to run hot -- hot enough to burn your skin if you touch them. The solution is to not touch them.
A few suggestions:
Q: I was researching brushless outrunners and I understand that the four digits that they are described with refer to the diameter and length [in millimeters], but do these values mean anything in terms of rpm or torque? For example, would a motor with a larger diameter always have more torque than a similar motor with a smaller diameter? [Anacortes, Washington]
Q: What are your thoughts on cheapo Chinese ESC's like this one? [eBay link redacted] It's been reported to work up to 4s by others, and I plan on using a 3s battery with a 1000rpm kitbots 25mm gearmotor.
A: [Mark J.] So, you want to trust your success at a tournament on a pair of $4.65 ESCs (shipped free from Hong Kong) pushed more than 30% beyond their rated voltage 'cause somebody says that they've gotten away with even greater abuse? Allow me to reprint an excerpt from Frequently Asked Questions #16:
Q: Where can I get really cheap combat robot parts?
A: You're going to spend a lot of time and effort building your 'bot and going to a tournament. When some component fails in combat and puts you out of the competition, you're gonna wish you hadn't gone cheap. In particular, don't scrimp on electronics! With experience you'll learn where you can save money, but it's not gonna be on key components like speed controllers and wheel hubs.
'Cheap' and 'Combat Robot' don't go together!
That said, some teams are willing to use the really cheap off-brand electronics, motors, and batteries available from China. Documentation is awful, customer service non-existent, quality control unheard of, and delivery uncertain -- use them at your own risk.
When 'bots with these ESCs start winning tournaments feel free to use them. Paying more for a component doesn't guarantee success in combat, but it does buy some peace of mind. If you pay $30 for an ESC and it fails you can blame the manufacturer, but if you pay $5 and it fails you can only blame yourself.
Comment: I believe the questioner was referring to a Reddit thread from a while back, the suggestion of those ESCs was made by Sean McCartin because he's used something like them successfully on 'Melanistic Leopard'.
I'm still shelling out for the more expensive ones, because it's only $30-40 difference in the end on a $300-500 bot, but you can't deny he's had tournament success with them. [Vancouver, British Columbia]
Reply: I'm not sure how long Sean used JMT controllers in 'Melanistic Leopard' or if he still uses them. He had been having drive controller problems a few months earlier and may have gone to them in desperation. Here's the unenthusiastic recommendation he gave in a Reddit post from March, 2018:
Note: The following post is from the Indian subcontinent where robot combat is MUCH different than you see elsewhere in the world. Arenas are small, availability of parts is limited, speeds are much slower, and battles are very cautiously contended. The featherweight drivetrain I recommend here is both too fragile and too slow to be suitable elsewhere.
Q: first,as you can see a wedge. & i plan for a motor powered lifter in future. i've hit my first event & had issues with keeping all wheels attached to the ground. Its a six wheel small compact robot which i think i need to improve a lot,specially keeping all in a same level. & I'm struggling with it so far.
now, here i want my suggestion. cases are Motors..
now,re caps.. I'm interested using 775's but would using 555's will do for me.?? & brushed ESC for a good price & i don't mind them used.. [Bangladesh]
A: [Mark J.] From what I've seen of robot combat in your region, six RS-555 motors geared down to 300 RPM in a featherweight would be entirely competitive with other robots. The small arenas combined with a builder preference for slow speeds don't require much power.
The
Tentacle Drivetrain Calculator
is handy for these design calculations. It says that a 30 pound robot with six 300 RPM 555 series gearmotors @ 12 volts with 6" diameter wheels will have a calculated top speed a little better than 4 MPH, reaching that speed in about 3 feet with a maximum current draw of just 3.5 amps per motor when pushing hard.
Not sure what ESCs might be available to you in Bangladesh. Given your budget you might be interested in the inexpensive Australian BotBitz controllers.
A: Here's a table showing performance with different wheel sizes. The 3.5" wheels are slow -- not even walking speed. If you were to drop down to four motors and wheels, you'd have room for larger wheels for greater speed. Four of your 555 gearmotors with 5" wheels gives a 3.6 MPH calculated top speed reached in about 3 feet with only 4.3 amps of current per motor at max push. That sounds like a good solution.
I'm not certain what your battery options are, but these motors will handle a bit of overvolting. Running 14.8 volts (4 LiPo cells) with four motors would boost your speed to 4.4 MPH and keep your acceleration brisk. Max current remains at 4.3 amps per motor because your max torque is traction limited.
Q: I'm looking to experiment with gearmotor gearboxes that would attach to the brushless motors for my beetleweight wedge. How do the gearboxes attach to the motors? will they attach to most motors of that relative size? Is the Hamburger bad? [Anacortes, Washington]
A: [Mark J.] The hamburger is fine. I've been avoiding questions about brushless drive, but I can only dodge the topic for so long and now feels like a good time to jump in. The basic brushless beetle drive conversion is mechanically pretty simple:
 + =
I'd suggest starting with 'Afro 20A Race Spec Mini' ESCs; they're popular for beetle drives, come with appropriate control firmware installed, and are widely available. If you're VERY LUCKY the default firmware settings are going to provide smooth and quick response for your beetleweight drive. If not, you'll want to join one of the on-line robot forums to find other builders with your particular setup to assist you in modifying those settings. It's WAY too much to cover here at 'Ask Aaron'. Best luck. Q: I'm down to the last stages on my beetleweight, but I've run into a snag. My receiver is working, the signal goes just fine to my brushless weapon motor when tested on all three channels I'm using. However, my brushed drive motors won't respond on any of the channels. The motors run fine when voltage is applied directly, so I've narrowed the problem down to the ESC. I'm using a Dual TB6612FNG from RobotShop, and I've triple checked all my wiring. Is there something I'm missing? Is the ESC compatible with an RC signal? It says it's PWM. Any help would be greatly appreciated. [Mark from Vancouver]
The reason it's so cheap is that is isn't a brushed motor controller -- it's a brushed motor driver, and there's a big difference.
Reply: Thanks once again! There's more of a learning curve on this stuff than I thought when I started out. I think that's the last hurdle for me to get over before Seattle. I decided to go with a pair of Wasps instead of the Scorpion because I simply don't have the space in my chassis for the Scorpion, but as I understand they're battle tested as well. Q: Is there a reason those who use brushless [MOTORS] only have used the ones made for airplanes? Why not cars? Something like this: Turnigy 1/8th Scale 4-pole 2100KV.[Hagerstown, Maryland]
A: [Mark J.] There are two broad types of brushless motor design:
As robots get larger you see more inrunner designs in use. Larger inrunner motors spin more slowly than the tiny ones which makes them a workable option for both drive and weapon power. The Turnigy motor you referenced has about the right amount of power for a hobbyweight weapon but spins at better than 38K RPM, which could be difficult for a belt drive to handle -- simpler to use an outrunner. For drive use, a pair of the Turnigy inrunners would have enough power for a featherweight 'bot, and this is probably the lightest class where I would consider using inrunner drive motors.
By the time you get up to heavyweight 'bots, the outrunner motor styles have disappeared and everyone uses inrunners if they're using brushless.
Q: How much can the performance of a brushless ESC be improved by swapping capacitors? I saw on Charles Guan's website when he was learning brushless drive setups that he swaps the capacitors out on all his ESCs for high-performance capacitors. I know this is supposed to help shield the ESCs and prevent burnouts from current surges when the motors are pushing at low RPMs, but how much does it help, and how much should I upgrade the capacitance of my ESC? To be more specific, I have a 30 amp brushless ESC with a pair of 25v 550 uf capacitors. Would it make sense to swap those for a pair of 35v 2200 uf capacitors? They only weigh 6 grams and the ones I'm losing are probably 3, so if I can get any increased survivability for my ESC that kind of weight would be worth it. The ESC will be running a .102 Ohm 4S 580kv brushless motor with a 1.8:1 pulley reduction to power a large horizontal spinner (about .0025 Kg M2). Also, I read on some hobby forums that ESC capacitors should generally be rated at twice the voltage of the system, is that true? My system is 15v, and the current caps on the ESC are 25v so is it worth it upgrading for that alone? Thanks again for all your help! The bot is coming together nicely and I've registered for the Robothon in Seattle in October. [Mark from Vancouver]
Charles Guan is an MIT engineer who designs and manufactures ESCs in his spare time 'cause he thinks it's fun. He absolutely knows what he's doing. If you're not a electrical engineering graduate of a respected school you're likely gonna to do as much harm as good by tinkering with the components of a complex digital electronic device.
That said, the little brushless ESCs robot builders 'borrow' from the hobby R/C aircraft industry are designed for very different performance targets than we expect from them in our applications. Bumping up the voltage rating of the filter capacitors will do no harm and may provide some level of extra reliability.
Blindly increasing the capacitance is another matter. I'm pretty sure that Charles Guan had his drive system fully instrumented and was watching voltage spikes on an oscilloscope as he changed the capacitor values. Simply pouring in more capacitance is shooting in the dark.
A drive motor ESC sees a lot of high-load/low-speed action that can benefit from added capacitance, but a spinner weapon ESC operates much closer to the propeller-spinning job for which the unit was designed. I'd suggest keeping the factory capacitance value as-is unless you discover a specific reliability problem as too much capacitance can adversely affect high-speed operation. If you really want to tinker, keep the increase to no more than double the original value. I suspect that most spinner ESCs fail for reasons that more capacitance isn't going to overcome.
P.S. - I'm happy to hear that your build is going well and that you're registered for Robothon - that's a great event. Say 'hi' to the Seattle crowd for me.
Q: I'd like to try brushless drive motors for a new robot, but I can't figure out how to input brushless motor data into the Tentacle drive train calculator. How do I calculate the stall torque and torque constant for a brushless motor? [West of the Pecos]
A: [Mark J.] The
Team Tentacle Torque/Amp-Hour calculator
was designed for brushed motors and assumes an inverse linear relationship between motor speed and torque. Common hobby brushless motors do not have that speed/torque relationship, and the entire concept of brushless 'stall torque' doesn't really make sense. To further complicate the issue, the torque curve of a brushless motor is highly dependent on motor controller firmware and on the specific user-adjustable firmware settings. The Tentacle calculator is nearly useless for designing a brushless drive train.
See what successful bots in your weight class are using for brushless drive motors, gearboxes, ESCs, and (importantly) controller firmware/settings. In general you'll find motors with high Kv constants, high reduction ratio gearboxes, and controllers flashed with SimonK or maybe BLHeli firmware. Brushless is a whole lot more complex than brushed drive, and I'd strongly recommend that you start with a proven combination. Experiment after you have a working drive train.
Q: My antweight horizontal bar spinner bot currently uses:
A: [Mark J.] There are some fairly common misconceptions about hobby-grade brushless ESC amperage ratings that I need to clear up:
Q: I see several different brands of brushless ESCs labeled 'OPTO'. What does that mean? When would I want 'OPTO'? Would it ever be a bad thing? [Pomona, California]
A: [Mark J.] A complete answer requires a bit of history.
Back in the day...
The small, high-power brushed motors powering early electric model aircraft created a huge amount of high-frequency electrical 'noise' and voltage spikes from the arcing of the motor brushes at high speed. This 'noise' traveled along the power wires and found its way to the sensitive and vulnerable electronics in the R/C receiver, wreaking havoc with reception.
To protect the receiver from this electrical noise, 'OPTO' speed controllers included a device called an optical isolator in the line between the receiver and the power circuits. This 'opto-isolator' converted the receiver output signal to a modulated light beam, then back to an electrical signal for the ESC. Electrical noise couldn't jump back across the optical gap to get to the receiver. Genius!
But there's a catch...
If you use a battery eliminator circuit (BEC) to power the receiver from the main battery pack the electrical noise can use it as a path to the receiver, nullifying the benefit of the opto-isolator. Fully isolating the receiver from motor 'noise' requires that the receiver have its own battery, so OPTO ESCs had no BEC. Inconvenient, but it worked.
Jump to present day
Brushless motors have replaced the 'noisy' brushed motors in model aircraft and the need for opto-isolation no longer exists in most applications. Few brushless 'hobby' ESCs have an optical barrier to electrical noise, yet many still carry the OPTO label. In current usage it's simply an indication that the ESC has no battery eliminator circuit. Eliminating the BEC shaves off a couple grams of weight and eliminates the need to 'clip' the red power line in the receiver cable when multiple BEC equipped ESCs are on the 'bot.
You need something in your power system to provide power to your receiver, and a convenient way to do that is to make sure that one of the ESCS for the weapon or drive motors has an BEC -- but you only need one.
Q: I'm new to brushless motors and having trouble setting up the Plush 30A ESC for my beetleweight weapon. When I switch on the robot power the Plush keeps beeping and I get no response from the motor. What's the problem?
A: [Mark J.] All the Plush ESCs use beep tones to communicate their status and alert you to problems. Since you say it 'keeps beeping' we can rule out the normal startup tones that tell you everything is OK. The 'keeps beeping' codes are:
Q: Got it! It was a low voltage issue. Now I have a new problem. My outrunner weapon motor spins up fine without the weapon, but with the weapon attached it spins very slowly and at an uneven speed. The battery is fully charged, I've checked all the connections, and the weapon spins freely by hand. Any ideas on what might be wrong?
A: Robot builders 'borrow' brushless motors and ESCs from the model aircraft hobby market. These products are designed to spin low-mass propellers, and the firmware in the ESCs are designed for that purpose. When you hang a big, high 'Moment of Inertia' bar/disk/drum off a non-sensored hobby brushless motor the ESC 'firmware' may not know how to handle that type of load and gets stuck in its 'startup' sequence. This is usually more of a problem with robot drive than with spinner weapons, but it does crop up and can be very frustrating. It's called 'cogging'.
There are adjustments to the Plush firmware settings that may help. Instructions for entering the ESC programming mode are given in the Turnigy Brushless ESC manual linked above, but I think you'll be better off watching a video of the programming process. It's more than a little confusing, and you might be interested in purchasing an inexpensive Plush Programming Card to avoid all the beep-counting and stick-throwing.
There are six settings that are user-adjustable, but for this problem you'll be interested in 'Timing' and 'Startup Mode':
Q: Hi Mark,
I have looked high and low and cannot seem to find the motor I would like. I would like a flat motor (no taller than 2") that runs on 24 volts. It will be used as a weapon motor for an overhead spinner with a 20" or so bar weighing between 4.5 and 6.5 pounds. By my rough calculations, 250 kv and 4 Nm of torque would be ideal, but 150 kv with 8 Nm of torque would also work. Any websites or brand suggestions would be much appreciated. Many Thanks! [Pittsburgh, PA]
A: [Mark J.] There's a reason you can't find that motor. Robot builders 'borrow' weapon motors from the model aircraft hobby industry, and the motor you're looking for is...
Q: So I have a question regarding battery eliminator circuits.
Also out of curiosity, I had no problems last year with both controllers providing power to the receiver, any idea why that might be? [Ballwin, Missouri]
A: [Mark J.] Versions 2.4 and later of the tinyESC have components to prevent multiple BECs from affecting the tinys. As noted on the FingerTech tinyESC page:
Note BECs don't 'stack' to overpower the receiver or other electronics -- they're all wired in parallel. The problem is that multiple unprotected BECs may confuse each other's voltage regulators and cause an unstable fluctuation that effects receiver performance. When in doubt, use only one.
Q: I'd like to try brushless drive motors in my next 'bot, but I'm really confused about the motor controllers. How do I tell if an ESC can be flashed with simonk firmware, and how do I actually flash an ESC? [Central Florida] A: [Mark J.] Builder Robert Cowan has a very nice video on his YouTube channel - SimonK Firmware Flashing Tutorial (Brushless Drive):
This video covers how to flash reversible SimonK firmware on to any compatible ESC. In the video, I'm using a HobbyKing F-60A, but any ESC with an ATmega-8 processor should work. This should allow you to use a brushless motor for a forward/reverse drive system for a robot.
If the process is a bit intimidating to you, there are small brushless ESCs that come pre-flashed with SimonK. Search your favorite hobby source for 'SimonK'.
Comment: Regarding your last answer. There's a table of ESCs and their relative SimonK & BLHeli compatibility in this Google docs spreadsheet.
It may not be 100% perfect, but it's the best I know of. [Mark from Vancouver]
Reply: That resource and several others appear in Robert Cowan's comments directly under the YouTube video linked above (click on 'show more'). I'll reprint those links here for those who missed them: Q: I have live steam train with two ampflow motors. I burned a motor up completely. Is ampflow still have issues? [Salem, Oregon]
A: [Mark J.] AmpFlow markets high performance motors designed for applications where weight and space are at a premium. They will not take the same level of abuse that larger, heavier industrial motors of similar output will absorb. When used within the performance envelope for which they are designed they are reliable and provide excellent performance.
You shared no details of the load or service level placed on the motors in your application, and neither did you specify which AmpFlow motors you are using. I can only comment that the AmpFlow motor failures I have seen have been the result of their being placed in applications poorly matched to their capability.
Q: Hi, again. Last time I asked about the rpm to run vertical disk spinners at, and now after spending considerable time optimizing my disk with FEA I am wondering about drive motors for mantis weight robots. I know of the bane bots gearboxes but those seem a bit heavy for mantis weight while at the same time other gearboxes seem too light duty, do you have any recommendations on where to look? I was almost considering just swapping to a belt drive system for gearing but that seemed to not be the smartest idea to me.
This thread on the SPARC forum has some discussion on eBay gearmotors for Mantis bots.
Q: Is 2x Castle 1406 1900kv @ 20:1 3in wheels drivetrain underpowered for 30-pounder? [Ann Arbor, Michigan]
A: [Mark J.] There's a "rule of thumb" for brushless motors circulating out on the builder forums that appears to work well for drive motor selection. The rule states that a set of brushless drive motors that together weigh between 2% and 3% of the total weight of the robot will have more than enough power and adequate 'thermal mass' to survive. Exactly why this works is a topic for a future discussion.
A pair of Castle 1406 1900kv motors weigh in at 2.9% of a 30 pound robot. That puts you on the high end of the golden range. On 4 cells with 3" wheels and 20:1 reduction you'll have about 12 MPH speed. Sounds healthy to me. Q: Can epoxy putty or “mighty putty” be used to “battle-harden” brushless motors? [Columbia, Missouri]
A: [Mark J.] I wouldn't. You're looking for an epoxy that is both light in weight and 'runny' enough to wick into cracks and small voids before it sets. Epoxy putty fails in both of those requirements.
Builder Robert Cowan has a nice video demonstrating how to harden a brushless outrunner motor for combat. Robert uses standard slow-cure two-part epoxy mixed with glass microspheres for both the rotor magnets and the stator windings.
My preference is to use the microsphere mixture to fill in the rotor gaps and strengthen the magnet array without sagging during the cure process, but brush pure epoxy sparingly onto pre-warmed stator windings. I warm the stator to around 100-120 degrees Fahrenheit (warm to the touch) with the gooseneck flood lamp on my workbench; the epoxy thins when heated and soaks into the windings where it can better hold the wire in place without interfering with stator cooling.
Q: Hi Mark. I am just wondering, are there some major differences between MagMotors and AmpFlow motors? I do believe that these two are more or less the same brand but...are there any build quality differences between the two companies? and Are AmpFlow motors more likely to burn out than its MagMotor counterpart models? [Secaucus, New Jersey]
A: [Mark J.] Back in 2010 the distributor of high performance Magmotor models to combat robot builders changed the name of the motors he was selling to AmpFlow. He told me at the time (I still have the email) that this was simply a marketing choice, but apparently the AmpFlow motors also switched to a new manufacturing source and no longer had ties to Magmotor. On casual examination the design remained the same, but there are differences in materials and processes.
The motor problems reported by some builders stem from the design of the motor brushes -- a design shared by both AmpFlow and Magmotor. Overvolting and/or overloading these motors causes brush failure that leads to rapid heat build-up and structural failure of the commutator. Any motor can be pushed to failure if stressed beyond its limits, and many builders have no problems at all with these motors.
Which motor line will take more abuse? I have no personal data on that, but if you blow up either an AmpFlow or a Magmotor it's likely because your design is at fault.
Q: Hello Mark. Our team is currently designing a heavyweight bot. We have heard from builders that describe Ampflow motors as inconsistent and unreliable. Are there any better alternatives? A: [Mark J.] AmpFlows/Magmotors become unreliable when loaded heavily for prolonged periods or overvolted on robot drive trains or lifter weapons. Their brushes are not up to sustained operation at very high amperage. They are quite reliable when running at nominal voltage with reasonable gear reduction. Q: Do amp flow motors only work with amp flow escs? [Arlington, Virginia] A: [Mark J.] AmpFlow motors may be controlled with any brushed motor ESC of suitable voltage rating and current capacity, and AmpFlow ESCs will operate any standard brushed motor within their current and voltage rating. Q: Can you make a calculator for how much current a given motor will draw at a given load?
A: Permanent magnet direct current brushed motors have a linear relation between torque output and current. If you know the torque output required, the formula is very simple:
Q: What makes the RageBridge 2 so good for lightweights? It’s amperage rating is so tiny. A: Previously answered. See this post farther down in this archive. Q: Would a sabertooth 2 x 60 driving 2 amp flow e30 150 motors be adequate for a lightweight? [Dallas, Texas]
A: [Mark J.] Motor current draw depends on voltage, wheel diameter and gear reduction ratio as well as motor type and robot weight. The
Tentacle Torque Calculator
can evaluate these factors and provide an estimate of maximum current draw as an aid in motor controller selection. It will also model the speed, acceleration, and battery requirements for a robot of a specific drive train design. Learn to use this valuable tool.
The Sabertooth ESCs have a poor reputation in combat applications. Builders using them have reported sudden and unexplained complete failure of the controllers following a high-energy impact. Although they work well in general robotics I no longer recommend their use in combat robots. In this power range I recommend the RageBridge 2.
Q: Hello Aaron. It’s my first time in your wonderful world of robots, and I hope you can help me with a problem. I have a Sabertooth 2x32 controller attached to two 200 watt motors. The system runs on 24 volts and is controlled by a wired analog joystick. The thing is that the motors are running weaker and slower than I expected. Checking the output from the controller to the motors with a DVM shows 12 volts instead of 24 volts. I checked the parameters with the software, and everything looks OK, but maybe I overlooked something? The joystick signal is 2.5 volts in the center, 5 volts at full forward, and 0 volts at full reverse. I don't know if that's OK or not - I thought it should give 5 volts in every direction. I appreciate your help in figuring this out. Arik [Tel Aviv, Israel]
A: [Mark J.] Hi Arik. The Sabertooth 2x35 controller is made for general robotics applications; its functions are much different than the dedicated R/C Sabertooth controllers used in combat robots. Still, I may be able to help find a solution to your problem.
Reply: Hi Mark, thanks a lot for your last answer! I sent the question to the controller company support, hope to find an answer shortly. Q: I'm having trouble with a pair of 550 series motors I purchased recently. At full throttle the motors spark and sputter with the power dropping off and surging. They do this even when powered directly from the battery (3-cell) so it isn't the electronics. I've never had this trouble before. What's going on here? [Northern Ohio]
A: [Mark J.] This sounds like motor brushes that aren't seated properly. Inexpensive brushed motors like your 550s are manufactured by several different sources, so even if you're buying from only one distributor you may be getting variation between one batch and the next. The care taken by the manufacturer to assure that the brushes are correctly contoured to match the commutator is one of the things that can vary.
Prior to subjecting to full power, brushed motors (particularly small, inexpensive ones) should be 'run in' at low-speed/no-load to allow the brushes to wear in and match the curve of the commutator. Failure to do this may result in electrical arcing that can damage the brushes and commutator.
Q: I'm currently doing the electronics for my 3 pound bot but I'm running into some issues:
A: [Mark J.] Let's look at the separate responses of the left and right motors:
Response: That makes 100% sense. Thanks so much for catching that. I tried swapping the plugs in the receiver and it did what you said it would do. I will contact botkits for a replacement immediately. Q: Why are bigger motors (example: Perm 132) slower in rpm than smaller motors? (example: Park 150) [Arlington, Virginia]
A: [Mark J.] Think about it this way:
General rule - everything else being equal, smaller diameter armatures/rotors spin faster, but provide less torque. There are other factors involved, but I'm not going to write a treatise on motor design in this short-answer format: more info.
Q: Other than shaft threading, are there any other factors pertaining to brushless motors that determine its intended spin direction? As always, your advice is greatly appreciated. [Albany, Oregon] A: [Mark J.] There is no preferred spin direction for hobby BLDC motors. A brushed motor commonly has the commutation 'advanced' a bit in one direction via the mechanical alignment of the commutator relative to the motor brushes to provide greater power and speed in one direction. Brushless motor controllers will adjust the motor timing automatically and are not affected by the direction of rotation. Spin 'em either way -- no difference. Q: Just out of curiosity, is there any combat robotics related application for the NPC-A0045512 Gearmotor? If not, what is this thing even used for? [Charlotte, North Carolina] A: [Mark J.] Nope, it's a military item (the color gives that away). This $9600 gearmotor is used to rotate gun turrets on armored vehicles. Q: I want to use a brushless motor for my spinner weapon. How do I calculate the 'Stall Torque' of a brushless motor to enter into the on-line Spinner Weapon Kinetic Energy Calculator? [The Aether]
A: [Mark J.] Motor stall torque is not listed in the standard spec sheet for brushless motors because hobby brushless motor controllers have a 'soft start' function limiting low speed torque. I've put together a simple JavaScript calculator that will give you an estimate suitable for the kinetic energy calculator. The Kv (RPM per volt) and Ri (internal resistance) values are commonly given in brushless motor specs.
A: [Mark J.] The aether was revived when physicists needed someplace to put dark matter and Facebook. Put a message in a bottle and toss it out onto the vast sea of the internet: it may drift close enough for me to snag.
A: [Mark J.] The Talon SR is a very simple brushed motor ESC that is entirely combat useable. Ratings are: 6 to 28 volts, 60 amp continuous, 100 amp peak. It lacks the fancy input protocols and external sensor capability of the SRX, but combat 'bots generally don't use those. A few things to watch out for:
If you need two brushed DC motors to power wheels on the same side of a tank-steer robot or gear together to a common output shaft, the motors can be wired in parallel to the output of a single-channel electronic speed controller; see the diagram.
A: [Mark J.] A threaded shaft is common on hobby brushless motors, but it isn't a problem:
Q: Hey Mark, just wanted to give an update in case some future bot builder reads my question and is confused by what diameter the shaft is, I think the actual internal motor "shaft" (the on that can be seen on the bottom of the motor) is 3mm and the threaded piece that sticks off the top of the can is 5mm. Maybe that's common knowledge that I'm just now learning, still, feels like there should be better terminology / better clarification of this.
A: That helps -- thank you! That mismatch of shaft and output sizes is not a standard configuration and the descriptions on the web are confusing. It's normal for a single shaft of uniform size to run thru the motor from end-to-end. I'm a bit worried about how the transition from 3mm to 5mm is accomplished. I appreciate your update.
A: [Mark J.] 'Ask Aaron' provides the tools needed to calculate battery capacity requirements given the details of the drivetrain and weapon system. There is no 'general' number -- the requirement varies with gearing, moment of inertia, wheel diameter, and expected loading. See this previous post in the Ask Aaron Radio and Electrical archive that links to the tools and gives example calculations.
Note I wouldn't run the E30 series AmpFlow motors at much more than 30 volts. The brushes overheat and 'burn' at higher voltages. If you need to run 48 volts I suggest switching to the fan cooled A28 series.
A: [Mark J.] One of the first things you learn in combat robotics is that the advertised specifications for electronic speed controllers are generally meaningless. There is no standard for determining advertised current ratings. Hobby level ESCs typically take their rating directly from the power control chip spec sheet, which is massively misleading. The power chip rating is VERY sensitive to temperature and as soon as the power starts to flow the temperature starts to rise, which quickly drops the power rating -- but they don't tell you that. RageBridge current ratings are based on actual testing under the worst conditions an ESC can face.
Then there's what happens when you exceed the current capacity of the ESC.
A: [Mark J.] The more info you can provide about your 'bot, the better chance I have to give a direct and useful answer. Given the symptoms, I'm gonna guess that the motor is a brushless outrunner, since brushless is 'in' and an 'inrunner' style motor would be a horrible mistake for this purpose.
Possible cause #1
Brushless motors rely on the brushless ESC to provide power to the correct motor windings at precisely the correct time. With the non-sensored brushless motors commonly used in insect-class combat robots, the ESC is not given data on the position of the rotor relative to the motor coils -- it has to take a guess based on small changes to the electrical properties of the motor as it rotates. At slow speeds and heavy loading, the ESC may guess incorrectly and send power to the motor windings at the wrong times. What you call 'sputtering' is in this case more correctly called 'cogging'.
I suspect that the ESC you have chosen for your weapon is a poor match to the motor. There are so many different motors and ESCs that it is very difficult to track their compatibility. I suggest that you join the Combat Robotics group on Facebook and tell them:
Possible cause #2
It's also entirely possible that your battery is not fully charged and/or does not have enough capacity to deliver full voltage under the heavy load of your unspecified weapon motor starting up from a full stop. The drop in voltage can 'brown out' your electronics and cause a sputtering of the type you describe.
LiPoly batteries are shipped with only a partial charge, so before you go off and ask the Facebook group for help give your battery a FULL CHARGE and try another test.
Five Days Later...
I haven't heard back from Arlington, nor have I seen any posts on this topic in the forums. I'm guessing that a battery charge did the trick.
A: [Mark J.] There's nothing special about an 'EDF' motor; they're just brushless inrunners suitably sized to power an 'Electric Ducted Fan' in R/C 'jet' aircraft. In general an 'outrunner' style motor of similar power will spin at lower RPM and will be easier to adapt to a combat robot weapon in a light combat robot.
A: [Mark J.] Ask Aaron no longer answers questions from builders that compete in India (why not?). You are free to search our site for answers; your questions have been answered here many times -- most recently just a few posts down this page. You may also find some info in our Motor Control by Solenoid guide.
A: [Mark J.] The Tentacle Drivetrain Calculator
shows that two of those motors in a 20 kg robot would exceed both their torque and amperage ratings by a factor of three under combat conditions. In addition the robot would be very slow, crawling along at less than walking speed. The motors are not suitable for your purpose.
I've been pondering a concept for a while and am wondering if it would even do anything different, or in fact, would create a new world of problems.
Many bots armed with spinning weapons use some sort of gear reduction in order to take a high RPM motor and bring it down a bit in order to increase torque and spin up time. But what if it were reversed? Instead of using a 1000kv motor with a 2:1 reduction, would anything change if we used a higher torque 500kv motor with 1:2 "reduction"?
David R. [Livermore, CA]
A: [Mark J.] Builders generally choose a weapon motor with the lowest RPM for a given motor weight and power so that the smallest gear reduction ratio possible is used. This assures that the reduction is efficient, using light and compact components. Consider:
A: [Mark J.]
Given your description I can't figure out what your robot is doing, let alone propose a solution.
Try asking at the Facebook page for 'Combat Robotics India' -- perhaps they know what you're talking about.
A: [Mark J.] Speed controller requirements depend on the specifics of the drive layout as well as the robot weight. Robots with large weapons may have drive trains with barely enough power to move the weapon into position, while robots that rely on speed and acceleration may have drive motors many times more powerful that require speed controllers capable of delivering huge current at higher voltages. No single speed controller is 'right' for all heavyweights.
Assuming use of brushed drive motors, here are some popular speed controllers used in heavyweight robots:
I have a question regarding different configurations of outrunner brushless motors. Specifically, I seem to see some that are designed to attach the load (wheel, prop, whatever) directly to the rotating "can", and some that appear to have an output shaft exiting the motor from the opposite end (the stationary part where the wires attach). Finally, I've seen some that appear to do both.
Am I correctly interpreting how these different designs work? And on the dual-output designs, is it possible to attach things to both ends at the same time?
Thanks for your help -- Ian M [San Jose, California]
Take a look at the photo at right. Outrunner brushless motors have two 'chunks':
It's good engineering practice to take power from a shaft as close to the motor mounting plate as possible. Forces applied at the far end of the shaft place additional strain on the bearing tube and can cause the mounting structure to flex. You might consider a support bearing block out at the end of the shaft if your design calls for a pulley or sprocket that far from the mounting base -- but you can take power from either or both ends of the motor.
Note: You mentioned a wheel in your question. You may attach a prop, sprocket, pulley, gearbox, or small spinner weapon directly to the motor -- but small brushless motors do not have enough torque to allow you to directly attach a robot drive wheel.
Comment: Hello Mark, Ian M again.
Thanks for your answer -- that helps a lot. For context, this is just for a design exercise I've been playing with that's unlikely to actually be built: an invertible HOG drive. In this case, a variable speed reduction is inherent to the design and the shafts would have extra support bearings, so that's not an issue, but having a dual-ended motor would simplify things a lot so it's good to know that's a real thing. In any case, thanks for the warnings about supporting the shaft and the speed reduction, even if they weren't needed.
A: [Mark J.] I recommend pre-running ALL brushed motors -- particularly small, inexpensive Chinese motors like those from robot component suppliers like Kitbots. See the next post down. Suppliers do not do this for you.
For more info on brushed motor break-in see this article at the RC Universe Forum. Note: I don't recommend the 'wet' break in process, and a few minutes of running-in is adequate.
A: [Mark J.] Insect class builders continue to report problems like these when running inexpensive brushed 'eBay Special' motors with certain speed controllers. However, the problem doesn't originate with the speed controller!
Where is all that 'noise' coming from? Power to the armature is transferred across the sliding connection between the brushes and the rotating motor armature. Efficient power transfer and correct timing of power flow to the armature coils depends on the motor brushes being correctly contoured and in full contact with the commutator. If the brushes are not in contact with the commutator over the intended full surface area there will be excessive heating at the commutator plus electrical arcing that generates the damaging voltage spikes.
Brushed motors (particularly small, cheap brushed motors) must be 'run in' for a few minutes at reduced voltage and loading to allow the brushes to wear away surface irregularities and contour themselves to the commutator radius. Exposing the motor to high voltage and loading before running the motor in may damage the commutator and brushes and cause permanent 'noise' problems. Here's how you do it:
A: [Mark J.] DO NOT OVERVOLT MOTOR CONTROLLERS!
A: [Mark J.] Build a lot of BattleBots in Nepal, do they? The power of the motor matters, and power is the product of torque and RPM:
American builders aim for about 100 watts of power per pound of robot weight, although successful robots can be built with as little as 10 watts per pound. When you have enough power you can adjust gear reduction between the motor and the wheels to get the required speed and acceleration for your design.
Team Run Amok offers tools to help with robot design and motor selection: here is an example drive motor analysis using the
Tentacle Drivetrain Calculator.
A: [Mark J.] If you tie two people together can they run twice as fast? Power is additive, speed is not:
You can gear up to get greater speed, but you will get a proportional reduction in torque. Ain't no such thing as a free lunch.
A: [Mark J.] It sounds like one of your re-soldered connections is damaged or has a high-resistance 'cold' solder joint. Check your connectors carefully to make sure they are engaging firmly and completely.
If you have an ohmmeter, check the resistance across each of the three possible pairings of the three lead wires. If one of the three readings is much lower than the other two, the wire not included in that low-resistance pair has the bad connector. If you don't have an ohmmeter sensitive enough to measure the leads, replace the connectors and/or re-solder them all.
The idea would be instead of trying to combine three brushless outputs I would give the three a common shaft and use two end bells for a really long motor. I know the windings of the three motors would increase resistance but that should also allow me to increase voltage to get my current back up where optimal should be?
I figured it would be lighter (one gear reduction instead of a three motor combiner and then a gear reduction). Only two end bells vice three. Thoughts?
Thanks - Team Boom Bots, Stafford USA. [Virginia]
A: [Mark J.] Oh hell no! Why do 'builders want to make things so effing complicated? Complex is the enemy of good.
Buy an outrunner with three times the power of your small motors that runs at whatever voltage you want and be done with it.
Q: I like to make things difficult and they don't make a motor in the size and shape I am contemplating.
A: Everybody likes to do things their way, but if you like things complex and difficult you're writing to the wrong guy for advice. Two motors on a common shaft isn't a bad plan to eliminate a 'combiner', but adding a third motor and series winding the whole stack is asking for grief. Best luck.
A: [Mark J.] It's simple math.
A: [Mark J.] I just did a re-write on Frequently Asked Questions #23 -- see if that helps.
I found that Kt can be calculated by 1352.4/kv, using this I enter the information from hobbyking into the robot drive calculator, but the numbers always seem very, very poor, I must be missing something.
I realize that "max amps" isn't stall amps on hk's motor stats for the calculator but it is the best I have, how do I make a decision on what motor to use, or is there any easy math to choose a brushless as opposed to brushed motor? [Dublin, Ohio]
A: [Mark J.] The formulas that apply to torque for brushed motors do not directly apply to hobby brushless motors, and there are good reasons why such motors don't give figures for 'stall torque' and 'stall amps'.
Brushless motors are not easy. Hook one up to a battery and it will twitch and then quickly melt. They require an external motor controller to supply to each of their three input leads with power of the correct polarity at just the right instant and then change everything an instant later as the motor rotates. The timing and amount of current supplied to the motor is controlled by software in the motor controller that monitors the speed of the motor and adjusts everything thousands of times a second.
Now here's the problem: hobby brushless motors and controllers are made for model aircraft and light little model cars that don't rely on low-speed pushing torque. The controllers aren't designed to provide the current needed to produce a lot of motor torque and their software severely limits current flow when the motor is operating at low RPM, protecting the controller but killing low speed torque. The motors are also designed to run in their upper RPM range and cannot handle the heat generated by high current levels.
This all explains why brushless motors don't list values for 'stall torque' -- low speed torque depends on the combination of motor and controller and is limited by the controller software. The torque and power curves for a brushless motor are not predictable from the motor specs, they don't look much like the brushed curves at all, and the curves for the same motor will be different with different controllers. It is possible to hack into the controller software and modify the parameters governing low speed torque, but you're about as likely to blow up the controller and/or motor as you are to improve performance. It's all a black art.
So, how do you select a brushless drive motor? Overkill. Builders are selecting drive motors based on rated output and mass. For guidance on picking brushless motors for both drive and weapons applications see: Brushless Motor Selection.
As with all component selection, I advise looking at competitors with similar designs to see what motors and controllers are being used with success. At least then you might have someone to ask for help if you can't get your drive to work!
A: [Mark J.] The answer depends on what drill motor, whether or not you want to save the removed pinion, and what tools you DO have.
The Talon SRX from CTR Electronics is a robust and capable brushed motor speed controller that should be finding broad use in larger combat robots. The controller was developed for use by competitors in the FIRST Robotics Competition with interface and control options that far exceed the needs of combat robotics. All those options inflate the size of the 'user guide' to 40 pages plus a 'software guide' that runs another 140 pages. The size of those manuals make it difficult find the information needed for combat applications.
I consulted with builders who have used the Talon SRX in combat applications and edited the manual down to a 'Quick Guide' just 3 pages long. The quick guide contains only the information needed to install and set up the Talon SRX in your combat robot.
A: [Mark J.] The hamburger is fine.
Once you've put the details of your drivetrain into the Tentacle Calculator the first thing to check is the Amps (per motor) to spin wheels. This value should generally be less than half the 'stall amperage' of the motors. Your design has the wheelspin amperage (0.23 amps) way under the stall amperage (1.97 amps). That's good, but being that far under the stall amperage may be a sign that the wheels are too small or the gear reduction is too great.
Next, click on the Acceleration Calculator button. Insert the width of the combat arena you expect to compete in and click 'ReCalc'. For an antweight the eight foot default is fine as a starting point. You'll notice that your design leaps to its top speed of 2.8 MPH in less than one foot! That plus the low amp draw noted above tells you that the gear reduction is too great.
Close the acceleration window and click on the Gear Ratio Tips button. You'll eventually want to read the whole document, but for now scroll down to the 'Adjusting gearing for special conditions' tab. You can follow the instructions there to try the different gear ratios available for the Polulo motors and different wheel sizes to optimize the robot's performance in an arena of any given size.
A quick recalc changing to the 30:1 Pololu Micro HP motors shows that the maximum motor amp draw is a very comfortable 0.45 amps while the robot now accelerates to 4.4 MPH in a workable 3.5 feet. That's a considerable gain in performance over the 50:1 motors for a six or eight foot arena. Now that you know what to look for you can evaluate other motors for your design.
A: [Mark J.] Yes, you can still get the classic Magmotors -- they're a sponsor of 'Bite Force' in the BattleBots reboot. The high-output robot motors aren't currently listed on their website but you can contact them directly by email (bill.tupper@magmotor.com) or call (508-459-5991) and ask about the classic S28-400 or S28-150 motors. They know about the needs of robot builders and can help with motor selection for big 'bots.
You can see the 'bot in action at the 2016 RoboGames in this Team Velocity playlist. This was the third event for the new brushless drive and Kevin says the brushless drive is working great but that driving feels different and "takes some getting used to."
A: [Mark J.] No. Drive motors in combat robots are wired in parallel to the battery via the motor controller(s). A single 18 volt battery will supply 18 volts to each motor. See Frequently Asked Questions #19 for a diagram of a typical circuit diagram.
Note: The 18 volt DeWalt drill motors are often run at up to 24 volts for greater speed and power in combat robot drivetrains.
A: [Mark J.] Weapons yes -- drive trains generally no. It boils down to expense, low-end torque, and thermal overload. See: 'Reddit: Brushless Motors for Large Bots' for an unusually thoughtful discussion of the reasons big 'bots run brushed drive motors -- at least for the present.
We did have a couple of 250 pound robots at ABC BattleBots season 1 with brushless drive, but neither was a conventional drivetrain:
A: [Mark J.]
A: [Mark J.] The Vantec controllers have an optical isolator between the receiver input and the control electronics. This isolator expects a nominal receiver output signal peak of 4.8 volts, and there are hints in the Vantec RET Series manual that the 411 series is particularly sensitive to signal strength issues (...not for model cars, ...use of y-cables or mixers not recommended).
Your Spektrum AR600 receiver has a signal output peak voltage down in the 3.3-3.6 volt range, so it's likely that the control signal isn't getting through the isolator. Call Vantec and explain your problem. They'll send you a plug-in signal voltage booster at no charge.
It uses 2 AmpFlow E30-400 motors with a AmpFlow 160 controller.
Here is a video of the test. About 10 seconds after the video ends, the controller had a failure, one wheel would only run one direction, the other did not move at all. Examining the controller showed that two MOSFETs had disintegrated.
I am searching for a solution / recommendation; perhaps some way to limit current to protect the controller or a higher capacity controller.
Best Regards
A: [Mark J.] Thank you for the kind words, Jon.
First, let's run your drivetrain thru the
Team Tentacle Drivetrain Calculator to see what type of current load you're putting on the controller:
Inputs:
Let's see the numbers if you change the chain drive sprockets to increase the overall reduction to 17:1 at the 250 pound design weight:
Back to your controller failure:
Suggest you review the controller manual and make sure the settings were correct at the time of failure. Also check the gearbox and drivetrain for binding, but even if there was binding the thermal and current limiting of the controller should have prevented the meltdown you experienced.
If everything checks out, contact the manufacturer (Robot Solutions - contact page) and give them the info and video you sent to me. Until we can figure out why the controller blew there isn't any point in sending in a different controller to fail.
Q: Dear Aaron Archives,
Just writing to update on the progress.
We competed in Robogames, the kids learned a ton from building the robot and competing.
A: [Mark J.] I'm happy to hear that the motor controller issue was resolved, and I'm VERY happy to see you and your boys out competing. You held your own in combat, gained good experience, and you must have had a good time. Nice job! Things like getting the start procedure right will come with a little more experience.
A: [Mark J.] Yes -- see Frequently Asked Questions #25. Linear actuators are powered by electric motors, and overvolting will make them faster and increase their output force -- double the voltage, double the speed and double the force. It will also decrease the life of the actuator, perhaps greatly. Use moderation.
A: [Mark J.] See Frequently Asked Questions #21 for guidance on motors. Assuming that this is for the lightweight flipper 'bot you asked about previously, you might be interested in the DeWut?! Gearmotors utilizing the 18 volt DeWalt drillmotors overvolted to 24 volts.
Q: we can use drill motor in lightweight? I knew u can in featherweight but i thought they were not strong enough for heavier class (lightweight, middleweight, heavyweight)
A: [Mark J.] There are drill motors and there are drill motors, and the DeWalt 18 volt is no ordinary drill motor. Safely overvolted to 24 volts it pumps out a reliable 1.2 horsepower at a very useable output RPM. That's a whole lot better than the common inexpensive hacked cordless drills used in lighter 'bots.
The DeWut Gearmotor Kit places the DeWalt motor and gearbox in a strong and easy to mount cradle with a well-supported 1/2" steel output shaft. That makes a great package for a new builder of larger 'bots.
Want crazy power? Use four of these motors.
A: [Mark J.] There is a LOT of variability in the "1000 RPM Motors" from KitBots, but your battery (assuming it's properly charged) should be ample to cover that variation, and a voltage drop shouldn't cause an ESC to reverse. The problem is likely in the ESC / receiver / transmitter chain. You haven't given me enough info to fully troubleshoot the problem, so let's try a little diagnostic tree:
It's possible that you've failed to mention something important, like a huge weapon motor that cuts in just before the problem crops up. I'll assume that you wouldn't do that to me.
Q: Mark,
P.S. - I forgot to mention that I'm using the built in mixing on the Spektrum DX5e. Do you think getting a vtail mixer might fix this problem?
A: OK, both motors can reverse unexpectedly and control is inconsistent. You've got caps across the motor terminals, all the batteries are charged, and clipping a red power lead from an ESC to the receiver makes no difference. The transmitter is used with other 'bots and performs just fine. Got it!
The transmitter-side mixing isn't the problem. I'm suspicious of two things:
Check for that flashing LED the next time the 'bot acts up -- it's an indicator of reception trouble. Get those antennas out where they can do some receiving!
Q: Mark,
A: Thank you for the follow-up!
I'm glad you were able to track down the problem, but you never told me that the 'bot had a weapon ESC -- even after I prompted you in the first step of the diagnostic tree. It's really tough to correctly diagnose a problem with a component that a builder won't admit exists.
A: [Mark J.] I think this is a record -- three different drive motors in one 'bot. May I assume that all the wheels are the same size?
You haven't told me enough about the motors to do a full analysis, but in spite of what some builders might tell you this shouldn't cause any major problems.
As your robot accelerates the torque available from the front motors will taper off more quickly than the available torque from the rear motors. At some point the front motors will no longer be able to 'keep up' and the rear motors will take over the entire load. There will be a little drag on the front wheels at speed, but it shouldn't be a cause for concern. Many all-wheel drive automobiles deliberately design a front/rear torque imbalance into the drivetrain to improve handling -- they drive just fine.
The small difference in speed between the front motors might cause a small 'pull' to one side, but again I believe this will be barely noticeable.
If you're really worried about this you could fit 40% larger diameter wheels to the front of the 'bot (or 30% smaller to the rear) or you could provide differing voltages to the front and rear to level out the motor RPM. I wouldn't bother unless a real handling problem develops. Give it a go; I think you'll be fine.
A: [Mark J.] That's a little bit like asking 'How long a rope do I need to climb down a cliff?" The answer depends on the cliff.
My bottom line advice is to not build a heavyweight robot on a limited budget with E-series AmpFlow motors and marginal ESCs. The other builders in the class are not going cheap, and you're not going to have much luck in combat if you do.
Now, about the weapon: see FAQ: Spinning Weapon Design for Combat Robots.
I'm designing a MW (120) and I'm planning on having 6" wheels. I understand that designing so that the wheel-break torque at 1/2 stall is best from a motor optimization perspective. But I seem to be stuck in a strange balance. For example, using an F30-150 geared down 10:1 will give me top speed of 12.2 mph, acceleration of 1.75 sec, and break torque at 55.6 amp which is perfect for most ESCs on the market that can handle 60A continuous. But this current draw is only 18% of the stall torque of this motor. Does that mean it's extremely oversized? I think finding an ESC that can handle almost 150 A continuous would be a little absurd for my bot, which is a spinner and won't be pushing most of the time. Downsizing to an E30-150 would make my accel a bit sluggish.
I guess my question is.. does the F30-150 seem like a good choice here? Or would I be wasting money since I won't be coming close to the peak HP of the motor?
Thanks and happy holidays. - James [Long Island, New York]
A: [Mark J.] I really like to see this type of question. You've been thinking your design thru and something isn't adding up -- a perfect use of Ask Aaron! Let's take a look...
Gearing the drive motors for tire breakaway at about 50% of stall torque allows the motors to achieve maximum horsepower output while protecting them from excessive current consumption. That's 'optimum' for the motors, but if the motor power is not adequate for the desired acceleration of the robot in a specific arena it may be necessary to change the gearing to something less than optimal to compensate. Other factors, such as your desire to use less expensive ESCs, can take you farther away from 'optimal'.
Assuming that you're building for the RoboGames arena, you'll have about 16 feet of space between the front of your robot and the center of the arena. That's a good distance to use for calculations. Here's what different gear ratios do to your maximum current draw and to the acceleration time and top speed in a 16 foot sprint:
You can see that in this arena dropping the gear reduction down into the 6:1 to 8:1 range improves the practical top speed of your 'bot without significantly impacting the acceleration time. It may also improve the drivability of the robot by reducing wheelspin on acceleration. Pushing power is unaffected. The cost is greater current consumption by the motors, which requires more expensive ESCs.
Note: the
Tentacle Drivetrain Calculator
assumes four wheel drive in it's calculations. If you have two-wheel drive the reduction in weight on the driven wheels will cause them to break traction at lower torque levels. See the pop-up help files in the Team Run Amok version of the Tentacle calculator to learn how to adjust the calculations for a two-wheel drive robot.
Bottom line: your motors aren't overpowered, you're simply choosing to not use their full power in order to save on speed controller costs. The motors are a good choice for a weaponed middleweight. You just have to decide if squeezing extra performance from them is worth the added expense given your weapon design and attack strategy. The F30-150s do give a good performance boost over E30-150s even at the reduced current level.
I've added a new section to the Optimum Robot Drivetrain Gearing page to better explain adjustments to 'optimum' gearing to compensate for arena conditions or ESC choice. Thanks for pointing out this source of confusion.
A: [Mark J.] Your IP address whispers San Mateo, but your 'bot design screams India. I no longer answer combat robot questions for builders competing on the Indian subcontinent.
Click here to learn why.
A: [Mark J.] The DeWalt 396505 series motors do have 15-tooth pinion gears, but those gears have a non-standard pitch that will not mesh with BaneBots gearboxes.
It is possible to remove and replace the DeWalt pinion with a standard pitch gear, but it isn't easy. The mounting holes for the DeWalt motors are also spaced differently than the mounts in the BaneBots P60 gearboxes -- they aren't gonna fit without extensive modification.
If you want to use the DeWalt 396505 series motors in a robot drivetrain I'd suggest using the DeWut?! gearmotor kits that are designed for that purpose.
A: [Mark J.] Your questions have been discussed previously here at 'Ask Aaron':
A: [Mark J.] I'm confused as well -- what exactly is your question?
A: [Mark J.] On my workbench I have a block of wood that is 9 cm wide and 4 cm thick. Can you tell me how long it is? You can't, because there is no formula to calculate length given only width and thickness.
With proper equipment the stall torque of a motor can be measured, but there is no formula to calculate the torque output of a motor based only on its speed and the applied voltage. If you know the manufacturer and the model number you may be able to find a data sheet for the motor online that includes the stall torque.
Thank you for the answer about the RS775s, it was very informative. Banebots also seems to be discontinuing their motor stock, so using something else is more desirable. I have a few more questions if you're willing to answer them.
For the 60lb wedge/lifter my friend and I are planning, do you think two Dustin motors at 24v would work? The veteran builder who is helping us might have some leftover from his heavyweight, and has offered them to us. They would drive the four 5" wheels via chain (we don't know whether they are the high or low speed, or if they can be switched, but we would use a 1.2:1 increase for low speed motors and 1:4 reduction for the high speed motors).
I've seen videos of 60lb robots going about 7MPH (with active weapons) and 12MPH (no active weapon). Using the motor torque calculator, this motor/wheel/voltage combination would give a top speed around 8MPH. Do you think this speed is enough to be competitive (assuming we get lots of driving practice)? [San Diego]
A: [Mark J.] The Dustin motors [no longer in stock] are easy-to-mount, battle-hardened conversions of the DeWalt 24 volt hammer drill motors and gearboxes. The modifications include improved brush holders and end-caps that provide good reliability up to 36 volts -- more than doubling the power output of the motors at 24 volts. A pair of Dustins at 24 volts would provide adequate performance for your lightweight lifter, but you're missing a bet by not running the motors at higher voltage.
You can certainly try the 'bot at 24 volts and see if it suits you. If you use speed controllers that can handle higher voltages you can bump the voltage later if you decide you need greater drivetrain performance:
A: [Mark J.] What new wedge? Original Sin has different wedges that it can mount against different opponents, but I don't know of a new one.
Do you mean the yellow spinner-trap? It isn't new -- it dates back about three years. The spinner trap was developed to control the massive horizontal spinning bar on 'Last Rites' and bring it to a stop. If you're facing similar opponents it's a reasonably effective design, but it doesn't work well against drum spinners.
A: [Mark J.] Magmotor Technologies Inc. is a manufacturer of industrial motors and related components. They make a wide range of industrial motors covering a broad range of specifications. Their motors are designed to provide reliable service over a long service life under continuous operation.
About 15 years ago, combat robot pioneer and champion builder Carlo Bertocchini worked with Magmotor to produce a special series of motors designed specifically for combat robots. These new motors were designed to provide much greater power output than standard Magmotors over shorter operating cycles. Carlo initially sold these high-performance motors as a variety of Magmotor, but for several years now he has marketed motors, motor controllers and gearboxes exclusively under the name AmpFlow.
The current AmpFlow motors resemble Magmotors, but they are built by another supplier. Magmotor continues to produce a wide variety of motors -- some suited to combat robot use and some not.
Q: Can you also let me know as to what is the best way for a drum bot to defeat an well armored, ramp or wedge, invertible bot?
A: Speed and maneuverability are important in drumbot attacks. The last thing you want to do is slowly move forward and grind uselessly on their wedge-- that's playing to their strength. You must get into a position where you can use your weapon against a part of their 'bot that has hard edges for your impactor to grab. Attack the outer edge of their wedge at an angle, or rush in against unprotected wheels. Keep hitting them 'til they break. Watch videos of 'Tauro' to see how it's done.
I'm using a 2.4GHz system.
Lastly, if you could share some comments on the Vantec, it'd be of great help!
A: [Mark J.] The classic Vantec controllers were designed more than 20 years ago. In the early days of combat robotics they were the best controllers available, but that is no longer the case. I can't think of a competitive combat robot built in the last 12 years that uses one. They are bulky, complex (as you have found from reading the manual), expensive, and lack features commonly found in more recently designed ESCs. If you look thru the Ask Aaron archives you will not find a single instance where we recommend a Vantec controller.
For purely informational purposes:
A: [Mark J.] You have several problems:
As a general rule, a combat robot requires a minimum power supply of 10 watts per pound -- that's 1320 watts for a 60 kg robot. You're trying to do it with 120 watts. You'll need to reduce the weight of your robot in order to provide even poor performance. A 30 kg robot with 4" diameter wheels powered by four gearmotors providing 40 kg-cm stall torque and 150 unloaded RPM will generate maximum pushing power without motor stall and give a (very slow) 1.7 MPH top speed. I think that's about the best you'll be able to do with such a limited power supply.
A: [Mark J.] A 60 kg robot is a very large undertaking for someone new to combat robots.
Motor calculations require a bit more information than you have given. The answer depends on the size of your wheels and the number of motors you will use. I'll assume that you will use one motor for each of the four wheels.
Four of your selected motors do not have enough torque to generate maximum pushing power for a 60 kg robot if the wheels are larger than about 8 cm diameter. Speed with 8 cm wheels is only 1.4 MPH -- about half as fast as a slow walk.
The motors are not suitable for a 60 kg combat robot. You'll need either much more torque to allow the use of larger wheels for greater speed, or much greater RPM to provide better speed with the small wheels.
Examples:
My question is should one chose higher spinning motor with larger gear reduction or lower spinning motor with lower gear reduction, and why? I understand that there are many other parameters to consider so that is a more theoretical question. As a practical question, I will use the RS-540 and the RS-555. I understand that my pushing power comes primarily from mass (weight) and wheel traction (friction), so mounting the most powerful motor in the world will not give me greater pushing power.
The RS-555 stalls at 15A, Stall Torque is 29.16 oz-in, Kt: 1.94 oz-in/A, Kv: 646 rpm/V, no-load 7750 RPM.
The RS-540 stalls at 42A, Stall Torque is 39.48 oz-in, Kt: 0.94 oz-in/A, Kv: 1400 rpm/V, no-load 16800 RPM.
So is the RS-555 a viable option? At first glance it seems that a smaller gearbox (less weight, less friction and other mechanical losses) and less stall Amps is a big bonus. Considering that I will be using an adequate power pack and a Scorpion XL ESC, the only things I'm noticing are a big difference in acceleration and the fact that the torque to spin wheels is much closer to the stall torque than for the RS-540. The problem is that I don't have enough experience with this to "feel" and interpret these numbers properly, so I can't tell if the RS-555 will be a complete dud compared to the RS-540 or not. Thanks! [Tarrytown, New York]
A: [Mark J.] Thank you for the very complete information in your question. It makes my job much easier if I don't have to guess at the details.
The issue of motor speed versus torque and the possibility of using a lighter gearbox pretty much disappears when simply matching up available components to desired performance. Bot Hockey matches (video) are all about close-quarters maneuverability and acceleration -- not top speed. A hockey robot geared for 13.8 MPH that takes 4 seconds and 50 feet to reach that speed is not making best use of its motors.
A top speed around 5 or 6 MPH is perfect, with acceleration to that speed in just a few feet. Although the weight savings are tempting, you're not going to get that performance from the 5:1 gearbox. The numbers I get from the Tentacle Calculator are a bit different from yours, but close enough. I can't find a good combination of speed and acceleration using the RS-555 motors and the available P60 gear ratios with 3" wheels, but other available motors come out well:
Four RS-540 motors with 26:1 P60 gearboxes and 3" wheels
Or, you can save a bit of weight with lighter motors and still have good performance:
Four RS-395 motors with 26:1 P60 gearboxes and 3" wheels
If you're willing to work with a different wheel size, you may be able to find a combination of wheel and gear reduction that works with the RS-555 motor, but you're not going to get the weight savings you were looking for. Plus, the lighter RS-395 motors put out the same power as the RS-555, just at a higher RPM.
A: [Mark J.] Very slow for a wedge -- less than 4 MPH -- but ample torque and acceleration. Six inch wheels would be better, or the 12:1 gearbox. Use the
Tentacle Drivetrain Calculator and you can check drivetrain performance yourself.
Q: CIM guy..
A: As I said, you may check drivetrain performance for yourself with the Tentacle Drivetrain Calculator. My recommendation for larger wheels or the 12:1 gearbox stands.
A lifting wedge? I cannot recommend a motor/gearbox for a lifting mechanism without full details and measurements of the proposed lifter. In general, the BaneBots P60 gearbox is unsuitable for such purposes as its maximum torque rating is 35 foot-pounds -- certainly not enough to transmit power for a lifter in the 100 pound class.
A: [Mark J.] Small and slow robots sometimes are controlled by mechanical switches with power cables running to the robot. This was the method used for years in Indian robot combat, but there are many problems:
Electronic Speed Controllers (ESC) offer smooth control of drive and weapon motors by providing full control of the speed and direction of rotation in response to radio control input. They plug directly into the radio receiver and are available in current capacities suitable to control any robot drive or weapon motor. Fine control of speed and direction is critical for high-speed attack and defense.
There are MANY posts about the selection and usage of ESCs for specific robot requirements in this archive.
A: [Mark J.] The AmpFlow G43-500 motor is a big, heavy, slow lump of a motor with low power output for its size. It's a poor choice for combat robot applications.
For $10 more you can buy the AmpFlow E30-150 that is smaller, weighs less than half as much, has a standard output shaft, comes with a proven record in combat robots, and delivers 50% more power.
A: [Mark J.] Mark Harrison gave me a photo of the Sabertooth 2x60 circuit board with the bottom plate removed and some information about it. The plate is held to the board by the two fan screws and a bead of adhesive. If you're careful you can remove the fan screws and pry the board free without damaging it, but I don't know why you'd want to.
A: [Mark J.] You may have noticed that there are VERY few electric lifters like 'Sewer Snake'. The problem isn't the motor -- lots of motors have the power needed to hoist your opponent off the floor. The problem is finding a gearbox that can take the huge torque loading without exploding! 'Sewer Snake' uses an AmpFlow A28-150 motor connected to an industrial 25:1 gearbox that feeds a 3:1 chain reduction [photo at right]. The torque available at the lifter axle is a whopping 10,650 kg-cm!!!
Here's the formula for the torque needed for a simple single-pivot lifter like Sewer Snake's:
So, for a 60 kg class robot with a 50 cm lifting arm the torque required is:
Now, your motor has to provide that much torque while still spinning at a reasonable speed, so you can't use motor 'stall torque' (zero RPM) as a starting point. A PMDC motor develops max horsepower when loaded to 1/2 stall torque, so we use 1/2 the motor stall torque in the next calculation.
Example: an AmpFlow E30-150 motor has 51 kg-cm stall torque. If you require 3000 kg-cm output torque at half motor stall torque, the reduction ratio required is:
When loaded to 1/2 stall torque, PMDC motor RPM is 1/2 the free-running speed. So the output RPM under load is:
The E30-150 motor has a free-running speed of 5700 RPM @ 24 volts, so:
You need less than 1/4 of a revolution to lift your opponent, so that's reasonable lifting speed. So the motor is suitable, but the problem is finding a 118:1 gearbox that can survive 3000 kg-cm of torque! A BaneBots P60 gearbox is rated for less than 500 kg-cm.
A: [Mark J.] First, why are you building a '25-30KG' robot for '60kg' events? That's like showing up with a knife at a gun fight. Give yourself a chance and build to the weight limit.
NPC-41250: heavy (7.5 pounds), slow (3 MPH @ 24 volts w/6" wheels), awkward mounting, cast aluminum gearbox housing is weak. Suggest that you use the
Team Tentacle Drivetrain Calculator to evaluate motors for your application.
A: [Mark J.] Your questions about the 85 amp BotBitz ESCs have been previously answered here at Ask Aaron. Search this archive for 'BotBitz' to find multiple posts.
BotBitz takes inexpensive Chinese brushless ESCs and installs new firmware for use with brushed motors. The '85amp' version is popular in Australia for featherweight (30 pound) combat robots, but these are very basic 'no frills' motor controllers. There seem to be some quality control issues with BotBitz controllers that may trace back to the original Chinese manufacturer.
You should consider the Talon SRX ESC as an alternative. The Talons have been well tested in recent US competition and have proven themselves reliable in the lightweight (60 pound) combat class.
A: [Mark J.] Your calculation of 571 amps is correct for a 'theoretical' stall current of the AmpFlow A28-400 motor at 24 volts, but the 'real life' stall current will be lower due to the added resistance of other elements in the motor circuit (controller, connectors, wire, battery...). I don't know exactly how those other elements are accounted for in the Robot Marketplace figure, but 390 amps is likely close to the true stall current of the motor in a typical combat robot application.
A: [Mark J.] Carlo Bertocchini went to considerable effort to design, and produce a series of PMDC motors ideally suited to the demands of combat robots. AmpFlow motors are so widely used in combat robotics because there is no cheap substitute.
You can't get a documented 2 hp in that RPM range with anything close to the reliability, efficiency, and ease of use of the AmpFlow in a 'bargain' motor. If you could, you wouldn't need to ask me what motor that was 'cause a great many robot builders would be using them.
A: [Mark J.] I can't tell you what motor will be best in your budget because:
'Cheap substitutes' are cheap for a reason. Without performance specs you're just guessing, and you're likely to throw your money away on an entirely unsuitable motor.
I suggest that you read the Ask Aaron Spinner Weapon FAQ for guidance in weapon design and motor selection.
A: [Mark J.] It doesn't sound good -- but if you're lucky the ESC may have just lost its calibration settings. Try resetting to the default:
A: [Mark J.] I'm very sorry, but Aaron is not available to answer your question. I'll do the best I can.
I can't recommend a speed controller based solely on the motors that a robot will use -- see Frequently Asked Questions #21. The speed controller amperage capacity requirement for your motors will depend on design factors such as:
The merits and drawbacks of the Roboclaw and Sabertooth controllers have been discussed in previous posts on Ask Aaron. I suggest that you search this archive for those posts.
Very briefly:
Do your homework and find out how much amperage your motors are likely to use, then you will be better equipped to select the proper controller for your purpose.
A: [Mark J.] Not terribly difficult -- but maybe a bit of an expense. Since you ask about radio gear farther down, I'm guessing that your current robots use wired controls to actuate the contactors and that you are interested in going wireless. Frequently Asked Questions #19 features a wiring diagram of the components for a typical combat robot. Compare that to your current robot and you should have a good idea of the modifications you will need.
Q: Also can you suggest me a good 2 channel speed controller giving out 30 - 40 amps per chanell continuous current at 48V or 36 V . I am using two motors rated at 36V 750 watt for drive overvolted to 48 V . I want to buy a speed controller or higher rating so that I can use it with other motors in the future .
A: It's a good plan to look to future uses of the controller and invest in something that you can use for a long time. Your list of requirements is difficult to meet, largely because of the need to control up to 48 volts. The only speed controller I can recommend is the RageBridge -- which is a very fine ESC that can handle up to 50 volts at up to 40 amps continuous current. Note that 50 volts is an absolute maximum for this controller, and that four freshly charged '12 volt' batteries in series can exceed 50 volts. Be cautious.
Q: Do you have any idea about roboclaw speed controllers ? the 30 and 60 amp version .
A: According to the RoboClaw manual:
The RoboClaw HV 2x60 Dual Motor Controller is more expensive than the RageBridge and much more difficult to set up. It has a great many features that are not used in combat robots, and lacks the true current limiting feature of the RageBridge. I can't recommend a RoboClaw for your purpose.
A: [Mark J.] I'd need much more information about your 'bot than just it's weight before I could recommend a specific gearmotor.
A: [Mark J.] The correct substitute depends on why you want a substitute.
A: [Mark J.] Estimating ratings for switches outside their listed voltage range is a tricky business.
If the switch does fail, it will likely weld the switch contacts 'on' and be impossible to turn off, which could be disastrous in a robot. I'd strongly recommend that you find an automotive switch properly rated for DC usage at the current you need for your motor control.
A: [Mark J.] Silicon controlled rectifiers (SCRs, also known as thyristors) are commonly used to control AC motors and can be used to control DC electric motors (examples). However, they are far from ideal for the type of control circuits desired in combat robots. Combat ESCs are based on highly efficient MOS-FET devices.
More to the point: I can't recommend trying to 'home brew' an electronic speed controller. A full-featured ESC is a terribly complex design challenge (example). I've seen many attempts by builders to wire up their own controllers. The best of them were disappointing, and the great majority were entirely disastrous. You can't afford to have poor, unreliable motor controllers in your combat robot -- use a reliable and proven ESC or take up a different hobby.
A: [Mark J.] That's not easy to do with push buttons. You'll need four double pole single throw (DPDT) momentary contact push-button switches. The switches must be rated for the voltage and amperage that the motors will use. You didn't mention the stall amperage of your motors, but I think you're going to have trouble finding push-button switches that are capable of handling that great a load.
Assuming that you can locate adequate switches, the diagram shows the circuit for one side of the robot. Duplicate the circuit for the other side.
A: [Mark J.] As I recall, you're building a lightweight (~60 pound) robot with AmpFlow motors geared 8:1 and 6" wheels. Your calculations indicate a maximum amp consumption of less than 30 amps per motor before wheel spin limits greater current use.
I hear multiple reports of Sabertooth controllers giving trouble in combat, but I can't comment based on personal experience. I'm really not sure how big a problem this is, but it may be prudent to pick another ESC.
The Talon SRX controllers are fairly new, but have been used in a few lightweight combat robots using AmpFlow motors with very good results. You should not dismiss these controllers just because they were built for FIRST competition. Note that the Talon ESC is not plug-in ready for R/C control -- it needs to have a standard connector added and may require a firmware update. See the User's Guide for details.
The choice between the two controllers is not simple:
Q: Hey Mark, Sage again.
A: Forgive my scientific skepticism, but I believe that what you've seen are videos where it was claimed that the match was lost because the ESC glitched. If a specific controller gains a reputation - deserved or not - for transient impact failure, it becomes very tempting for a builder to blame a 'bot failure on the easy explanation. How many of these claimed failures are actual ESC glitches, I can't say.
I hesitate to recommend a specific ESC because there are elements in your design that I just don't know about that could greatly influence that choice. The best I can do is to share some comments that may help you in your decision:
Feedback: Thanks a lot Mark! Your reply was bang-on. It was all I needed! I'm going for the Ragebridge...
A: [Mark J.] I suggest you contact AmpFlow for guidance on shaft radial loading limits.
The big terminals on your solenoid are the main power connectors, and the smaller connectors are for the coil. The diagram at the right shows how to wire the solenoid to activate it with a small toggle switch and control the main power output. You will likely want to wire the coil power line into a circuit controlled by the ignition switch, because if you leave the coil energized it will drain the battery. That's what you get for asking a truck question at a robot Q&A site.
Note: disconnecting the main battery power in a vehicle with the engine running may cause a voltage spike that can damage the alternator. A more complex switch is needed if you want to safely shut down the electrical system in a running vehicle.
A: [Mark J.] There are LOTS of different automotive starter motors with a very wide range of performance, and none of them come with the type of specs a robot builder needs to calculate their suitability for a spinning weapon. Also, starter motors are designed to operate intermittently for a few seconds at a time, not continuously for several minutes. I can't think of any champion robots that use starter motors to power their weapons -- can you?
I really can't give you any advice on the unspecified motor you plan to use. It might work well, it might be completely unsuitable. Personally, I recommend using components that have full specifications and a proven record in combat robots. Let other builders research new motors and risk failure in combat conditions.
A: [Mark J.] They didn't send you the user manual? Download it:
A: [Mark J.] You made a chassis before you decided on motors??? All the components of a combat robot must work together, and stuffing motors into a chassis not designed to match their requirements and capabilities is a poor start.
See Frequently Asked Questions #21 for general motor guidance and a link to a tool that is very useful in evaluating possible motors.
I don't know what motors are available to you, and you haven't given me enough information about your robot for me to be able to make a recommendation. If you would like to send the specifications of a motor you are considering I might be able to evaluate its performance in a 20 KG robot -- but I can't blindly recommend a motor.
A: [Mark J.] You bent a 1/2" hardened steel motor shaft? I'd carefully evaluate the weapon/drive design to better isolate the motor shaft from such high loading. AmpFlow motors have a nice long shaft, but you should mount the pully/sprocket/gear as close to the motor bearing as possible to keep excessive loading off the shaft.
Straightening a large diameter shaft is unlikely to yield useable results, and it isn't practical to attempt a shaft replacement in an AmpFlow. I think you mark it 'bad' and save it for parts. Combat robots are an expensive hobby.
Thank you from San Diego.
p.s. I just rediscovered your site after a few years, and I am sorry about Aaron's passing.
A: [Mark J.] Thank you for your mention of Aaron. He is always on my mind, and I miss him terribly.
Dale Hetherington has used DeWalt motors mated to his own gearboxes in several of his projects. He offers advice on removing the nonstandard pinion using a custom gear puller. I suspect you could also carefully grind away one side of the pinion with a Dremel tool and more easily pop the gear off. The motor shaft is 5mm and can be adapted to a variety of power transfer options.
You also may be interested in the DeWut?! 3 Speed Gearmotor Kit which adapts the 'new-style' DeWalt motor and gearbox to a strong package with a standard output shaft.
The cordless drill gearboxes also vary a great deal:
In contrast to drill motors, the BaneBots gearboxes:
When I enter the values for my drive motor (2500 no-load RPM, 4.6 Nm stall torque) I get a watt output number much larger than I expected. What's wrong? [India]
A: [Mark J.] You're getting a number about four times too large, right?
You're entering the no-load RPM of the motor and the stall torque -- things that do not happen at the same time. Stall torque comes at zero RPM, and at no-load RPM the motor torque is zero. You need to enter a speed and torque that happen together.
The maximum power output of a Permanent Magnet Direct Current (PMDC) motor is achieved when the motor is loaded to 1/2 the no-load RPM, at which point the motor is producing 1/2 the maximum (stall) torque. To calculate the maximum output of a PMDC motor, the correct numbers to enter into the formula are the values at the maximum power point:
= 1250 RPM × 2.3 Nm torque × 0.105 = 302 watts output
And one weird thing came in my mind that for wedge 600w is nt enough so i was thinking to use two of those motors on one wheel's shaft. [India]
A: [Mark J.] I don't know this 'pankmotor', so I can't comment on how well it will tolerate overvolting. See Frequently Asked Questions #25.
If the motor can survive a 25% overvolt, speed will increase by 25%, torque will increase by 25%, and current consumption will increase by 25%. Since power is the product of speed and torque, the power of the motor will increase by a factor of (1.25 x 1.25) = 1.56 -- so the motor will have a peak power of (1.56 X 300) = 469 watts. More power, but still slow.
Running two of these motors per wheel will generate double the torque for better acceleration, but will supply very little extra speed.
If you have full specs for this 'pankmotor', the Tentacle Torque-Amp Calculator can give you performance parameters for changes in gearing, voltage, wheel size, and number of motors. It's what I use to answer questions like yours.
A: [Mark J.] OK, let's walk thru it. Here are the specs for the Maxon 148866 motor at it's nominal 12 volt rating -- assuming that you can supply the full stall current :
A: [Mark J.] There seems to be some considerable confusion about electric motor power requirements amongst Indian robot builders. The relationships between voltage, current, and motor performance have been discussed in many previous posts to Ask Aaron, but let me summarize them here.
About current An electric motor will consume current based on the voltage supplied and the work load placed on the motor.
A: [Mark J.] You've found a very simple ESC with no reverse current protection, no current limiting, no thermal protection, no LiPo voltage shut-down, no mixing, no midpoint throttle sensing, no DIP switches, no jumpers, no cooling fan, no mounting bracket - no frills at all. I rather like that. There's no complex manual to read thru and no way to get the set-up wrong for your combat application. Just make sure you wire it correctly or you'll fry the ESC.
Most ESC 'protection' circuits function very simply: when they sense a problem they shut down the robot.
A combat robot should turn off ALL 'protection' in combat: no fuses, no LiPo voltage, no thermal, no overcurrent - NOTHING that lets the machine decide it should shut down on it's own except radio fail-safe. It's fine to have sensors trigger audio or visual warnings about overloads, but let the driver decide if it's a good time to ease up or shut down.
A: [Mark J.] I'm still sad. Lucky. Still very sad.
We've discussed your drum weapon extensively, but I don't think I have the details of your robot drivetrain. An AmpFlow E30-150 motor can pull as much as 125 amps at 24 volts under heavy loading, so it seems like wishful thinking to hope that a 32 amp motor controller might be enough -- but there is a simple way to check...
The Tentacle Torque/Amp Calculator can quickly answer the question of the maximum sustained amperage that your motors might be expected to pull given the weight of the robot, wheel diameter, gear reduction ratio, and battery voltage.
Suggest you do a little analysis and see what your actual current consumption might be before selecting a motor controller. You really don't want to be worrying about 'proper care' of your ESC in the middle of a combat match!
Q: lucky again aacording team tentacle amperage calculator motor will need 20 amps continous and 40 amps peak . and [the Sabertooth] 2*32 [motor controller] is able to supply 64 amps at peak and 32 amps regular then sabretooth 2*32 will sufficient I think so?
A: You're misreading the results from the Tentacle calculator, Lucky. If you click on the friendly 'Help' button on the right side of the middle section of the Tentacle calculator display, you will find detailed descriptions of the input and output fields.
Now that the output of the Tentacle calculator is clear - yes, I agree that the capacity of the Sabertooth 2x32 ESC is sufficient for the normal combat operations of your robot.
I do wonder why you believe your SLA battery cannot provide more than 70 amps. I've never seen a SLA battery supply data on absolute peak current. The 'Cold Cranking Amps' (CCA) often cited in SLA specs refers to the maximum current a battery can deliver at very low temperature (0°F/-18°C) for a continuous 30 seconds. Peak burst amperage at normal ambient temperature can be much greater, and certainly enough to fry a small motor controller. Make sure you know what the numbers mean or they'll turn around and bite you.
A: [Mark J.] What you've found is NOT an Electronic Speed Controller (ESC). The Phoenix Series motor drivers are somewhat deceptively advertised relay boards for simple forward/off/reverse operation of two motors. They provide no speed control.
Manufacturers are interested in selling their products and will describe them in the best possible terms. I suggest using products that are in wide use by robot builders with a well-established record of performance. Don't be the first guy in line to discover the problems with a new product. Talk with other builders and use products that are proven to stand up to combat robot use.
Note: it's rarely possible to add current capacity to a commercial ESC with additional heat sinks and fans. The power chips must be designed and mounted in a manner that will allow the added cooling to efficiently draw heat away from the crucial junction where the heat is being produced. If additional heat sinking would add significantly to the capacity of the ESC, the manufacturer would already be doing it - or would at least provide information on additional heat sinking in the product documentation.
Q: Hi Mark, it's Sage. Thanks for pointing out that the Phoenix series was a relay board, I wouldn't haven't even known that! I completely understand that an ESC is an essential part but I just don't have the money left to buy a high end one. I know that using the above relay board has demerits as compared to an ESC out of which lack of speed control is probably the biggest one. Can the Phoenix series still be used in my robot WITHOUT TOASTING IT?? (the motors draw 27A before the wheels lose traction at 8:1 reduction and 6 inch wheels) i'm pretty curious about it's suitability I just wanna know will it handle the load or just burn.
I know that it only offers bang-bang controlling so it's either full forward, off or full reverse... But I'll work my way around it.
A: As I recall from your earlier questions, you're building a 'Breaker Box' style robot -- about which you correctly observed that "being fast & controllable is a necessity." Now you've got motors that will give you the 'fast' part, and you're trying to get the 'controllable' part from a relay board? Rethink.
If the specs on the Phoenix board are reliable it should be able to control your drivetrain without toasting, but you won't like it.
Q: You are right, I guess I shouldn't skimp on an ESC. Originally I thought that the Sabertooth 2X32 Dual Motor Speed Controller or Roboclaw 2X30A Dual Motor Controller would suffice. But I kinda have a bad feeling about them and my gut tells me that I should get the Sabertooth 2X60 Dual Motor Speed controller or Roboclaw 2X60 Dual Motor Controller. As I have to import them the 60A ESCs would end up costing about $150 more than the 30A ones. Should I go for the 60A ones or save them for my next bot? And which one do you think would be a better choice Sabertooth or Roboclaw?
Thanks again and sorry for bugging you with a hell lot of questions!
A: [Mark J.] No need for an apology, Sage. If I didn't want to answer robot questions, why would I run a robot question site?
I would not hesitate to use a 30/32 amp ESC for your current project. But if there is going to be a 'next bot' that will have greater amperage demands, why spend money now for an ESC that you will soon outgrow and replace? It may make better economic sense to buy a larger ESC now.
It's difficult for me to offer a recommendation between the Sabertooth and Roboclaw controllers. Each has drawbacks:
A: [Mark J.] No.
The very first comment at the top of the discussion section just below the product listing answers your question. This is a model aircraft ESC, and airplanes don't back up. General rule: if the product description for an ESC does not specifically say that it features reverse, it does not.
Note: I don't recommend ANY products from the 'H.K.' hobby warehouse for use in combat robots.
Some builders use a few of their products (batteries, radio gear) with some success, but H.K. is known for very low prices -- NOT for high quality, customer support, readable documentation, or bulletproof reliability. In the case of ESCs, their stated amperage ratings are entirely fictional and a fair number of them fail immediately on power-up. IMHO their products are marginally suitable for backyard hobby use, but not for competitive robotics.
A: [Mark J.] As a general guide, a combat robot with an active weapon needs a bare minimum of 4 watts of drive motor output per pound of total weight. Effective rambots, wedges and spinner killers need about three times that much. By that fomula, your 132 pound robot should have at least 524 watts total drive power -- pretty close to what a pair of these motors claim to deliver.
However: I have some doubt that these motors actually deliver 250 watts of output power. If the pole resistance is 2.3 ohms, that works out to about 10.5 amps stall current at 24 volts (24 volts ÷ 2.3 ohms = 10.43 amps). You can get a rough estimate of output power with the following formula:
That works out to less than 65 watts output power (24 volts × 10.5 amps × 0.25 = 63 watts). If your pole resistance reading is accurate these motors will not deliver enough power for your 60 kilo combat robot.
Note: a 250 watt peak output PMDC motor with 3000 RPM unloaded speed would produce about 440 oz-in torque at stall and draw about 45 amp stall current.
Q: hi mark .. the pancake motor guy here ... I think you are right ... my multimeter is having some error for the resistance , probably the wire leads have some of its own resistance
I did one thing ... I have run the motor on 1 volt from the DC power supply and stalled the motor ,it takes 2.2 amps .... from this can I calculate how much power it is rated on 24 v ....is there a formula?
A: The relationship between stall amperage and voltage is linear: 1 ÷ 2.2 = 24 ÷ X, solve for X. That gives a stall amperage at 24 volts of 52.8 amps. Plug that into the formula I gave above for approximate power output and you get: 24 volts × 52.8 amps × 0.25 = 317 watts output power. That is a rough approximation, but it's close enough to the rated output of 250 watts to give me some confidence in that rating.
A pair of these motors geared down 9:1 (330 RPM output) with 4" wheels for a 60 kilo robot would provide reasonable acceleration and good pushing power, but a low top speed for a wedgebot -- less than 4 MPH. You might want to consider four motors and larger wheels.
Save a few dollars in your budget to buy a new multimeter.
A: [Mark J.] Short answer -- no, the Sabertooth controller should be fine.
Longer answer - a motor controller's current rating refers to the maximum current that can safely flow from the battery, thru the controller, and then thru the motor.
Note: most AmpFlow motors will attempt to draw well more than 60 amps under heavy loading. If such amperage is not available, they will not be able to produce their full rated torque and motor performance will suffer. You didn't mention which AmpFlow motors you are using, their gear reduction, the wheel diameter, or the weight of your robot -- so I cannot calculate the current the motors might require under a heavy pushing load. A 60 amp controller may not be sufficient for the motors to perform well in your specific application.
A: [Mark J.] There are two general rules about combining cells in series, and you're violating both of them:
A 30% overvolting of a motor will result in a 30% increase in speed, a 30% increase in torque, and a 30% increase in current. Since power is the product of speed and torque, the output of the motor would increase by 69% (1.3 × 1.3 = 1.69).
How well an AmpFlow E30-400 might cope with this overvolting would depend on how heavily it is loaded. As you have told me nothing about the load you plan to place on the motors, I cannot comment on the wisdom of a 30% overvolting.
A: [Mark J.] The power of a motor is the product of its torque and RPM. Gear reduction reduces the output RPM and increases the output torque, but the product of the two remains unchanged -- save for the mechanical losses caused by gear friction.
Automotive windshield wiper motors have a high ratio worm gear reduction -- around 60:1. Removing the gear assembly would increase the output RPM by a factor of 60 and reduce the available torque by the same factor of 60. You'd be left with a rather heavy and bulky motor that provides around 6000 unloaded RPM with perhaps 60 oz-in stall torque at 12 volts -- about 1/10th horsepower. That's not nearly enough torque to direct-drive a wheel for a robot drivetrain, and not close to enough total power to provide adequate robot performance if it was properly geared.
The smaller, lighter, and less expensive BaneBots RS-550 motor produces similar stall torque and three times the RPM at the same 12 volts, giving three times as much total power. Even if overvolted, wiper motors are inadequate for combat robot propulsion.
A: [Mark J.] The current requirement for an electric motor depends on the load placed on that motor. An automobile will consume much more gasoline running at high speed, uphill, and pulling a heavy load than it will idling along a level road with no load -- and so to will an electric motor consume more current when asked to produce large amounts of power than when is is spinning along under a light load. To calculate the battery requirement for a motor you must first estimate the loading that you will place on that motor.
There are many posts in this archive that discuss estimating the battery requirements for motors used for robot drivetrains, and many posts in the
Examples:
Also, what is the best way to attach drill motors to our wheels? The wheels that we bought have a 1 inch hole in the center where we could attach the motor axle but this hole is too large. Would welding some metal into this hole to reduce the diameter be effective? or is there a better way? [India]
A: [Mark J.] Several problems:
With the gearboxes locked in 'high', the 1300 RPM output with 6" wheels gives a calculated top speed of about 23 MPH but with very poor acceleration. In this configuration the motors do not deliver enough torque to prevent stalling the motors under heavy pushing loads, and because of the poor acceleration true top speed in a 12 foot arena is only about 5 MPH.
Locking the gearbox in 'low' with 6" wheels will give much better pushing power, acceleration, and a useable top speed around 7 MPH. This is still short of good rammer performance, but much improved over locking the gearbox in 'high'. Maximum current draw under heavy pushing will be about 27 amps per motor in this configuration.
It's a poor idea to buy drills to convert to a robot drivetrain without checking to see if other builders have had success using those drills. You will have several problems converting this particular Einhill drillmotor to robot use. The drill housing is all that holds the motor to the gearbox -- there don't appear to be any screws attaching the two components. The housing also holds the components that keep the gearbox in high/low speed mode. It may be best to retain the drill housing and trim away the excess material you don't need. If you're unwilling to retain the drill housing, you'll need to fabricate something like the Dewut?! Gearmotor Kit to securely hold the components in place and lock the gearbox in your preferred speed setting. It's a lot of work, but some bodged solution will likely come apart in combat and leave you without propulsion. A single-speed drill with the motor firmly attached to the gearbox is a MUCH simpler solution.
The junction between the gearbox output and the wheel -- the 'hub' -- is a critical area in combat robot construction. The hub must:
Can you tell us which motor is better to use? The wiper motors seem to be extremely slow so i guess we would have to do some kind of gearing to increase the speed. They also look hard to attach wheels to.
We found some drill motors that had ratings of 15N-m and 500 rpm. Would two of these be enough to effectively drive a 25kg rambot? [India]
A: [Mark J.] I'm weary of saying this -- the Tentacle Torque/Amp Calculator can answer this type of motor selection questions for you. The calculator will provide speed, acceleration, current consumption, and even battery requirement for a robot of a given weight, specified motors, number of motors, wheel size, and battery voltage. Learn to use the calculator!
In general, automotive windshield wiper motors are slow and limited in total power. If you were to gear it up for more speed you would lose too much torque to maintain pushing power. I don't recommend wiper motors for any combat robotics drive.
A pair of drill motors with the rating you give would be much better than wiper motors, but still poor for a 30 kg ramming robot. Acceleration and top speed in a small arena are both less than adequate ramming 'bot performance, although pushing power would be adequate.
Seriously, learn to use the Tentacle calculator.
Q: would 4 30kg-cm motors with a 7.5 stall current and 10 cm diameter wheels be enough to effectively drive a 25 kg rammer bot? I used the tentacle calculator but im not sure what weight to enter since the rammer bot would need to overcome its own weight plus the opponents
A: [Mark J.] ALWAYS use ONLY the weight of YOUR robot as input to the Tentacle calculator. Only the weight pressing down on your drive wheels counts toward pushing power and drivetrain performance. Unless your opponent is on top of your 'bot, its weight does not enter into the drivetrain calculations -- and if it is on top of your robot the only impact it will have on a level surface is to slow your acceleration.
You haven't given me enough information about your proposed motors for me to calculate their performance with the Tentacle calculator. I need the speed (RPM) of the motors and voltage.
The best I can do is to take a guess at voltage and estimate the total output power of the motors to see if that is in the appropriate range for a 55 pound robot. Assuming 12 volts, the peak output power of each motor is about 20 watts (search this archive for the formula to estimate output power from stall amperage and voltage). Total for four motors = 80 watts. That's a bit less than 1.5 watts per pound of robot. A ramming 'bot needs more than ten times that much power to have effective speed and acceleration in a small arena. Even if these are 36 volt motors, you'd still be far underpowered. These motors won't do for your purpose.
A: [Mark J.] See answer to previous post on motor selection three questions down this page. In order to select drive motors you must first define your performance requirements -- in this case acceleration and top speed. Once you have those requirements, the Tentacle Drivetrain Calculator can assist in matching motors to your robot.
Note #1: Mecanum wheels add complexity to the calculations. A robot with Mecanum wheels will have both speed and torque reduced by up to 30% along diagonal axes (download table). To correct for this, increase minimum acceptable acceleration and speed performance by 41%.
Note #2: Mecanum wheels typically have poor traction, so high acceleration rates may be very difficult to achieve with Mecanum wheels regardless of the motors selected. This may severely limit the performance of an omnidirectional badminton-playing robot.
A: [Mark J.] If the cables and switches for your wired remote are capable of handling the large amperage required by the AmpFlow E30-150 motors, you can control the motors directly from your remote. However, the E30-150 motors can pull as much as 125 amps @ 24 volts if stalled; high current requires heavy cables and switches. It would be more effective to use solenoids on the robot and run only a small controlling current thru the cables and switches in your remote.
You will also find it very challenging to properly control your robot with simple forward/off/reverse control of the drive motors. Read thru our Solenoid Control FAQ for more information.
A: [Mark J.] See previous answer two posts down this page.
Team Run Amok does not compete in sumo so I cannot personally recommend motors for the sumo classes. I do know that some successful competitors use the 12 volt Maxon RE40 motors in the 3kg class, but they are VERY expensive.
I suggest that you find builders who compete in sumo to discuss drivetrains.
A: [Mark J.] The answer depends on how fast you want the 'bot to push that weight, how fast you want it to accelerate, and the diameter of the driven wheels. A motor with very little torque and speed can be geared down to push a 60kg 'bot and a 40kg weight slowly, but if you want more speed you'll need more power. Examples:
You can't calculate the power you need until you figure out your performance requirements. There are MANY posts discussing motor selection in this archive.
A: [Mark J.] We don't compete in sumo, and my initial reaction was that a pair of RS-550 motors can provide WAY more than enough power for a 3 kilo robot, even with magnetic downforce. Then I went out and found some video of recent 3 kilo sumo matches. YIKES! Take a look!
If you're competing at this level, I think you need to find someone who is a current competitor in 3 kg sumo to discuss drivetrains. The RS-550 motors may be adequate, but the BaneBots P60 gearboxes are probably too heavy for your purpose -- they are built to take heavy impacts that you aren't going to have in sumo. I suspect you'll want more than a 16:1 gear reduction to get that lightning acceleration in a very small arena -- maybe 26:1? Sorry I can't be of more help.
A: [Mark J.] Let's see if I have this straight: your 30 kilo robot with two E30-150 motors (2 horsepower total) got out-pushed by a 30 kilo robot with four gearmotors (2 horsepower total), and you've decided that the problem was that you didn't have enough horsepower? Think about that for a minute.
Adding more power can give you greater speed and/or greater acceleration -- but once you have enough power to break traction and spin your wheels in a pushing battle, adding more power will just spin your wheels faster without generating more push.
Pushing ability is limited by the weight pushing down on your drive wheels and the 'grippiness' of your tires. If you want more pushing power you'll need to look to those factors to find it. Throwing money into more power isn't going to get you there.
Back to your original question: no! AT A GIVEN VOLTAGE a permanent magnet DC motor generates maximum horsepower WHEN THE LOAD PLACED ON THE MOTOR RESTRICTS ITS SPEED to 50% of maximum RPM. Artificial methods of restricting the RPM (reduced voltage, artificially induced load) simply reduce the total power available. See this post in the archives for a more complete explanation and a graph.
A: [Mark J.] I'm rather sad, Lucky -- but thanks for asking.
I've previously answered this question. Search this archive for "typical PMDC" for the explanation of why this doesn't work.
A: [Mark J.] A properly geared robot propulsion motor should be able to provide more torque than the wheels have traction. When torque exceeds traction - when the robot is pushing hard against an opponent - the tire will break traction and start to slip. You'll still be getting full thrust from the slipping tire, but the slippage will prevent the motor from stalling and will keep amperage consumption under control. I generally design for between 50% and 100% more stall torque than the wheels can directly use for traction.
The Team Tentacle Torque Calculator will help with the calculations to determine the best gearing for your robot's weight, motors, wheel diameter, and voltage.
See also: Optimum Gearing for Combat Robots.
A: [Mark J.] For any gearbox,
I say 'approximately' because there is some small loss of power due to the gearbox friction and inefficiency -- maybe 15%.
For the BaneBots 18 volt RS-775 motor and a 64:1 gearbox.
Note: that is a HUGE stall torque output -- more than 50% over the rated torque for the gearbox. Make sure your drivetrain design does not allow the wheels to approach stall under operational conditions or you'll risk gearbox failure. Depending on your robot weight, wheel diameter, and number of gearmotors used, you may be better off with the BaneBots 16:1 gearbox.
My question is, where does the power for the coil draw from? The particular solenoid I'm planning to use has a 'coil voltage' of 24V. The 'contacts' are rated up to 48V- In this instance I am powering a 36V motor by means of the contacts. Does that mean the 24V to activate the coil must come from my receiver battery, since that is what my Battleswitch is wired into? That battery is only 6V - am I risking not having enough power to activate the coil? What else can I do?
Thanks for the help. [Philadelphia, Pennsylvania]
My recommendation is that you find a solenoid with a 36 volt coil to match the voltage you will supply to your motor. If you absolutely have to use the 24 volt coil solenoid, write back and I'll sketch a diagram of how to 'tap' your battery pack to supply 24 volts to the coil and 36 volts to the motor.
Thanks!
A: Yes, you can get away with running two identical spec NiMH packs with different voltages in series. Check the diagram on right for the circuit to run 24 volts to your coil and 36 volts to the motor. The current draw for the coil is small (about half an amp) compared to the motor draw, so it shouldn't cause an imbalance problem. I think you'll be OK.
Q: Hey Mark. Philadelphia here again. I was testing my robot's weapon using the 24V-12V combined batteries to get my 36V to my motor, with the 24V split to activate the coil as we discussed. First I ran the weapon using only a 24V battery and I simply used a 'bypass link' so i didn't have to attach the 12V battery just yet. Everything worked fine. I then plugged my 12V battery in to see how it did with more juice. When i turned the weapon on, I immediately drew so much current that both batteries (NiMH) began to smoke and melted their shrink wrap. I didn't expect the additional voltage to drastically increase the current draw. Is this expected? Motor is a F30-150 by the way. Should I play it safe and keep it running at 24V? thanks for the help!
A: We hadn't discussed the details for your weapon system. Increasing the voltage by 50% will also increase the maximum amperage draw by 50%, and at 36 volts the F30-150 AmpFlow motor can pull well in excess of 500 amps near stall. A 500+ amp draw for even a moment could seriously tax even a very high capacity battery.
Check the recommended maximum current draw for your battery packs and your relay, then carefully check the electrical circuit against the diagram. If everything checks out and the weapon runs well at 24 volts, I'm going to recommend sticking to 24 volts to avoid overloading the battery and relay.
A: [Mark J.] What's so magical about 500 RPM? The gear reduction should be chosen to make best use of the motor power to maximize performance of the robot in the arena, not to achieve a specific RPM output.
I suggest you use the Team Tentacle Torque Calculator to determine the best gearing for your robot's weight, number of motors, wheel diameter, and voltage. If you're really set on 500 RPM, the Small Johnson spins 18,720 RPM @ 12 volts -- do the math yourself.
Shaft diameter is the same as the standard RS-550 motor, so you could adapt a gearbox for that motor if you like. Even better, just use a BaneBots RS-550 motor instead.
A: [Mark J.] The 'Johnson' motors available in India must be a much different motor than what we call a 'Small Johnson' in the US. A US Small Johnson geared down to 200 RPM would have more than 500 kg-cm stall torque at 12 volts.
In my opinion, 24 volts is too much for a 'throwaway' 12 volt can motor. A 'Small Johnson' is only $6.49 -- you can't expect a lot of unused capability in so cheap a motor. You might want to consider finding out if a 'real' Johnson or RS-550 motor might fit your gearbox. Running six of those motors at 12 volts would give you a much more powerful and reliable drive than the six weak motors you're running at 24 volts.
A: [Mark J.] 'Brushed' motors like the AmpFlows do not require an external controller to run -- simply provide suitable DC power across the leads and they will spin. 'Brushless' motors have no mechanical commutation and require a suitable electronic controller to supply power to the correct field windings at the correct time. Without a controller, brushless motors will not spin. See Frequently Asked Questions #23.
For general information on overvolting see Frequently Asked Questions #25. Use caution in overvolting high performance motors like AmpFlows - they are already providing very high output for brushed motors of their size and weight. I recommend evaluating the performance and function of the motors at their rated voltage in the specific application you have for them and increasing voltage ONLY if the motors are showing no signs of stress.
A: [Mark J.] Doubling the voltage will approximately double the RPM, and will double stall torque if adequate current is available from the battery. Doubling the voltage will also shorten motor life.
A: [Mark J.] The word from Indian builders is that you can order AmpFlow motors direct from the AmpFlow website for shipment to India. Be prepared to pay as much as $150 for shipping, a 41.5% duty (payable in person at the customs office in Mumbai), and a few hundred rupees in assorted courier charges.
The preferred method of obtaining an AmpFlow motor is to find someone travelling from the US willing to purchase your motor and bring it to India. They will pay only a 37% customs tax at the airport.
I can't confirm that info, but it should give you some idea about what is involved.
A: [Mark J.] Unfortunately, there are a few catches to your plan, Las Vegas. The graph shows power, typical efficiency, and torque/amps at constant voltage when RPM is limited by the working load.
A Permanent Magnet Direct Current (PMDC) electric motor produces its greatest power when the work load upon the motor requires a torque level which restricts its speed to half the unloaded RPM for a given voltage. If factors other than motor loading are restricting the RPM (reduced voltage, artificially induced load), the net motor output is also reduced.
The only valid method for doing what you suggest is by use of a continuously variable (CV) gearbox that would decrease the gear reduction as the robot gained speed to keep the motor loaded to a specific RPM. CV gearboxes are larger, heavier, more expensive and much more complex than a simple single-speed gearbox -- which defeats the gains you were seeking.
A: [Mark J.] A rambot/pushybot needs more drive power than a robot with an active weapon. For ramming speed and pushing power in a small Indian arena I'd suggest a combined output of least 360 watts from your motors.
I don't know all of the motors available in India, but I understand that a version of the RS-550 motor (also called a 'small Johnson') can be purchased from local suppliers. The versions of this motor available from US robot suppliers are typically in the 250 watt output range. A pair of these motors geared down about 20:1 with 2.5" to 3" wheels would give good performance in a 20 kg 'bot.
The Team Tentacle Torque Calculator is an excellent tool for selecting appropriate motors for combat robots of all weights and configurations.
A: [Mark J.] The output of an R/C receiver cannot directly control a motor -- you will require an interface to convert the receiver output signal into current/voltage to supply to your motor.
Depending on the specifics of the use you have for that motor (robot propulsion, spinner weapon, lifter, etc.) you may need simple on/off control, the ability to reverse direction, or full variable speed control.
Q: Hi Mark!
A: The answer to your question is in the blue box at the very top of the solenoid guide that I suggested you read:
For weapon or drive motors that must be able to reverse direction, an electronic speed controller costs very little more than a solenoid system of comparable capacity. An ESC is more reliable, lighter, less bulky, and provides much greater control.
For on/off with direction reversing you'll need two reliable DPDT relays with large power handling capacity and a pair of R/C switches for each motor. The relays will weigh more, take up more room, give much less control, be more vulnerable to mechanical failure, and cost very close to as much as an ESC of comparable capacity.
A: [Mark J.] Video: How a Starter Motor Works. It would have been faster for you to look this up yourself.
Q: (2)which kind of moter will use to work(move) on car like suv and how many moter will use?
A: The electric Chevrolet Volt is propelled by a single electric motor with a peak output of 149 hp (111 kW) delivering 273 lb-ft (370 N-m) of torque. That's equivalent to about 33 powerful AmpFlow A28-400 robot motors.
Base Curb Weight of the Volt is 3786 lbs. A similar power to weight ratio will deliver similar performance for an SUV.
Combat robot questions, please.
A: [Mark J.] The more important question is 'Should I use this motor for my 30kg robot?' I'm not familiar with that specific (wiper?) motor, but if you plan to use it for the robot drive system you should reconsider!
If you plan some other use for the motor and require only on/off control, a simple automotive relay will suffice. For more information on using relays and solenoids in combat robots, see our solenoid guide.
A: [Mark J.] You can, but window gear motors (like the AME-210) are relatively heavy, bulky, and very slow (about 100 RPM). Even with large 4" wheels, the speed of your 'bot will be barely more than 1 MPH.
Something like the PD27M gear motor weighs half as much, would give about three times the speed, and still has more than enough torque to maximize sumo pushing power.
A: [Mark J.] Note: it's heavyweight 'Tauro Maximus' that uses the Scorpion brushless motors for propulsion, not middleweight 'Tauro'.
Few robots use brushless motors for propulsion. I consider brushless propulsion to be experimental at his time -- you'd better be ready for many trials and many failures. Amongst the challenges:
My advice is to leave the expense and frustration of brushless drive motor experimentation to other builders. Use the proven AmpFlow motors for propulsion.
A: [Mark J.] My suggestion is to contact the vendor, explain that you got a bad motor, and request a replacement.
The Mabuchi RS series motors are not built with easy disassembly in mind. The motor endbell is secured by metal tabs bent into recesses to clamp it in place. It is difficult to access these tabs well enough to bend them out without damaging the endbell or breaking off the tabs. The best approach is to grind away the base of the tabs to remove them. On reassembly the endbell may be secured with small screws inserted thru holes drilled in the metal can. You're better off replacing the motor.
A: [Mark J.] The AmpFlow E30-400 motors use a high grade ferrite material that is resistant to demagnetization under the conditions you describe. I doubt that you will notice any significant power reduction as a result of disassembly and reassembly of the magnet array. Put the motor back together and give it a test.
The most common cause of ferrite magnet problems in combat robots is extreme heat generated by stalling or otherwise overloading the motors. Again, the AmpFlow ferrite magnets are high grade material designed to perform well in combat and resist demagnetization.
A: [Mark J.] Many, many different motors are used in robot combat. I do know that the High Performance Ampflow motors are very popular in heavy Indian robots for both drive and weapons -- particularly the 'Economy' E30-150 and E30-400 models. They are available as bare motors or with attached gearboxes ready to mount wheels.
Selection of specific motors and gear reduction ratios depends on many design factors that are extensively discussed in this archives. You can also read down this page to find several questions from Indian robot builders about specific motors they are using.
The system seems straight forward,although I cannot find information on the Invacare products. My drive are right angle gear-motors,rated at 24v with a max of 18amps. My application a German Jagdpanther 1/6th scale tank. My previous system was 2-pole wiper motor drives using the same Saber-tooth control.
The difference now being of course 4 wires instead of two. I cannot see voltage feedback occurring,but would like to be sure. Any help would be excellent. I do have the original schematic from my first drive system,would you care to examine that doc in a PDF FORMAT? I cannot see these motors operating differently? Please advise.
Thanks Pal, Blake [Memphis, Tennessee]
A: [Mark J.] I've never used the Invacare motors, Blake -- but I've used a good number of other wheelchair motors.
The first thing I can tell you is that any wheelchair motor can consume a great deal more than 18 amps. The NPC-B82 wheelchair motor pulls well in excess of 100 amps at stall, and the more powerful NPC-T74 wheelchair motor can pull more than twice that amperage. The 18 amp figure may refer to some anticipated consumption at cruise speed, but it certainly isn't a 'max'.
Your Sabertooth controller has overcurrent protection, so you should still be fine using it with the wheelchair motors, even under high amperage conditions.
Four wires, you say? There are several different Invacare wheelchair electric motors, many of which have an 'electric brake'. The two larger wires (likely red and blue) are the motor leads, and the two smaller wires are for the brake. You'll need to disassemble the brake housing at the end of the motor and remove the brake pad for robotic operation. Here's a video of a builder disassembling the brake. Once the brake is removed, you may ignore the brake wires and connect the motor leads to the controller just as you did with the wiper motors.
Write back if you're stumped and we can puzzle this out.
A: [Mark J.] DO NOT attempt to reverse your spinner weapon without first bringing it to a stop!!! Max current figures for your motor all assume that it starts at zero speed, not spinning with high momentum opposite the direction you will power it. A full-power reversal from full speed would be catastrophic, and damage to the battery, motor, or contactor is almost a certainty. Don't do it!
Shorting the motor leads on a spinning weapon is not particularly abusive -- considering that robot combat in general is abusive to all components. Shorting the leads is known as 'dynamic breaking' and essentially converts the motor into a generator, bringing the weapon to a stop more rapidly than allowing it to coast to a stop. The breaking process generates a fair amount of heat, but otherwise should cause no damage.
Also, I wanted to ask If any additional parts (gear, pinion, screws etc.) are required to couple Banebot P60 gearbox with RS-775? [Maharashtra, India]
A: [Mark J.] Last question first. The BaneBots gearbox comes with a mounting kit which contains the correct pinion gear, motor mounting screws, and lock washers. All you need is the P60 Gearbox: RS-775 Mount, 16:1, the RS-775 motor, and a supply of lithium grease to pack the gearbox -- available at your local auto parts store. The gearboxes are shipped 'dry' and must be packed with grease before use.
The warning about high torque for the P60 gearbox only applies to the high-ratio boxes (64:1, 256:1) when used for lifter weapons. The boxes are rated for 35 foot-pounds of torque at the output shaft, and a 16:1 gearbox attached to a RS-775 motor at 18 volts can generate no more than 12 foot-pounds of torque even if stalled. Plenty of reserve, and entirely suitable for your purpose.
I would recommend the BaneBots gearmotor over the Magnum 775 for your application. Four of the Magnums weigh a full two pounds more than four of the BaneBots, and you won't need the Magnum's heavy steel gears for a non-pushy 'bot. Both the BaneBots and the Magnum are more reliable and much easier to mount and implement in a 30 kilo robot than hacked drill motors. Make it easy on yourself and use the BaneBots.
A: [Mark J.] The E30-400 motor (diagram) does not have the large mounting bolt holes found on some other AmpFlow motors. You haven't told me anything specific about your design, but I can offer some general suggestions.
The 'Ask Aaron' project was important to Aaron, and I have decided to continue the site in his memory. Thank you for the many kind messages of sympathy and support that have found their way to me.
- Mark Joerger, Team Run Amok
A: You're confusing 'torque' and 'power' -- they are not the same thing! As discussed previously, power is the product of torque and speed:
Example: if that 100 Nm 'wrench motor' you mention spins at 400 RPM, and the 40 Nm 'drill motor' spins at 1000 RPM, they both have the same power [100 × 400 = 40 × 1000].
Since you haven't given me information on the RPM of either motor I cannot make a comparison of the relative power of the two gearmotors. More importantly, without RPM and wheel diameter information, I cannot evaluate the speed performance of a robot powered by the motors. A very powerful motor will do you little good if it spins too slowly to give the robot decent speed.
I suggest you evaluate the performance of your candidate motors with the Tentacle Drivetrain Calculator and discover for yourself if either is appropriate for your purpose.
Q: Ok. In the website it is mentioned that cordless drill 0-950rpm max torque is 40 nm..for wrench max speed 0-2200 rpm and amax torque is 105 nm...wheel diameter is 9cm...
A: Oops -- no! What you've found is a cordless electric impact wrench. The torque given is the maximum pulsed impact torque the tool can deliver. It greatly overstates the true power of the motor and cannot be directly compared to the stall torque rating for the cordless drill. An impact wrench is not suitable for direct conversion to a combat robot drivetrain -- go with the drill motors.
Robot haiku:
A: [Mark J.] The number of speed controllers you will need depends entirely on your design and the type of controller(s) you select. Many common robot designs use one dual-channel controller for the drivetrain and one single-channel controller for the weapon motor (if you have one). There is a great deal of discussion about controller selection in this archive.
Draw up a design, figure out how much current the motors will use, then start shopping for controllers that meet your need. For a simple 15 pound 'bot, a dual-channel controller like the 'Sabertooth 12 RC Dual Motor Speed Controller' might be a good place to start your search.
A: [Mark J.] Extreme caution!
DISCLAIMER - connecting a single 4-brush PMDC motor (like the A28-150) to the outputs from two non-synchronized motor controller channels is not good practice. I have seen it done successfully, but I do not recommend it. You're spending a good amount of money on this 'bot, and I really don't recommend that you risk the reliability of the machine with a highly questionable speed controller bodge. Destructive failure of the controller is entirely possible. I strongly suggest that you simply obtain properly rated controllers.
That said, I'm willing to tell you how to do it -- proceed at your own risk.
The Ampflow A28-150 motor has four motor brushes rather than the usual two, and four corresponding power leads (2 red - 2 black). The two-channel Ragebridge has four motor output connections (MA1, MA2, MB1, MB2). This set-up powers one set of motor brushes with one controller channel and the other set with the second channel. Be very cautious when powering up the motor the first time; apply partial power and shut down quickly if the motor does not respond as expected.
You get no diagram. If you can't do this without a diagram, you shouldn't do it. In fact, I don't recommend that you do it at all.
Q:Sir, I asked you earlier about having bought 2 Ragebridges to control my two Ampflow A28-150 motors. Now after connecting 1 controller per motor each, how to mix the transmitter set up ?? I am not getting how to provide IN1 and IN2 inputs to receiver and how to control them. Do we have to program the transmitter?
A: The IN1 and IN2 on each controller MUST get exactly the same input signal from the receiver. The simple way to do that is to use the R/C splitter cables I mentioned above. A splitter cable plugs into a single receiver port and duplicates the receiver signal to two plugs that attach to IN1 and IN2 on the same controller. There are other ways to do this, but the splitter cable is the cleanest and easiest to implement. They are available from most R/C suppliers.
You haven't told me what transmitter you have. Assuming that it is a dual-stick design that has elevon mixing, you'll probably want that option. See Appendix A in our Radio Function Guide for full info on setting up elevon mixing.
When you get everything wired up and ready to test, be VERY cautious. As I said before, this is a risky setup. If the motors do not respond to throttle input, power down immediately and tell me exactly what's happening. I'd sleep better if you'd taken my advice and switched to different controllers.
Q: Sir, I was the one who asked you about Ragebridge....No worry now I have bought Ampflow motor controller......hurrrreeeyyyy....No problem about current discharging...
A: [Mark J.] Thank you! I'll rest easier -- and you can too.
A: [Mark J.] You've got all the information you need to answer your own question:
Rather than switching to another speed controller, I'd suggest decreasing the wheel diameter. A smaller wheel will give better acceleration, allow full pushing power, and reduce the peak current drain. If your design allows for use of 4" diameter wheels, the acceleration, speed, and current draw numbers all look much better.
A: [Mark J.] Yes. If you pick a gearmotor from the list, the output speed is the speed of the geared output shaft, and any additional gear reduction selected reduces that speed further. If you pick an ungeared motor from the list, the only gear reduction will be the value you input to the gear ratio box.
A: If you've read thru earlier posts in this archive you have noticed that I often recommend various cordless drill motors for combat robot drive trains. You should also have noticed that I do not suggest the use of automotive windshield wiper motors for that purpose.
Wiper motors (example) are both heavy and bulky for their output. Worse, they are geared down too much for easy combat robot use. Even at 18 volts and with 4" wheels, top speed would be barely 2 MPH. Pick another motor, and learn to use the Team Tentacle Torque Calculator to evaluate your motor choices.
Robot haiku:
A: Short answer - no.
Longer explanation - Electronic Speed Controllers function differently than variable resistors and can't be used as voltage dividers.
An ESC works by rapidly (thousands of times a second) switching the current on and off for variable periods of time. At any given instant each ESC is either passing full current/voltage or none. The timing of those on/off cycles will not be synched between the two ESCs, resulting in both erratic partial-speed control and the full 36 volts being 'seen' by each of the
controllers at least some of the time. You'll very quickly fry the capacitors and/or the power transistors.
Robot haiku:
I did a small calculation realizing that Sabertooth gives me more power, since I can use my RS775 at 18V and with 120 amp in stall which provides 2160W and for the case of victor it provides 150 amp in stall while the voltage is limited to 15V, which means 2250 W. however the stall current is ~130 amp (considering that it does not drop at lower voltage), I will provide the motor with 15*130 = 1950 W, which is less than 2160W. It is also note that Sabertooth 2X60 if burns you need to change the whole unit, where as for victor each motor has its own controller and if one burns, you only change one.
One more issue is that in my application I am sure that the motor is abused so I have to make sure my controller lasts long, and there is a high chance the rs775 spins from outside at 18000RPM which should be taken care of. Since they are closed source, I can't check if they have an additional diode for regeneration rather than the internal diodes in MOSFETS.
Thank You [Malaysia]
A: Mark J. here: you never got around to asking a question, but I can guess that you're asking for my input on which of the speed controllers might be the better choice.
The Sabertooth 2X60 ESC does have full regenerative braking which recovers some electrical energy for the batteries. It also has current limiting to assure longevity of both the speed controller and the motors. The Victor controllers do not have regenerative braking or current limiting. They also do not have a battery eliminator circuit to power your radio receiver -- you'd have to add a stand-alone BEC.
I'd use the Sabertooth.
Robot haiku:
Q: Hi, I am the guy who posted about rs-775 and the two motor drivers (Sabertooth 2X60 and Victor 888). I apologize that I did not ask the question properly. I was thinking about the topic for some time that I almost forgot you are not in my brain :)
A: My bad -- my psychic powers are weak.
Q: There are a few issues. Our robot has a chain which can be stopped from outside if something gets between the sprocket and the chain. I have designed the drive system in such a way that the stall torque is equal to more than twice the mass of the robot × G, and I believe it is a safe choice and I like it since it can climb. The previous one could climb up the wall and flip, which was really cool. something like this. I have temperature monitoring on my motors and speed controller and I can see actually what is happening inside the robot.
The current limiting that you mentioned in Sabertooth is promising, but do you think it can take care of few short period stalls? In last competition there was a robot with spikes on it and the spike used to get between the sprocket and chain. We would totally stop from moving. Happily, he did not have enough torque to push us around but it was keeping the motor in stall mode. At that time I was young and crazy, so I was going full throttle and the motor and driver were running okay, without any failure. This time for Mark 2 we have solved some of the potential problems so that spikes won't reach the chain but it is still possible for something to jam in there.
A: Glad you took some steps to keep your opponent from jamming your drivetrain.
The product description of the Sabertooth 2X60 at the Dimension Engineering website says:
"Overcurrent and thermal protection means you'll never have to worry about killing the driver with accidental stalls or by hooking up too big a motor."
A good implementation of current limiting can completely protect the speed controller (and motor) from cooking. I know from experience that you can hook up a current-limited 4QD ESC to a dead short, push the throttle to full forward, and leave it there 'til the battery goes dead without any ill effects on the controller. Some implementations of current limiting are better than others, and I haven't torture tested the Sabertooth, but I suspect that it will be fine for your purpose.
The best possible current limiting device may be a cool head on the robot operator. I'm happy to hear that your 'young and crazy' days are behind you!
Q: One more question, why do we care about maximum power of the motor in a combat robot? In my mind in combat robot you need high torque to push people around, the maximum possible traction, and high speed to damage opponent with your momentum which is directly related to mass and velocity. Since mass is usually constant in combat robot (Unlike gypsy from Pacific Rim) the only way to increase your momentum is your velocity. However you should make sure you are solid enough for the impact. Am I missing something here?
BTW thank you for reminding me that I = V2 ÷ R. I feel really sorry for myself for being this dumb. [Malaysia]
A: Few things can make you feel stupider than designing a robot. So many things to keep in your head, and so many things that you can frack up!
You're welcome for the reminder on Ohm's law. Let me remind you of another equation that can help to answer your question about motor power. Negating units for simplicity, power is the product of torque and speed:
You want torque to 'push people around' and speed for kinetic energy at impact. Since you want both, that equates to power.
Robot haiku:
A: No - several problems:
Robot haiku:
A: No! Never connect two individual ESCs to a single motor.
Each ESC has its own internal 'clock' circuit that regulates the power output pulses. Two ESCs driving the same load would sometimes cooperate and sometimes be completely out of synch. The result would be unpredictable speed control and unpredictable load on each of the motor drivers.
See this previous post for details and exceptions.
Q: Hi Aaron,
A: Not usually. If the ESC has a single clock circuit, the specifications will clearly say that the outputs can be combined to drive a single motor. Example: the description for the Pololu Dual VNH5019 Motor Driver Shield for Arduino says the unit...
"...can deliver a continuous 12 A (30 A peak) per motor, or a continuous 24 A (60 A peak) to a single motor connected to both channels."
If the ESC description does not specifically say that it can combine outputs to drive a single motor, don't try to use it for that purpose. If in doubt, ask the manufacturer.
Robot haiku:
A: Way too little information for me to answer. See
Robot haiku:
See Appendix 'A' of the Run Amok Radio Functions Guide for help in setting up the radio system for proper robot control.
Note: the AmpFlow dual-channel ESC actually has what the manufacturer describes as an "additional third channel for low-current switching of auxiliary device". This is a simple on-off R/C switch that can be used to activate a relay or some other low-current device, but it cannot be used to directly drive a large weapon motor.
Robot haiku:
A: I suggest that you take some time to learn how an ESC works. An ESC does not 'consume' current any more than a light switch 'consumes' current. Both devices simply control the current passing thru them to the device that actually does consume the current -- such as a motor or a light bulb.
The ESC amperage rating is very similar to the rating for a switch -- it is the maximum current that the device can reliably control without electrical failure. A '15 amp switch' can safely switch (on/off) current up to 15 amps, and a '90 amp ESC' can safely control current up to 90 amps. Both devices will handle lighter loads, but they may be heavier and more expensive than devices designed for those lighter current loads. If your motor is consuming 20 amps, the battery current draw will be very close to 20 amps; the current used by the ESC is negligible.
Example current consumption: a 30 KG robot powered by a pair of AmpFlow E30-150 motors operating from a 24 volt battery.
I should point out that the
Team Tentacle Torque / Amp-Hour Calculator
will calculate the anticipated battery capacity required for a given robot drivetrain in combat conditions given the weight, gearing, wheel diameter, voltage, motor type, and number of motors. It can save you a lot of work.
Robot haiku:
A: Several problems with that ESC.
Q: AND ALSO HOW MANY AMPFLOW E30-400 CAN I CONNECT TO THIS SINGLE ESC????
A: That would depend on what use you plan for those motors. If your use loads the motor heavilly and brings it close to its 266 amp stall current, the answer would be 'one'. Note that the Kelly ESC is a 'single channel' controller. All motors controlled by the ESC would operate at the same speed and in the same direction. Control of motors on opposite sides of a skid-steer robot would require either a 'dual channel' ESC or two 'single channel' ESCs.
Q: hey aaron I'm tat esc guy again wat abt this RoboteQ VSX1850 esc for connecting ampflow f30-150 for tat shell weapon and two e30 150 for drive train weight of our bot is 30-35 kgs
A: No -- that RoboteQ is entirely wrong for your purpose.
A: The low-budget Harbor Freight 900 RPM 18v Drills I recommended in your earlier post - now in the
Q: Yea that is what I used. I attached the motors and gear heads directly to the wheels but for some reason it doesn't seem to move at all, should I have both batteries or is one enough, 18 volt battery that came with it? - VA Engineer
A: You're telling me what the 'bot ISN'T doing (moving), but you aren't telling me what it IS doing.
If the motor is spinning slowly or not at all, make sure your battery has a full charge and that the settings on the ESC are correct. A single battery should be plenty to power your twin-motor 'bot for a full match.
Q: Hey Aaron sorry to be bugging you like this, from what we designed I need at least 4 inch wheels instead of 3, when attached to the 18 volt battery, the motors themselves are moving as well as the wheels but there isn't enough torque to actually make the bot move. Thought maybe I need a set of 20 volt motors but before I get them I wanted to get your opinion on the matter, thanks again for all your help. -VA Enginner
A: No need to be sorry -- you've got me curious, and I'm happy to help.
Something is VERY wrong. Even if you had 8 inch wheels on the 'bot, it should still have plenty of torque to accelerate well, although it would pull a lot of current. Several things to try:
Robot haiku:
Q: Hey Aaron, it's VA Engineer again. I have no further questions for you, all the advice you gave me really paid off and worked significantly. My battle bot is finally working and and I'm ready to compete in my first competition with it. Thx again for all your help. I will definitely be back if any new questions pop up, and I'll recommend this site to any future robot battle builders that I meet. If any people new to this site are reading this, this site really helps you out.
A: Glad to hear you tracked down the bugs! Best luck in competition -- be sure to write back and let us know how your first tournament goes.
Robot haiku:
A: We have discussed evaluating drive motors for a specific robot design MANY times here at Ask Aaron -- browse this archive. Additionally, in the top left margin of this page are links to tools and discussions on robot design that apply to your question. Specifically, I'd suggest that you read the
The equivalent BaneBots P60 26:1 Gearbox mated to their RS-550 motor is a reasonable alternative, but the BaneBots version of the RS-550 motor is not as powerful as the Duratrax RS-550 motor provided with the PDX gearmotors.
I encourage you again to learn the use of our robot drivetrain design tools to better evaluate your options. Depending on wheel diameter, arena size, motor selection, and battery voltage you may be better off with another BaneBots gearbox (16:1, 20:1, 64:1).
Robot haiku:
A: The DeWut?! 3 Speed Gearmotor Kit from Equals Zero Designs is an attempt to make the 'new style' DeWalt motors and transmission useable for robotics applications by providing a structure to hold the motor and gearbox together and link them to a standard output shaft. I haven't had 'hands on' with the kit, but the description and photos look very promising and the price is certainly reasonable.
Robot haiku:
A: Mark J. here: tell me, are these Indian robots that use starter motors successful? Are they champions? If not, why do you copy their design?
The typical series wound stator starter motor - as previously stated - is inherently unsuited to combat robotics use. These motors are designed to produce high torque at low RPM for short periods of time. Torque falls rapidly with increasing RPM, and small changes to the load can cause large changes to the output speed. Operation under high load for more than a handfull of seconds can cause heat damage to the motors, and allowing them to run at high RPM can cause mechanical damage. These are functions of the basic design of the motor, not something that modification can correct.
Accurate specifications (stall torque, stall amperage, RPM per volt...) for starter motors are difficult/impossible to obtain, which makes weapon design pure guesswork. The Team Run Amok Excel Spinner Spreadsheet cannot be used to model weapon performance with series-wound motors because their torque curve is much different than the curve for permanent-magnet motors for which it was designed. It will take a great deal of experimentation (and probably several fried motors) to get the design right.
A few manufacturers use a permanent-magnet design for their starter motors. These motors would be better suited to robotics applications, but are likely still designed for high-torque, low-RPM, short run-time operation. As I have not seen examples of successful starter-motor powered weapons, I cannot recommend their use.
Robot haiku:
A: Black & Decker offers many 18 volt drill models with very different speed and torque specs. All of their drills are of good quality and power, but you'd have to research the speed and torque of the specific model to determine a proper wheel size.
There are large differences in the power 18 volt drills provide. The Harbor Freight drills are good for about 1/2 horsepower each, while overvolted DeWalts pump out 1.25 horsepower. Other brands fall somewhere in this range, but specifications are hard to obtain.
Robot haiku:
A: See
As a starting point in your calculations, consider a pair of BattleKit drive modules mated to AmpFlow E30-150 motors and two Victor 888 speed controllers. That's just about $600.
Q: which motor &speed controller team.toad use?
A: Team Toad has several middleweight robots.
Robot haiku:
A: See
Robot haiku:
I'm building a robot with 4 wheels, each of which are driven by an individual motor per wheel... These motors will be driven by two modified rc car ESC one for forwards/reverse, the other for steering...
My question is, would a gyro improve the handling of this setup (assuming wheel size and spacing has already been optimized for stability)? what I'm hoping a gyro would do is tighten up the robots movements so that it moves with a greater degree of precision, and in turn give the user more control.
If so... am I correct in thinking that I would only install this between the RX and the ESC for the steering?
or would I need one gyro per ESC?
and finally, is there a specific type of gyro I want to be using for this purpose?
I've been searching for days through hoards of gyro types, most of which seem to be for planes and helicopters, and for each of those formats there seems to be units for very specific purposes... yet not much info on how they fundamentally work, and there's not really anything that I could find to do with robotics...
Thanks for your time
A: First, I need to talk you out of your ESC setup. It's generally a really awful idea to attempt to control a motor with two ESCs -- particularly if those two ESCs are providing different power levels and polarities! An ESC controls power by switching it on and off very rapidly. The relative time spent 'on' determines the power supplied to the motor(s). With your setup, if you are moving forward and attempt to turn, the motors on opposite sides of the robot will get conflicting power pulses from the ESCs:
Your questions about robot gyros are answered in the Team Run Amok Beginners Guide to Combat Robot Gyros. I'd suggest you test the robot without a gyro and consider adding one only if you have trouble maintaining a consistent turn rate or have issues with the robot 'spinning out'. Note the warnings in the gyro guide concerning invertible robots.
Robot haiku:
A: Mark J. here: a typical starter motor is of an essentially different design than the AmpFlow motors.
Many small robots use brushless 'outrunner' motors for their spinner weapons, but an outrunner motor powerful enough to spin a weapon for a large robot would be expensive, difficult to find, and untested. All brushless motors require a special brushless motor controller to provide correctly timed and directed current pulses to the field windings of the motor. A diagram depicting the electrical layout of the brushless controller/motor can be found here.
Robot haiku:
A: A pair of the 'new style' DeWalt 18 volt motors running at 24 volts would typically be used in a robot no heavier than 120 pounds, but will give fair performance for an 80 kilo 'bot in a small arena. I suggest locking the transmission in 'low' and fitting wheels about 4" in diameter. Peak current consumption will be close to 50 amps per motor at full push.
Many builders prefer the 'old style' 18 volt DeWalt motor and 2-speed transmission, viewing it as stronger and easier to adapt to robot applications. With either motor you will need to develop your own output shaft and bearing support solution. Team Delta offers 'old style' DeWalt motor mounts, bearing blocks, and output shafts that you may find useful -- but I'm not certain that they are still in stock.
There aren't many proven motor options available for your weight class and budget. If you're willing to consider 4-wheel drive, four of the inexpensive Harbor Freight 900 RPM 18v Drills may be disassembed to provide motors and gearboxes suitable for robot applications. Many budget robots have been powered by these motors with good result -- a web search for "harbor freight drill hack robot" will take you to info on using these motors in a combat robot. Four would give power comparable to two DeWalts.
Robot haiku:
Q: After reading the product manual for the 18 volt Harbor Freight cordless drill the max torque shown in specs is 7ft-lb... and 84ft lbs on the link you provided. Can you please sort it out?
A: One of the things you pay for when buying premium components is correct specifications. Cheap gear typically has either no specs or unreliable specs. The 84 ft-lb torque number on the webpage is a typo that should read 84 in-lbs, which correctly converts to the 7 ft-lbs figure given in the manual.
Robot haiku:
Now for the NPC motors they have give power from dynamometer result and the power of NPC 02446 is 0.7 hp . So this power is even good than F series ampflow in less price .
So what will you suggest AmpFlow or NPC ?
A: Mark J. here: I don't think the RioBotz tutorial got this bit quite right. In real-life conditions, the NPC-02446 is in no way comparable in power to the AmpFlow F30-150 -- or even the E30-150.
Here's the deal: in actual robotic application, the additional resistance of the battery and speed controller can be important when calculating the maximum (stall) torque of high-output motors. The lower the armature resistance of the motor, the larger the effect will be. However, a permanent magnet DC motor generates maximum horsepower at 50% of maximum RPM -- at a point where the counter-electromotive force generated by the rotating armature adds a considerable effective resistance. The impact of battery/controller resistance on horsepower is much smaller than the values cited by RioBotz.
Second consideration: I suspect that those NPC numbers from the dynomometer use a really huge low-resistance battery and no controller at all. You must be VERY careful when comparing motor power measured under apparently different conditions.
A simple method to compare the approximate output power of two brushed motors is to use the following formula on each motor:
The formula makes many assumptions, including equal efficiency of the motors, but it gives an estimate of the power output that is independent of variables such as battery and controller resistance. Using this formula, the numbers come out as follows:
Robot haiku:
A: Mark J. here: brushless motors rely on special motor controllers to perform the function of the commutator in a brushed motor. The controller must be able to determine the relative positions of the rotating (rotor) and non-rotating (stator) parts of the motor to correctly direct the electrical current to the stator coils. There are two ways to do that:
Most all hobby brushless motors are sensorless. Sensored motors will be clearly identified in their description, and will have a sensor cable in addition to the three power leads. A sensored controller may be used only with a sensored motor. A sensorless controller may be used with either sensorless or sensored motors, but the advantages of a sensored motor will be lost.
Sensored or sensorless motors with proper specs will both work well for a spinner weapon. A sensored motor may have a little quicker spin-up, but a suitable sensored motor and controller will be both more difficult to find and more expensive.
Robot haiku:
Q: Hey aaron...
A: Many high performance brushless model aircraft motors can meet or exceed a 4 horsepower per kilogram power to mass ratio. For example, the AXI 5345/16 weighs 995 grams and has a peak output of about 4.7 horsepower -- but only for very short periods of time.
The problem is that model aircraft motors are designed to function with a cooling airflow from the propeller and to operate at fairly constant high RPM. If allowed to drop down below 10,000 RPM at full throttle for more than a couple seconds, rapid heat build-up will destroy the AXI 5345/16. That drawback makes most brushless motors unsuitable for robot drivetrains, but useable **with caution** in light robot spinner weapons.
Robot haiku:
A: Depending on the gear ratio and wheel diameter of your robot, you may be able to hold current consumption of the E30-150 motors close to 50 amps, even when the robot is pushing hard against an immoveable object. For example, 20:1 gear reduction with 6" diameter wheels will deliver enough torque to the wheels to break traction and spin free at about 42 amps current. The bad news is that with that much torque, applying full throttle from a standstill would likely spin the wheels and send you off in some unpredictable direction -- a serious problem with all-or-nothing relay control.
Next problem: if you are moving forward and try to reverse direction without first coasting to a stop, the switching amperage the relay must meet can rise sharply, and can stay high until the robot has reversed and is moving away. Pulling 100 plus amps thru a 50 amp relay even momentarily can weld the relay contacts closed and make it impossible to turn the motors off. Very dangerous, and disastrous in competition. I'd recommend against it.
Robot Marketplace offers a selection of high-amperage power solenoids with contact ratings up to 600 amp inrush. I suggest you read thru the
Robot haiku:
A: Those are very common questions from Indian robot builders. I've answered those questions SO MANY TIMES that I decided to edit together some of the previous Q&A into a special 'Frequently Asked Questions' page covering just solenoid/relay/contactor control of robot motors. You'll find all your answers here:
Robot haiku:
Q: Hey, i have an outrunner brushless motor that is acting funky. When I try to make it turn and it sort of shakes back and forth. It dosn't even come close to making a full rotation. Do you have any idea as to what is going on? [Seattle, WA]
A: First, check and re-seat the three connectors from the motor to the controller.
Next, check continuity on the three motor leads with an ohmmeter. If one pairing of leads is an open circuit, the motor is fried.
If all motor lead pairings check out, the motor controller is toast.
Robot haiku:
Q: O.K, so I have figured out that my motor is fried, is there anything I can do about it besides buy a new motor?
A: Let me guess... Hobby King motor? It's not worth your time trying to fix it. Spend ten bucks on a new one, and buy a couple spares while you're there.
Take another look at your weapon design to see if you can figure out what toasted the first motor.
Robot haiku:
A: Mark J. here: the Open Source Motor Controller (OSMC) project is supported by Robot Power Products. You can find the OSMC circuit diagram, board layout, assembly documentation, and other support info about half-way down the Robot Power downloads page. Read thru the assembly documentation to see what you're getting yourself into.
You'll also need the circuit and board layout for the MOB Interface Board that allows an R/C receiver to 'talk' to the OSMC. Those files are down near the bottom of the same download page.
The Victor ESCs are not open source, and I have no info to share about the Victor 885 circuitry or board layout.
Back in the BattleBots era, a number of teams attempted to build their own controllers. The task is difficult and the results were generally disappointing. There is an active OSMC discussion group that can offer support and answers during your build, but about half-way into this project you're gonna start wishing you'd bought the Victors. Best luck.
Robot haiku:
A: Money is always tight, but you REALLY don't want to go cheap on your ESC. If the controller packs it in, your robot is toast. The Sabertooth is a reliable product at a good price -- there is nothing less expensive that I could recommend in the same capacity range. See
If your funding won't allow you to build a quality 30-pound robot, consider dropping down a weight class or two.
Robot haiku:
Q: Hey Aaron, we decided to go with the sabertooth speed controller as you suggested, luckily we're getting sponsored and can now afford it. Another question that we have is that if we will need another speed controller for the weapon? Its gonna be a spinbot and we are trying to not go over 400 rpm, we understand that the gear ratio will need to be 3 to 1 to get a speed in that area. Also we were wondering what kind of battery will we need for the receiver?
A: Last question first -- you will not need a receiver battery. The Sabertooth ESC you are getting has a Battery Eliminator Circuit (BEC) that reduces the voltage from the main battery pack to about 5 volts and feeds that to the receiver thru the receiver cables. Slick, huh?
You have told me very little about your spinner weapon. I cannot specifically comment on it without knowing much more about its design and components:
You will need some method of controlling your weapon motor. If your weapon spins in one direction only and is powered by a brushed motor it is possible to use a relay or solenoid to start and stop the motor -- but you will also need an R/C switch interface. There are previous posts on this topic in the
Robot haiku:
I just wanted to know what are 'victors' used in all combat robots? What is it used for? Can you please help me out because I have no friend from electronics branch who can help me out with it.
A: There seems to be a great deal of combat robot activity in India. Team Run Amok was invited to a tournament in India earlier this year, but my classwork didn't allow us to attend. Sorry to have missed it!
A 'Victor' is a specific brand of motor speed controller made by IFI Robotics. All combat robots have motor controllers, but they are not all Victors. An electronic motor controller is wired between the main battery pack and a drive or weapon motor. The controller interprets the signal from the radio receiver and provides proportional power to the motor for speed and direction control.
There are several models of Victor controller that differ in current capacity, voltage range, and features. Victors are popular in moderate to large sized combat robots due to their simple set-up and relative low cost, but there are several other manufacturers that supply controllers with more advanced features.
There are many posts in this archive that discuss motor controller selection -- you may wish to read thru them.
Robot haiku:
A: Yes, I've seen video of wild robot fights from India: small arena, thick power cables trailing from the robots, and guys actually standing in the arena keeping the cables from tangling. Terrifying!!!
You may or may not need speed control for your robot, but you do need to be able to reverse the motor direction for your drive motors. If you don't use a speed controller you'll need two high-power solenoids for each of the drive motors -- and two more for the weapon if it needs to be reverseable. Each relay will connect to the R/C receiver thru a small R/C switch interface to interpret the receiver output and control the solenoid activation. There are several recent posts in both this archive and the
The drawback to this type of system -- beside the lack of speed control -- is is that all those high-power solenoids and interfaces are both bulky and expensive. You'll likely spend as much for this system as you would for real speed controllers. Go read thru the archives for more detailed discussion.
Robot haiku:
A: Motor selection depends on more than the weight of the robot and the wheel diameter. Factors to consider include:
As a starting point, a pair of DeWalt 18 volt drill motors with the gearboxes locked on 'low' would be someplace close to correct for your weight and wheel diameter.
A: The amperage consumption of an electric motor varies with the amount of load placed upon it. If your motor is heavilly loaded to its peak 600 watt capacity, it would consume 75,000 mAh of current in 3 hours. If simply free-spinning, the same motor might consume as little as 3,000 mAh in three hours.
A: Cut two -- scroll down to the next question for a full answer.
A: Both of your tinyESCs and the Turnigy Plush 12 ESC have Battery Eliminator Circuits (BECs), and you do NOT want more than one BEC attempting to power your radio receiver. Pull the red pin connectors out of the R/C plugs for BOTH of the tinyESCs (leave the brown and yellow pins intact) and allow the higher capacity BEC in the Plush 12 to power the receiver.
See the generic wiring diagram in
A: This is a job for the Tentacle Drivetrain Calculator!
Fill in the motor values as follows:
Next, set the 'Weight Class' to 12 lb and 'Motors per Side' to 2. Leave everything else alone for now.
With the three inch default 'Wheel Diameter', the performance section of the shows a top speed of about 3.7 MPH with enough torque to break traction and spin the wheels at about half the stall amperage. That's pretty slow.
Click on the 'Acceleration Calculator' and you will see that the 'bot will accelerate to top speed in less than 4 feet. That's about right for a small (8 foot) arena. For a larger arena, the wheel diameter could be increased to 4 inches for improved top speed (about 5 MPH) in a larger arena.
Overall, the performance of the robot would be barely adequate.
Q: Is it possible to swap the motors for a hacked Battlebot toy antweight? In another antweight I built I used some Tamiya Hyper Dash-2 motors. They're the same size as the ones in the Battlebots toy. Or do you think this could cause damage either to the integrated ESC/RX?
A: The BattleBots toy Controller/RX should be fine, but the Hyper Dash-2 motors won't do well at 6 volts. The HD-2 is a 3.0 volt motor that shouldn't be run over 4.8 volts. If you want more power, I'd suggest overvolting the stock motors. I've seen BattleBots Custom Series toys run at over 9.6 volts without trouble -- but are you sure you really need/want more speed in a small insect arena? It's gonna be a real handfull to control!
Q: only 4.8v for the Hyper Dash 2? I figured since the robot marketplace sells them as a drop in for the Tamiya Tracked Chassis, and says that can be ran safely up to 7.2V...I was planning on using that chassis and putting a dual gearbox with the Hyper Dash 2 motors installed...
Mark J. here: you can run the stock Tamiya motors up to 7.2 volts. The high-performance HD-2 is wound with larger wire and less of it than the stock motor. It spins as fast at 3 volts as the stock motor does at 7.2 volts, and it draws a lot more current. Run the HD-2 at 7.2 volts and you'll likely melt the brush holders, fry the commutator, or just burn the windings. See my comments on overvolting high-performance motors in the post below.
A: There are many posts about overvolting in this archive. Also, take a look at
Overvolting will shorten the life of the motors. The impact on motor life will depend on the design specs of the motor, the load placed on the motor, and the degree of the overvolting. In general:
It's also critically important to countour the motor brushes to match the radius of the commutator by running the RS-550s for 10 to 20 minutes at about 6 volts. It's a good idea to do this for ALL brushed motors, but it's particularly important for motors that will be overvolted. This 'break-in' will maximize the contact area of the brushes and spread the current/heat over a larger area and prevent destructive arcing.
Mark J. here: when you see a motor rated at a nominal 12 volts, that rating assumes a certain usage for the motor and a specific expectation for longevity and reliability.
A: Most all brushless ESCs - including the one you want - are made for aircraft. Since airplanes don't back up, aircraft brushless ESCs don't reverse. Reversing brushless ESCs are very rare, and they will very clearly say that they are capable of reversing. Unless you want to hack into the firmware to try to activate reversing, look for brushless ESCs made for R/C cars and boats.
A: I don't generally suggest a specific drive motor, but I can point you in the right direction. Final selection of the motor and gear reduction will depend on factors such as the size of the arena and exactly what you mean by 'move fast'.
The
Team Tentacle Torque/Amp-Hour Calculator
suggests that a pair of AmpFlow E30-150 motors geared down about 16:1 would provide fair performance in a 4-wheel drive 150 pound robot with 8" wheels. A top speed of about 8 MPH is reached in 16 feet, and there is sufficient torque to maximize pushing capability or carry an opponent without difficulty. An upgrade to the Ampflow A28-150 motors at a 12:1 gear reduction would provide greater speed and acceleration better suited to a larger arena.
If you're interested in a gearmotor, the DeWalt 18v or 24v drill motors are possible choices -- as is the venerable NPC T64 wheelchair motor. I suggest that you use the Tentacle Calculator to experiment with different motors, gearing, and wheel sizes to find an optimal performance and cost balance for your drivetrain.
Six-wheel drive heavyweight 'Sewer Snake' uses an AmpFlow A28-150 motor geared down 75:1 for their lifter weapon. For a 150-pound version, lifter motor selection will depend on the length of your lifter arm and the lift speed you are willing to accept. I suggest starting your calculations using the specifications of the AmpFlow E30-150 motor and adjusting the motor requirements as needed.
Mark J. here: we rarely recommend a specific motor solution because there are so many variables in robot design. Is the builder willing to accept a 'bare' motor and construct their own gear reduction system? Does the builder require a gearmotor with a sturdy output shaft ready to bolt into a chassis and use? Is some intermediate solution acceptable? How much money is in the budget for motors, and what level of performance is expected? If we make a specific recommendation, it's quite likely that the hamburger will be bad.
If you just want to turn a brushed motor on and off, you can use a less expensive R/C switch. An R/C switch will give you single direction on/off control. Again, the switch must be matched to the voltage and current requirements of the motor.
The diagram illustrates how the components of a typical combat robot connect. There is a great deal of discussion about speed controllers in this archive.
I think i have to operate them at 12V because the torque after reduction is huge and may cause wheels to slip.
A: You will certainly have huge torque -- but VERY little speed. At 12 volts with 6" wheels (just a guess) your top speed will be barely over 2 MPH. That's nowhere close to enough speed for a combat robot.
Operation at 18 volts with the 16:1 P60 gearbox and about 3" wheels (depending on arena size) would be much better. Top speed jumps to more than 10 MPH, and acceleration is adequate for a robot with an active weapon system. If your design requires greater acceleration and power to be effective (wedge, rammer...) I would suggest four RS-775/P60 gearmotors and four wheel drive.
Wheel slippage is not an issue -- you need to be able to break traction and spin the wheels freely under hard pushing conditions to avoid stalling the motors and overloading the speed controllers. With two RS-775 16:1 gearmotors at 18 volts and 3" wheels, the motors will each draw about 50 amps before breaking traction (several assumptions included). An equivalent four motor, four wheel drive version would require about 30 amps per motor to break traction.
A: The solenoid you found gives no specifications beyond '100 amp' -- there's no way for me to comment on its suitability to control a motor that can pull more than 120 amps at startup. I can say that the body of the solenoid appears to be made of a material similar to the brittle body of the Team Whyachi solenoid that has been reported to fail under impact.
Any time you use a component that hasn't been tested in combat you are in uncharted territory. I recommend using known and proven components upon which you know you can rely.
Mark J. here: the peak 120 amp consumption of the AmpFlow E30-150 motor assumes that it is starting from a dead stop. If the motor/weapon is spinning in one direction and is being reversed without first bringing it to a stop, the potential peak current can be much greater and can continue for longer period of time.
A: DO NOT connect the outputs of a two channel ESC in parallel unless the user manual specifcally says that it's allowable, and never connect the outputs of two individual speed controllers in parallel. Each ESC has its own internal 'clock' that regulates power output pulses. Without a mechanism to synchronize those clocks, both the load on each ESC and the speed of the motor being controlled are unpredictable. This applies to both brushed and brushless motor controllers.
Seach this archive for "one motor with two" for more detail and for a special case where two ESCs on a single motor can work.
A: Mark J. here: I'm guessing that you want to use the gearmotor as a generator.
The question cannot be answered with the information given. Just as motors of differing design will run at different RPM at a given voltage, different motor designs will produce differing voltages at a given speed when used as generators.
If you know how many volts the gearmotor requires to spin at 3000 RPM as a motor, you can estimate that it will produce 'a little less' than that voltage when spun at that speed as a generator. How much less depends on the efficiency of the generator design.
A: For any practical purpose - no.
Mark J: entirely correct. That helicopter control board appears to be an 'all-in-one' receiver/servo-driver/motor-controller. The two brushed motor controllers are most likely uni-directional (no reverse) with TINY current capacity. Compatibility of the receiver with standard R/C transmitters is an unknown (I'd bet against it) and fail-safe capability is likewise unspecified. Get something with known specifications and compatibility.
A: The E-flight 10-Amp Pro ESC comes 'out of the box' programmed to cut off power when the battery voltage falls to 9 volts. That's the correct value to protect a 3-cell LiPoly battery, but it almost immediately shuts down when hooked up to your 2-cell LiPoly. Consult the ESC manual for instructions on how to re-set the low voltage cut-off to 6 volts. Read the rest of the manual while you're there.
A: Mark J. here: both that weapon ESC and the Sabertooth have battery eliminator circuits (BEC). If you're lucky, the problem is the two BECs 'arguing' about the power needed for the receiver. Fix that by disconnecting the red (power) lead from the weapon ESC receiver plug. Take a good look at the connector plug and you'll spot how to push in a metal 'finger' to release the red socket and pull it out. If you have trouble with that, just cut the red lead wire and tape it back out of the way.
Q: Your idea worked! Thank you guys so much for all your help. I was discouraged after putting all the time and work into it that it didn't drive and activate it's weapon. thanks so much again.
A: Glad to help.
Problem is with forward speed. On first forward movement I have normal forward speed as long as it's held at max throttle. Stop and reverse and have same normal speed. Ditto on max turns, spinning tread is running at normal speed. Go back to forward, and now speed is 1/2 what it was. Reverse and normal stay at original speeds, only forward is reduced.
I've swapped motor leads, swapped motors, tried the various Dip switches in different positions to no avail. Even swapped out from a dry cell to a gel cell battery. Different rx's too. To me it all seems to be in the Sabertooth? Tried using DE's help section but never got a reply from them.
Any help or suggestions would be greatly appreciated. Thanks, Brad
A: Mark J. here: there is a known issue with some specific Spektrum receiver models sending a signal at power-up that can confuse the auto calibration feature of Sabertooth motor controllers. Try turning auto calibration off by setting DIP switch 5 on the Sabertooth 12 R/C to the down position. If this solves the problem, you may be able to turn auto calibration back on and avoid the confusing signal from the receiver in the future by always turning the transmitter on before powering up the robot -- ermmm, tank.
Q: An update to a question I just asked on a Sabertooth loosing 1/2 forward speed after turning.
Dimension Engineering answered back saying to try a recalibration. Switch #6 down to set end points then turn switch up to lock in settings before powering down. Worked fine until I powered back up to check, no joy same problems.
Tried something on my own. Set switches 5 and 6 both down, ran through forward/reverse/turns, then back to forward and reverse. All working as intended. Flipped switch 6 back up to lock in. Shut down system, fired it back up and happy happy joy joy, IT'S WORKING!!!
Shut off and fired back up several times and all is good. Hopefully this might help someone else in the future. - Brad
A: Great result, Brad! If the robot -- ermmm, tank -- is responding well you can certainly just keep DIP switch 5 down and leave well enough alone. Happy tanking.
Note: DIP switch 5 controls the auto calibrate function on the Sabertooth 12 R/C and Sabertooth 5 R/C. Auto calibrate in R/C mode for other Sabertooth motor controllers is typically part of the function of DIP switch 6. Consult the manual for your specific motor driver.
When both switches are closed the diodes are both bypassed and current may flow in either direction to the motor. When a switch is forced 'open' as the gun reaches the travel limit, current flow in one direction is blocked by the diode. When the controller is reversed the current can flow in the other direction to reverse the motor and back it off the limit stop.
Test to find out which switch stops motion in which direction, and install accordingly.
A: Mark J. here: a hobby brushless ESC is essentially three independent Pulse Width Modulators (PWM) controlled by firmware that interprets the signal from the radio receiver. It's possible to re-write the control firmware to make two of those PWMs provide independent controlled current to two brushed motors, and use the third PWM to provide a common ground. Flip the polarity of all three PWMs to give forward/reverse directional control. Note my careful use of the word 'possible'.
The crazy Aussies [no offense
Note that there are control issues with the completed hack. The 'spin in place' rotation speed is slow because full power cannot be provided to both motors if they are turning in different directions [no thwackbots!]. Also, standard transmitter mixing will not work for this hack since the 'reverse' signal comes in on a different channel.
If you want to invest several weeks of your time to:
A: Mark J. here: I can't comment on the switches, but the BotBitz ESCs are Hobby King brushless ESCs that have been reprogrammed with new firmware to work for bi-directional control on a brushed motor. These very inexpensive Chinese ESCs are not known for their high quality or accurate 'real world' power ratings, but the BotBitz units have 'added value' in that they work with brushed motors. More info on BotBitz products.
Note: buying from foreign suppliers can carry unexpected costs. In addition to shipping, you may have to pay US customs import fees. Most credit/debit/check card companies will also add a 'currency conversion fee' to foreign transactions.
A: Mark J. here: it depends on how the motors are wired to the battery. In a typical robot drive system the motors are wired in parallel to the battery. Each of the two motors has access to the full battery voltage -- each will 'see' 8 volts.
If two motors are wired in series to the battery, the voltage share available to each motor is complicated by the electrical behavior of motors under varying load. Under equal load each motor will 'see' 4 volts, but if one motor is placed under a greater load the other motor will receive a greater share of the voltage in proportion to the difference in loading.
A: As you might guess, the brake function slows the attached motor more quickly than allowing it to coast to a stop. Handy for a weapon. Note that this is a single-direction ESC with a greatly inflated amperage rating. Use at your own risk.
Q: Could in theory I use two 'Turnigy 20A Brushed Speed Controller' powered one source for drive motors? Would it be reasonable at all?
A: You could run two (or 3 or 4) from a single battery, but these are 'single direction' ESCs with no reverse. Potentially useful to control a very low budget spinner weapon, but useless for drive motor control.
Q: Of course I wouldn't use those ESC's into a tournament but could in theory I practice with them untill my good ESC comes in the mail?
A: They Have No Reverse! They are made for airplanes -- airplanes don't back up. You could practice driving forward and making wide turns until you ran into something and got stuck, but how would that help? If you're in a hurry buy overnight delivery for your 'good' ESC.
Q: Could you elborate on those control issues?
A: Motor speed controllers are either 'single direction' or 'reverseable'.
A: The short answer is 'yes'. Toy electronics combine a simple radio receiver and a crude motor controller in a single unit. A hobby-grade radio receiver itself cannot directly control an electrical circuit. A variety of different devices can be plugged into the output ports of the receiver depending on the specific control needs of the vehicle being controlled:
There are many posts dealing with selection of ESCs in this archive and the
A: RS series motors are available from many sources. Some may have attached wires, most will not. Wires attach to the two metal tabs sticking out of the end of the motor opposite the output shaft -- it will be obvious when you see the motor.
Q: If so, where is the tab that i put the positive cable, and the tab where i put the negative cable?
A: Your RS motor may have a red dot or a '+' mark at the base of the positive tab, but the motors are reverseable. If the motor doesn't spin in the direction you want you can swap the wires to reverse the spin direction. Note that most RS motors spin faster in one direction than the other.
Can i use either the SyRen 25 or the BB-12-45 as an alternative to the Victor 884?
A: You've got to tell me what weapon blade and reduction ratio you've settled on. The larger blade you've mentioned recently will have more rotational inertia, will cause your RS-550 weapon motor to draw more current for a longer time, and will require a speed controller with greater capacity. Settle on all of your weapon physical details, then you can pick an ESC.
Q: Battery guy again.
okay, i've decided to use the 50cm X 3cm X1cm blade, and i've settled on the 8:1 ratio you've told my about the previous blade.
A: Mark J. here: good choices. With that blade and gearing the weapon motor will still pull a lot of current during a spin-up:
An ESC stressed beyond its continuous current rating starts building potentially damaging heat that must disipate before the ESC can repeat the overload cycle. ESCs with large heat sinks (like the SyRen) or with active cooling (like the Victor) can shed heat faster and recover more quickly. ESC manufacurers do not provide enough data on overcurrent thermal behavior to precisely calculate performance under all conditions. Here are my takes on the three ESCs you are considering:
A: Either motor will provide plenty of power and do nicely for your general purpose. As you point out, the Dustin gearbox will simplify drivetrain design. Mounting options for the two motors are different, and one may be a better fit for your chassis design. You might also consider a design with four DeWalt Powerdrive motors to completely eliminate the need for a drive chain.
A: I don't think you're going to have much luck finding a similar gearbox. My understanding is that the shortage of certain BaneBots gearboxes is temporary. If you don't have time to wait for a P60 for the 3xx motors you can switch to the RS-545 motor and the 16:1 P60 for the 5xx motors - which is in stock. The RS-545 is heavier than the RS-380 (6.2 ounces vs. 2.8 ounces) but will provide similar performance in your hobbyweight and will not overstress your Sabertooth 12 ESC. If you can afford the extra weight, go for it.
A: That depends on what you're planning on doing with the RS-550 and which DeWalt motor you're talking about.
I can tell you that the RS-550 is much less expensive, and that it is not made to survive the abuse a DeWalt motor can handle. You're getting what you pay for.
Q: i'm planning to use the RS550 for my drivetrain of my hobbyweight robot. The robot's size will be: 30 cm X 15 cm X 2 cm. The body will be made from aluminium and my weapon is a horizontal bar. For the Bar's motor i'll be using EFL-25-870 Power 25 Brushless Outrunner Motor and for the Weps ESC i'll be using Phoenix 45 Brushless Speed Controller. And for the Drivetrain's ESC i'll be using a Sabertooth 12. i've just ordered the Sabertooth yesterday.
A: How do you plan on installing a 3.8 cm motor in a robot only 2 cm tall?
A pair of RS-550 motors is overkill in a hobbyweight drivetrain. Geared for reasonable speed an a typically sized arena, they would provide ten times the torque needed to break the wheels free and spin them uselessly. RS-380s might be a better choice, particularly for a 'bot with an active weapon. Keep the gearing and wheel diameter reasonable or you'll overload that Sabertooth 12 -- bogging a motor makes it use more current.
Q: i'm not planning to use the RS-550 as it is, i'm planning to use this PDX16 - 16:1 Gearmotor. Is this still overkill for a hobbyweight?
A: Even more so. The Duratrax 550 motor is more powerful than the BaneBots RS-550 -- about 50% more powerful. Stall current is 148 amps -- rather too much to control with a 12 amp ESC! A pair of these motors will provide more than one horsepower. Massive overkill.
Q: so, what motor do you recommend which is powerful(but not overkill) and not a DeWalt?
A: Control and smooth response are more important than raw power in the drivetrain of a robot with a large active weapon. You need enough torque to achieve the full pushing power available in your weightclass without bogging the motors, but tire-burning acceleration and ramming speed are neither needed or desireable.
I mentioned above that a pair of BaneBots RS-380 motors would be reasonable to move around a hobbyweight spinner. You could mate them to BaneBots P60 gearboxes. Gear ratio for the P60 will depend on the wheel diameter you plan to use and the size of the arena -- around 26:1 with 3" wheels or 20:1 with 2.25" wheels is close.
Q: I've got a question for you about both of the RS-380 and P60 Gearbox. 1.what size of the bolt should i use to connect the RS-380 and the P60? And is the bolt included when i buy the P60 gearbox?
A: P60 gearboxes for the 3xx series motors come with a mounting kit that includes mount hardware and the required motor pinion gear.
A: That designation refers to the number of Lithium Polymer cells recommended in the battery. One LiPo cell provides 3.7 volts, so a 2-3S LiPo recommendation calls for 7.4 to 11.1 volts. I don't recommend overvolting hobby brushless motors.
A: I don't know why you would want to know, but about 9 minutes into this 2008 combat robot webisode of 'Systm', Dave holds up a Barello Ant 150 ESC. Barello ESCs are no longer in production, and Dave ends up spending more like $400 on a 'pit pass' robot without a battery charger.
Q: Hi Aaron, just asking, what ESC that can control multiple channels? i'm asking that question because i'm fascinated with that ESC because Dave Calkins said that this ESC can control multiple (3) channels.
A: ESCs that control multiple channels are not unusual. The Barello Ant 150 could control two forward/reverse drive motors (2 amp continuous) plus a forward-only brushed weapon motor (8 amp continuous).
Many ESCs control two drive channels (Sabertooth, Robo Claw, some Scorpion, some Vantec...) but three channel ESCs are pretty much extinct because of the increased popularity of brushless weapon motors that require their own specialized ESC.
A: Aha! The 'flip channel' is used to fix the throttle response when an invertable robot gets flipped over. An inverted robot with differential steering will still steer correctly when flipped, but the throttle is reversed -- forward is reverse and vice-versa. Plug this lead into a spare receiver channel controlled by a transmitter toggle, and if your robot gets flipped over you can flip the transmitter switch to restore correct throttle direction.
A: Line followers and maze robots are much different than combat robots. The gearmotors you found are converted servos with FAR too little torque (0.9kg-cm = 12.5 oz in) for a hobbyweight. A typical hobbyweight gearmotor would have similar RPM but 20 times that much torque. The motors could work for an antweight, but no way for a hobbyweight.
A: Victor IFI speed controllers were designed for use with a special radio system (no longer available) that provided a stronger than normal control signal from the receiver. For reliable use with non-IFI receivers, a special signal booster cable is typically needed. Like most robot supplies, signal booster cables for Victor IFI speed controllers can be purchased from the Robot Marketplace. The cable is NOT included with the Victor controller.
Q: One more thing....... is the Sabertooth 12 RC Dual Motor Speed Controller good for a horizontal spinner hobbyweight?
A: Proper selection of an ESC depends on more than the weight of the robot and the type of weapon. Motor type, gear reduction, wheel size, voltage, and weight on the driven wheels all go into the calculation. The Sabertooth 12 is probably a pretty close match to your needs, but if you don't do the math you're just guessing. See
A: Mark J. here: a forward/reverse DC motor speed controller with 10 amp capacity is not a common item in consumer products. Add to that an interface that allows R/C receiver output to operate that speed controller and you have a specialty item that you aren't going to find in 'hackable' form.
See the second part of
[10 hours later] Q: I have decide to ditch the B & D motors and get the RS-380 motors where can i find a gear box.
A: Ummm... have you confirmed that the RS-380 motor is different from your B&D motor? If it isn't, why ditch it? BaneBots makes gearboxes for the RS-380 motor.
[15 minutes later]
Q: Whould the BB-12-45 ESC be ideal for 2 RS-380 motors?
A: Did you not read my answer to the ESC selection question you asked earlier today? Have you told me the weight, wheel diameter, or voltage of your robot? The tools to select an ESC are given in my answer to your previous ESC question.
Note: the BaneBots BB-12-45 ESC is a single channel unit -- you would need two to control two motors in a differential steer robot.
A: I'm not sure what you're asking. A cordless drill motor itself has far too little torque to directly drive a robot wheel, but a great many successful combat robots have been built with the motor/gearbox combinations taken from cordless drills. Note that the gear ratio provided by the drill gearbox may dictate a wheel size that is inconvenient. See
A: I gotta say that picking a weapon motor based on what you happen to have lying around may not be the best approach.
550 size R/C car motors come in a wide range of performance capability and are notorious for a lack of reliable specifications. A 55 turn motor like the Terra Claw is designed for low RPM and low power consumption -- not high performance. As a PURE GUESS: stall amperage around 15 amps at 12 volts.
Search this archive for "determining motor stall current" to find the D-cell method of actually measuring stall current.
A: That trick only works in large brushed motors with two pairs of brushes, like the Ampflow S28-400 motors that power 'Sewer Snake'. Search this archive for "two identical ESCs" to find an explanation.
A: I hope you aren't trying to go very fast or do much pushing with a pair of those motors in a middleweight. That 250 watts is input -- output is no more than 1/4 horsepower per motor. That's much less than current middleweights run. It might be adequate to move around a 'bot with a really big weapon, but not for a wedge or rammer.
Anytime you grab a motor that is untried in robot combat you run the risk of uncovering weaknesses. The high stresses of abrupt reversing are not something that a chinese scooter motor is going to be designed to handle. We recommend sticking with products proven in robot combat rather than trying to go cheap -- particularly with key components like motors.
You can use a good-quality epoxy (slow set -- not that 5-minute stuff) to re-bond the magnets to the can. Disassemble the motor, clean all the old adhesive from the can and magnets, and make sure you get them back where they originally sat. Cover the entire back surface of the magnet with epoxy and clamp it firmly but gently in place 'til they set. Ceramic magnets are brittle and can break under pressure, so be careful. It may be easiest to reposition the magnets one at a time.
A: All the Ampflow motors have keyed shafts. The E-150 has a 12mm shaft with a 3mm keyway.
The process of forming a keyway is called broaching and requires a tool called a broach.
A: Page 6 of the Robot Power Wasp manual says:
"The internal voltage regulator of the Wasp is a National Semi LM2940IMP-5.0 regulator. This is a linear type regulator and generates significant heat when the difference between the battery voltage and the 5.0V regulated output voltage is large. If the voltage regulator gets too hot it will shut off to prevent damage. This disables the Wasp since the internal CPU uses the output of the regulator. The Wasp has been tested up to 21V input without overloading the regulator. For battery voltage above 21V it is recommended to use an external 5V supply connected through the RC cable (receiver battery or BEC) and clip the 5V disable loop on the Wasp."
According to that, your 16.8 peak voltage is not a problem. If the regulator overheats it shuts down to avoid damage, so I'd certainly test under full combat conditions to make sure that doesn't happen.
Do NOT try to undercharge the battery! Battery voltage does not have a linear relationship to the percentage of charge. A battery that is 50% discharged still has very close to the full charge voltage. You'd be shorting yourself a lot of charge in order to drop a little voltage.
What could be the issue?
A: First, bring that battery up to full charge. NEVER try to set-up an ESC with anything other than a fully charged battery.
Next, consult the Sabertooth User's Guide to verify the DIP switch positions for your application. In particular, make sure switch 5 is in the 'up' position to activate auto-calibration for your transmitter end points.
Q: Oh, and another thing: I want to put a wedge on my bot, which is basically a box. I was hoping to bend a piece of soft aluminum, bolt a hacked apart kitchen cleaver to the front, and then bolt two reinforcing brackets along the side. What are your thoughts on that idea, if you have enough info, at least. (I really don't know how else to make a wedge with the weight allowance and machining skills my team has.)
A: A kitchen cleaver makes a fine budget wedge. A full-width brace along the bottom of the wedge is better design than side braces. Search the
Q: I am not sure how or if my team can make the brace shown in the diagram, let alone make it so it will match the angle of the wedge. (I can't say what angle the wedge will be, because the bending process consists of placing the soft aluminum (aircraft aluminum snapped during this process) and banging it with a hammer.) I am not saying that I know for sure that we can't do it, but if we can't, is there an easier way to support the wedge?
A: Sure -- how about a block of wood? Go ahead and bend your top bracket to the angle you want, then mark that angle on a block of wood the width of the wedge and whittle away on it 'til it fits. Fasten to the wedge and robot body with screws. Wood is an underappreciated material in robot construction.
A: You can't. Efficiency is the mechanical output power of the motor divided by the electrical power consumed. It varies with motor speed. You can calculate the power consumption, but the 'usual given specs' do not include mechanical output numbers.
A: Mark J. here: any ESC can fail if called on to perform above capacity, and some builders are quick to blame the ESC rather than their own design skills. Many combat robots successfully use ESCs from the manufacturer you mention. I know of no inherent or recurring problems. Like any ESC: keep them within their amperage rating, provide ventillation, and protect them from impact shock and debris.
A: The DeWalt 18 volt drill motors are high performance motors that deliver a lot of power for their size and weight. As with all high performance motors, overvolting should be done with moderation. I cannot recommend running these motors much over 24 volts.
A: Too little information. See
A: Many manufacturers produce '385' motors with highly variable specifications. The specs for the BaneBots RS-385 in the TTTC (reloaded) came from the BaneBots spec sheet. I don't have a current source for that specific motor, but the Speed 400 motors (from multiple manufacturers) are the same size, bolt up to the same gearboxes, and offer performance equal to or greater than the BaneBots RS-385. I think you might find one that meets your needs.
A: Mark J. here: no, it would not. I would guess that a set of Axi 'gold' magnets might cost more than the whole Turnigy motor. That approximation of 'similar performance' is made by the people who sell the Turnigy, so I wouldn't count on that either. You get what you pay for.
A: The obvious third option is to dial down the motor power! Four RS-550 motors in a 15 pound bot is HUGE overkill. Applying as little as 4% of the motor torque will spin the wheels uselessly. Undriveable! Four RS-550's could power a 120 pound middleweight quite nicely.
The Tentacle Torque Calculator says you'll never draw more than 3.2 amps per motor (@ 12 volts), but I wouldn't trust that number. You're understressing the RS-550s so far that your peak amp draw will likely happen when applying full throttle before the motors overcome drivetrain inertia - a condition the calculator does not consider.
Seriously, drop back to RS-385 or RS-540 motors and consider less gear reduction (20:1 or 16:1) for better speed. With that set-up the Scorpion XXL speed controller would be a good choice.
Q: [received before above was posted] I'd like to add to my earlier question about ESC's. I'm trying to work out my drive train for a 15lb Wedge/rambot, and am getting some weird values when I use the Team Tentacle Torque Calculator.
My basic predicament: I'm getting very low current values at the point of wheel-slip. I would like an accurate value so I can determine if using smaller ESCs is possible.
Please look at the linked images to see what I am talking about. I'm getting very weird amperage and torque values. They are particularly important to me because I'm trying to choose an ESC, as well as make sure my motors have enough torque to spin, and get a battery with enough Amp-hours that can handle the four motors. Please tell me if I'm using this tool incorrectly.
Thank you for helping me.
A: First, I suggest using the Team Run Amok 'reloaded' version of the Tentacle Torque Calculator. It includes data for the ungeared RS-series motors and has a less confusing (IMHO) information layout.
From what I can make out from your screen grabs, you are getting essentially correct results -- the motors are HUGE overkill. See my comments above.
A: The EV Warrior motors were wildly popular in robot combat ten years ago. They put out a fair chunk of horsepower (1.55 @ 24 volts), didn't weigh much (3.24 pounds), and put up with a good amount of abuse. However, their main attraction was that they were available from surplus dealers for VERY little money. Eventually the surplus warehouses ran dry, although you can still find a few of the motors from time-to-time on EBay.
There are certainly better motors available. The Ampflow A28-150 Motor is of comparable size and weight, is more efficient, and produces about twice the horsepower. The Ampflow is about $300, while the EV Warriors were once available for as little as $15.
The EV Warriors could still work for a low-buget build. If the specs work for your design, use them -- if you can find a pair.
Motor stall torque can be directly measured with a simple aparatus. Clamp the motor securely to your workbench and clamp a lever arm securely to the motor shaft. Place a scale under the end of the lever arm (or attach a 'fish scale' to the end of the arm), and switch on motor power. Quickly get a reading from the scale and switch the motor power back off. Here's your formula:
A: A quick check with the Tentacle Torque Calculator shows that a pair of Magnum 775 gearmotors are generally suitable for a lightweight wedge. Actual performance will depend on the voltage at which the motors are run, and wheel diameter.
The classic motors for a two-wheeled lightweight wedge are the DeWalt drill motors, but a proper DeWalt setup is considerably more expensive than the Magnum.
A: Mark J. here: the electric motor braking described in section 7.2 of the RioBotz Combat Tutorial is correctly known as 'dynamic braking'. In dynamic braking the motor brushes are shorted, which turns the motor into a generator. The kinetic energy of the spinning motor is converted into electric energy and is dissipated as heat. A more sophisticated braking system known as regenerative braking uses the energy produced to restore some charge back to the battery.
Of course, if you need to stop really quick you can always throw reverse power to the motors and spin 'em backward. It's hard on the motors and controllers, but combat is hard on all the components.
A: Mark J. here: too many variables for me to put in a good estimate. How effective is the heat sink? To what is the heat sink attached? How hot is the motor when it stalls? Have the brushes been properly broken in? As a pure guess, I'd be fairly confident that you'd have more than a second under best conditions at 12 volts -- but that is a guess.
A: Mark J. here: the problem with brushless motors in combat robots is not stopping or running at reduced speed at small torque loading; brushless motors throttle down perfectly well. The problem comes when the motor must produce a lot of torque at low RPM when the robot is pushing hard against another robot. Under those circumstances the motor consumes a great amount of amperage and creates a large amount of internal heat. Hobby brushless motors do not tollerate that type of use -- they melt.
[This is an exerpt from a longer question. The full question and answer are in the
A: Although Kitbots does not give full specs, I believe that the motors themselves are very similar but that the gearboxes are different. The B16 is lighter, and note that Kitbots offers different mounting plates for the B16 and '1000 RPM' motors.
A: The FF-180 motors are used in a wide range of products (electric toothbrushes, cordless shavers...) so you can count on finding them somewhere. Try a web search.
A: I'm gonna say no. The motor's output is well less than one watt, and we recommend a minimum 4 watts per pound of robot. Move up to at least the FF-050 motor for a beetle.
Q: Are any FF-050 motors still around?
A: Bare FF-050 motors are available direct from BaneBots and many other sources [August, 2016].
A: I'm not current on brushless ESCs. I think this is a perfect question for the RFL forum.
Q: Does the Skyartec 18A Brushless ESC work legtally with the Spektrum AR6110 DSM2 Microlite?
A: See above.
A: I'll assume that you know that either of those motor choices would be massive overkill -- that's why you refer to your design as a 'super wedge'. The gearing and wheel diameter selection will hinge on the arena size in which you expect to compete.
The HTI motor does not bolt right up to the current BaneBots P60 gearbox for the RS-550, and there is no 26:1 ratio currently offered for the BaneBots P60 or P80 gearboxes that mount the RS-775. You'd have to modify the gearbox to mount your HTI motor to the RS-550 version of the P60 gearbox.
The DeWalt motors are both more powerful and reliable than the HTIs, but they are also considerably more expensive. Given that you are already overpowered, you will need to make your own decision as to which path to take. Either way, the 'bot is going to be a real handful to drive. Consider adding a peizo gyro to the radio system if it's uncontrollable.
I tried using the HTI motors with a 20:1 reduction, 3" wheels, and 16.8v driving it on a ~30 lb robot (1 per side), but they smoked. I opened one up and discovered that the brush arms are alarmingly thin copper leaf springs. Do you think cobalts would perform better, or that I should try a motor from a name brand drill?
A: I'm really surprised that you cooked the HTI motors at that voltage and gearing. I would have thought you could push a brick wall at full thottle 'til the battery went flat and not stress that drivetrain. The motors were only drawing about 15 amps to break the wheels loose - very disappointing!
The critical factor in drive motor selection is not whether the motor is brushed or brushless, but rather the design purpose of the motor. Hobby brushless motors and the Astroflight Cobalts were built to spin model aircraft propellers. They are not suited to applications where they bog down and pull high amperage for extended periods. A quality drill motor, in contrast, is made to produce big torque and survive bogging down at high amperage.
There are very good reasons why so many successful robots use drill motors. Save the brushless and cobalt aircraft motors for weaponry. DeWalt rules the drivetrain kingdom!
A: Mark J. here: I'm puzzled - what makes you believe it's shorted to the case? What symptoms does it show? Does it run?
A continuity or resistance tester will tell you quickly if it's shorted - put one lead on the case and the other to either of the input leads. If there's current flow, it's shorted. Unless there is something very obviously awry with the brush holders, the problem is likely insulation melted off the armature windings. That isn't something that can be reasonably repaired -- spend $17.50 and buy a new one.
A: You're providing way too little information for me to give you an answer. Read
A: RS-370s are commonly run at 12 volts, but overvolting always carries some risk. See
A: No - the gear reduction ratio is too large. Even if greatly overvolted the robot would be way too slow with any reasonably sized wheels. See
A: Greater motor power does not translate directly into greater pushing force. Pushing force is limited by the weight on the driven wheels, the tire compound, and the arena surface. Excess power will simply spin your wheels.
How much speed you can use is limited by the size of your arena and your skill as a driver. More speed than you can control is more dangerous to your robot than to your opponent.
Torque can be turned into RPM and vice-versa by changes in gear ratio. The question you want answered is 'how many watts of output power should my motor have?'
See
I know that you don't have enough information to diagnose the problem exactly, but could you tell me what are several possibilities?
A: Things to check:
A: It isn't just that motor -- BaneBots has dropped its entire line of small planetary gearmotors and replaced them with larger, heavier gearboxes aimed at BotsIQ and FIRST teams. Rumor is that there just weren't enough sales to justify ordering fresh stock of the small gearmotors.
A few of the old gearmotors are on their 'Closeout Specials' page, but when they're gone I don't think you'll see any more."
Q: Following up on my question regarding the banebots motors, what would be an alternative beetleweight 2-wheel drive motor that isn't too heavy?
A: The B16 is about as close a match as I can find. At 11.1 volts the performance is similar to the BaneBots RF-370 16:1 gearmotor at 7.4 volts, and it weighs about an ounce less."
Q: Should I keep the banebots already in my robot, or should I replace them with the B16s?
A: I don't see an advantage to switching unless you're out of spares and worried about it. You can still get replacement motors from BaneBots - it's the 24mm gearboxes that have gone away.
The B16 has a little less torque and a little less speed, so I'd stick with the BaneBots for as long as I could.
Q: I don't have any spare banebot motors. Should I still stick with the current ones?
A: Your call. If I had a 'bot design that exposed the wheels to impact damage I'd want spare gearboxes in my repair kit. You'll have to judge for yourself how likely it is that you'll need spares.
A: The Scorpion HX is available by special order direct from Robot Power. I suspect the increasing popularity of brushless weapon motors has caused a drop in demand for three-channel ESCs with brushed weapon motor support.
You can always go with a two channel ESC for the drive and add a stand-alone single channel ESC for the weapon motor - your choice of brushed or brushless.
A: All of those solenoids are big, heavy, and likely have more capacity than you need -- but I need to know full details
about what you're doing with that 550 motor before I can recommend a control device for it. I may be able to come up with
something smaller and lighter.
Q: i am planning to use the motor to drive my 3kg sumo robot. i would need something powerful enough to control the motor.
A: You REALLY don't want a sumo robot controled by solenoids. A solenoid is a simple on/off switch. It will provide either no power or full power - nothing in between and no reverse.
To have any reasonable control of the motor you will require an Electronic Speed Controller (ESC). An ESC will be lighter, will require no interface to work with your R/C system, and will provide full-range speed control in both forward and reverse. See
P.S. - That motor is wild overkill for a 3kg sumo bot. It will overpower the tire grip, spin the wheels, and add no additional push.
Q: can you recomend one?
A: In order to recommend an ESC I need much more information about your robot. I know it's weight (6 kg) and the type of motor you plan to use (Duratrax 550). In addition I need to know:
Example: in a 6 kg robot with two Duratrax 550 motors running at 12 volts with 3" wheels and a 16:1 gear reduction -- each motor will consume 14.3 amps in a full throttle stalled push, so a 20 amp ESC would do very nicely. Changes to the wheels, gearing, or voltage could change the amperage requirement by a large margin!
The calculation above confirms my opinion that the motors are massive overkill for a 6kg robot: the maximum traction the robot can provide requires less than 10% of the maximum torque of the motor to overcome. That much power will be useless in a small sumo arena. Pick some smaller motors!
A: Previously answered - with a diagram. Search this archive for "Can you run 4 motors off a 2 channel speed controller?".
A: That's a mighty thick dustpan, Anthony. You sure of that measurement? The test of a 'best' robot is winning matches, so we'll wait and see about your hammerbot.
The
Team Tentacle Torque & Amp-Hour Calculator says your current drivetrain is really slow. Even running at 12 volts, top speed with 1" tires is well less than 0.5 MPH. It's hard to find something that would work well in a beetle with only 1" wheels. The BaneBots 10:1, 25mm FF-180 Gearmotor [no longer available] is a possible choice since your wheels are well protected. Drop the voltage to 6.0 volts and you'll get zero to five MPH in less than 5 feet with 1" wheels. Amp drain is less than 2.5 amps at full push.
A: It's typical for a brushless motor of a specific physical size to be offered in several different versions based on the diameter and length of the wire used to wind the stationary magnet coils. Larger diameter wire of a shorter length gives a motor that spins faster, produces more power, and consumes more amperage. Amperage produces heat.
In robot weapon applications, brushless motors are stressed close to their maximum ability to get rid of that internal heat before it builds up and literally melts some motor component. The Park EFL-370-1080 is wound with a greater length of smaller diameter wire than the EFL-370-1360. It will produce less internal heat and has a better chance of surviving in a robot weapon application.
A: The Ampflows run so quietly that you might forget to turn it off. Combat robot builders wouldn't care if they screamed like banshees but if you have a suitable power supply and want a quiet, reliable, and compact motor you are well advised to buy an Ampflow.
A: See
A: See
Q: I will be using it in a heavy pushing and lifting.
A: Read
A: Directly drive? How are you going to do that? Something has to interpret the output signals from the radio receiver and use that information to control the power flow from the battery to the motors. That has to be either a speed controller or a crude relay system with an interface. You can plug a servo directly into the receiver only because it has a small Electronic Speed Controller (ESC) built in -- you can't just plug a motor directly into the receiver.
For a start I'd suggest you browse this archive to get some background and useful links for ESC selection and usage.
First of all, what is your opinion on using them for driving the robot around the arena? From the specs manufacturers give me, it looks like brushless motors are much more powerful than brushed motors for the same weight. As any 15lb robot builder knows, any way in which one may save some weight is welcome. However, after looking through some of the other posts and answers, this doesn't seem like such a good idea because of
A: As discussed previously in this archive, it is our opinion that brushless motors are in general >> NOT SUITABLE << for use in combat robot propulsion drivetrains. Briefly:
Q: My second question about brushless motors is in regard to using them to drive a weapon. This year we would like to drive our weapon weighing about 5 lbs at 15,000-20,000 rpm. We realize it will take quite a bit of power to get a weapon up to this speed, as brushless motors are designed to turn lightweight airplane propellers, not heavy, destructive weapons. Therefore, we have been considering using the combined power of two brushless motors to drive our weapon. However, we are worried that the motors might spin at slightly different speeds, and therefore will constantly be fighting each other and draining power and performance. Is this true? If we could in fact run two brushless motors on our weapon, could we even possibly run both of them on the same shaft? Thanks so much!!
A: Motors for rotary weapons spend most of their time sustaining spinning mass at high speed, which is entirely within the design scope of brushless motors. However, there are serious drawbacks to spinning a weapon at 15K+ RPM. Check the
You will be interested in the Team Run Amok Spinner spreadsheet [requires Excel] which models rotary weapon performance based on the mass and shape of the spinning element and the power input from the weapon motor[s]. The spreadsheet graphs RPM and energy level of the weapon over time and can calculate the approximate battery power required.
I have seen many robot weapons powered by multiple motors. A full explanation of why this isn't a problem would take too long, but consider:
A: I'm guessing that you have a BaneBots 36mm gearbox fitted with the RS-385 motor [others were offered] and that you are happy with the performance of that motor in your drivetrain design.
It isn't out of the question to run the RS-385 at 13.2 volts [I've seem them run at 12 volts] but it's really pushing it, and the gear ratio would be very wrong. You could also 'turn down' the drive motor voltage with the 'Travel Adjust' function on your transmitter, but since you asked for a replacement I'll find one for you.
The simple replacement would be a 'long can' version of the same motor, like the RS-395. Speed and acceleration at 13.2 volts are a little better than the RS-385 at 7.2 volts. Maximum amperage draw in your drivetrain is about half that of the RS-385. Weight is about 0.6 ounce greater per motor. The motor shaft has to be shortened a bit to fit within the gearbox.
The battery capacity requirement for the RS-395 is less than that for the RS-385. You haven't told me which of the HK-30 series motors you plan to run or details of the weapon, but you should still be fine running two drive motors plus an HK-30 weapon motor from your 2300 mAh pack.
Q: Thank you so much! This will help me enormously in my design, and the Scorpion motor I would be using would have ben the HK30/1000kv to a 1:1 ratio or someting similar. Thanks again!
A: You're welcome, Daniel. The RS-395 makes a nice drive motor for hobbyweights but is generally overlooked.
You don't mention what type of weapon you plan to spin with the HK-30, but you probably know I don't like direct drive spinners -- a weapon spinning 13,000 RPM isn't going to get any 'bite'. See the
However, there is a special case where two ESCs can be used to drive a single motor. Some large robot motors, (Ampflow, Bosch GPA) have two pairs of brushes set perpendicular to each other in the commutator plane. If you electrically isolate these brush pairs, each pair can be driven by its own ESC and the amp draw will be split between the controllers. I have seen this done successfully, but it is not a recommended approach -- buy a properly rated ESC and avoid risky design.
A: Mark J. here: electronic speed controllers are very carefully designed to make best use of their electronic components. Modifying is both risky and ill advised. Read Chuck McManis' article on Understanding MOSFET Current Ratings for a good background in power transistor application.
The Infineon BTS7960B power chips used by the Scorpion XXL are not well designed for external heat sinking, but you might gain a very little capacity by adding a dab of heat sink grease to each of the power chips on the bottom of the board and mounting the ESC with the chips pressed down to an aluminum panel. My advice: mount the XXL in a well protected area with good ventilation and let it operate as intended. If you need more current capacity, buy a higher rated ESC.
Q: What do you think about attaching heatsinks to the MOSFETs, sealing the assembly from the air (to protect against condensation), and putting the board in a very cold environment between matches, like a refrigerator or ice bath?
A: like I said, the TO-263 case on the BTS7960B does not lend itself to an add-on heatsink. You would need to remove each chip from the board and re-mount them to gain access to the metal heat plate on the back - a tricky process with a surface-mount chip. You could also replace the BTS7960B chips with BTS7960P chips and gain a TO-220 case with extended metal heat sink tab, but that's a lot of work (and again tricky).
Chilling the whole ESC would be like putting a light bulb in the refrigerator -- a couple seconds after it was switched on, it would be up to the same temperature an unchilled controller would be. You have to gain greater heat dissipation if you're going to gain current capacity. My advice given above stands: if you need more capacity, buy a different ESC.
A: Mark J. here: in theory a fanless Victor 883 will handle 30 amps continuous. Team Test Bot uses fanless Victors to control DeWalt drive motors in their current 30-pound robots, but I figure their motors are only drawing about 15 amps. I assume you're not really pulling 30 amps 'continuously' for three minutes.
Victors have no current limiting circuitry and no heat sink; they rely on airflow to dissipate heat from the power FETs. That makes them comparable to a hand grenade, and removing the fan is like pulling the pin. With no heat sink to provide 'thermal mass', a quick power peak can cause an equally quick temperature rise and an ESC failure with absolutely no warning.
I can't recommend a specific maximum power rating at which a Victor 883 is 'safe' without the fan. If I did want to run a Victor 'sans fan' I'd add at least a small aluminum heat sink to each FET to provide a little thermal mass and heat dissipation boost.
A: Immersion in clean fresh water shouldn't cause any direct damage to small gearmotors, but if you let them sit around damp they may quickly rust internally. I'd suggest a tear-down ASAP to dry everything out. Put a small drop of oil on each of the bearings as you reassemble, check that everything turns freely, and you should be ready for a low speed test.
I've heard of people heavily modifying 18v hammerdrill drivetrains to cut out the superfluous components. This is usually done for cost efficiency, but is very complex. I know it depends on many variables, but have you heard anything about these custom mods being more powerful than the dewalts? I would spring for the DeWalt Powerdrives, but they seem overly heavy, especially when I want to use high speed which doesn't even use one of the stages.
The only alternative I've found is the chinese Magnum 775 Planetary Gearmotor. The size 775 motor on the end is quite poor, so I'd replace it with a more powerful one. Should I totally ignore this option because of it's assumed lack of quality, or is there a possibility of some minor modifications that would make them more reliable?
I've been considering various motors to replace the stock one: the BaneBots RS-775, the HTI Motor, and the Mini-EV. The Mini-EV seems to have the highest rpm, but I don't know if it can withstand the resulting current draw. The HTI is run @ 12v in the equipment it was originally in, so is it possible that I could run it @ 23.1v or is that crazy talk? The BaneBots is the slowest and draws the fewest amps of the three, and actually seems pretty similar to the DeWalt 18v in terms of power and RPM.
I've been using the Team Tentacle calculator but I've been struggling to reach a conclusion. Here are some of the specs: the robot will be @ 23.1v, weight 45 lbs (15 lb + magnet), and be using wheels somewhere around 3", though that's a variable.
Thanks so much for your time.
A: You haven't told me what type of robot this is and what your performance expectations are. I can't recommend a drivetrain without knowing what will be expected from it. Fifteen pounds plus another thirty pounds of magnetic downforce -- must be some kind of pushybot, no?
I can tell you that any of the motors you mention are overkill for a 15 + 30 pound robot. The DeWalt setup in 'high' gear with 3" wheels will break the tires free at only about 1/6th of the available torque, and acceleration would be blinding because you're only dealing with 15 pounds of true mass. Power is NOT going to be a problem.
I can also tell you that the DeWalt Powerdrive is a very reliable drivetrain that can take enormous abuse. They have powered a great many successful robots. I have no experience with the Magnum gearmotor line and cannot comment on their quality.
The HTI motor is commonly run at 24 volts in combat robots. The Mini EV and RS-750 class motors are commonly run at 18 volts but should be OK at your suggested voltage in such a light robot.
With the information you have provided I really can't offer any additional guidance.
A: It isn't clear from your question whether you are looking for upgraded replacement gears for the Harbor Freight gearboxes or for tougher gearmotors to completely replace the HF units.
Getting a better set of gears for the planetary gearbox isn't really an option. Several problems:
A: You're not doing anything wrong with your search, you're just searching for three incompatible criteria in an ESC: good, light, and 65 amps. I can get you any two of those but not all three. You might consider a small power polenoid.
A: We've never used AstroFlight motors, although they have found application in a number of successful robots. The 'cobalt series' are all quite small and light for their output and have proven to be high-quality. It would help if you told me what use you had planned for the motors.
A: Previously discussed. Short answer: hobby brushless motors are built for model aircraft and airplane guys don't buy motors based on torque. Our Run Amok Spinner spreadsheet includes a calculator to estimate stall torque from specs that brushless motors do list.
A: We have no experience with the Magnum 550. It looks like the old style Banebots gearbox, but I can't comment further.
A: Not very much info to work with... I need some clarification.
When you say channel 2 does not work, do you mean that in mixing mode (DIP switch 1 'up') both motors run forward and backward in response to channel 1 input, but that you have no steering control from channel 2 input? What happens if you swap the ESC plugs in the receiver -- plug Ch 1 into Ch 2 and vice versa?
Q: In mixing mode, the motors had no response to the Turn(ch.2) whether the the Ch.2 was plugged into the aileron channel or elevator channel, the motors would not respond to the steering
A: That rules out a problem with the radio. One more quick test: set DIP switches 1 and 3 'up' (on) and the rest down (off). Leave the 'flip' wire unplugged. If you still get no channel 2 response I suspect that your ESC is faulty -- time to send it back.
A: I suggest that you:
A: That flashing blue light likely means the ESC is in 'lithium mode' and is attempting to count off the number of lithium cells it detects in your battery pack. Continued flashing means that it isn't happy with what it finds. Are you using a lithium battery? Is it fully charged? Check the manual and correct any ESC DIP switch setting errors. Switch #3 should be down (off) for lithium batteries and up (on) for other battery types. Charge your battery before continuing.
A steady dim blue light indicates that the ESC has power and functioning correctly. Turn on the transmitter and the blue light will brighten, indicating the presence of a radio signal. You should be set to go.
A: Yes, but the key word is 'sometimes'. Victors rely on airflow rather than heat sinking to stay healthy. Without the airflow they will handle a great deal less power. Victors don't have current or thermal limiting circuits and will simply melt if they can't get enough cooling. I don't have a factor to de-rate a Victor without the fan -- if you try it (I wouldn't) you're on your own.
If you're hurting for weight/space, you might consider the SyRen 25 amp controller. It is less than half the weight of the Victor 884, more compact, and has no fan. Rated at 25 amp continuous, 45 amp peak - it has the current/thermal limiting circuitry that is missing in the Victor and is less likely to melt if overloaded.
A: See
Thanks, Carlo Bertocchini
A: Glad to help, Carlo. Thank goodness for 'search and replace'. Also changed all the links to point to your AmpFlow website.
A: I suppose you've tried to run the motor with unsatisfactory results? You can try putting it back in the vise 90 degrees off from the way you squashed it before and see if you can bend it back sorta round. That's not exactly a high-precision repair, but you don't have anything to lose by trying. If all else fails it will make a very pretty refrigerator magnet.
A: Stalling a motor for two minutes(!!) is a very bad thing. It is entirely likely that the motor was heat damaged during that time. The wire insulation on the armature may be cooked, the commutator may be warped or pitted, and the brushes may be burned. Normally I would be suspicious that the new gearbox had some type of clearance problem when installed and was binding -- but if the motor is slow and overheats even when not attached to the gearbox then it's toast. Replace it.
A: NO!! BaneBots will sell you the correct 2.3mm pinion gear for a Speed 400 motor. Use it!
Q: Speaking of which, how do you take the motor out of the banebot gearmotors?
A: Remove the screws holding the gearbox together and pull the gearbox free from the motor mounting plate. You'll see the two screws that hold the motor in place.
Q: Pinion guy again. I was actually hoping to put a speed 400 into the 24mm 16:1 gearbox. My hope is to use 2 16:1 banebot motors to power a hammer. The arm is 6 inches and the weight is 2.5 ounces. Could they survive being directly connected to the hammer arm?
A: The Speed 400 does not have the same mounting screw spacing as the BaneBots 24mm gearbox -- you'll have to modify the mounting plate a little. BaneBots does sell a 2.3mm pinion for the 24mm gearbox, and this modification has been done before.
I'm curious -- how did you settle on two of the Speed 400 motors conected to a pair of the 24mm 16:1 gearboxes? Why not the 9:1 or the 20:1? Why two? I haven't done the calculations for the weapon, but your choice seems kinda arbitrary.
I can't tell you how long the motors would survive as hammer actuators. Anytime you use a component for something it isn't built to handle, you're on your own as far as reliability goes. Just guessing, I don't think the gears are going to last long -- the sudden stop is going to be hard on them.
A: Mark J here: the short answer is that greater power handling requires an increase in the capacity and number of power transistors and power capacitors. Increasing the number and size of those components also increases the need for cooling systems (fans, heat sinks) to remove heat build-up. For a full answer see 4QD-TEC: Electronics Circuits Reference Archive - PWM speed control.
A: One of the benefits of paying for a quality brand motor is that it comes with specifications that can be relied upon to be accurate. A Chinese knock-off may or may not come with specs -- if it does you shouldn't count on their accuracy (IMHO).
A: You need to remember that brushless outrunner motors are designed for model airplanes where they are commonly mounted behind a propeller that pumps a whole lot of air around them. Their amperage rating already assumes continuous airflow cooling -- they should be de-rated for still air applications (like robots). The killing heat is deep in the dense and compact windings of the motor that don't have enough surface area to dissipate heat any faster. If there was a way to get more power per ounce out of these motors, the airplane guys would already be doing it.
A: That's right, Kv is no-load RPM per volt. If the motor in question will operate at 20 volts, it would spin (with no load) at 40,000 RPM (2000 times 20) at that voltage.
Q: Is there a limit to the speed of brushless motors? What if I have a 5000 kv brushless inrunner at 20 volts, would it spin at 100,000 rpm? Is that possible?
A: Mark J here: any specific brushless motor system has an RPM limit. Brushless motors come with an operating voltage range, and pushing a high-performance motor beyond that range is done at your own risk. The RPM limit may be the point where:
A: Second question first - you have to have SOME WAY to control motor direction and speed. I have seen (and built) some slow moving robots with motors controlled by power relays that provided very simple forward/off/reverse control of each motor. While this is adequate for slow robots the savings in expense or weight is small because large power relays are both expensive and heavy, and you will also require an interface between the relays and the radio system. Electronic Speed Controllers (ESCs) are used because they are the best option for weight, control, and reliability.
Back to the first question - different manufacturers rate maximum current capacity in different ways, and trying to compare numbers can be very frustrating! Ratings are often quoted for a 'cold' (room temperature) ESC and with an associated time limit, such as '220 amps for 30 seconds'. There may also be a higher 'peak' amperage rating that may be sustained only for an instant. As the temperature of the ESC rises, the amperage capacity will drop. Mounting the ESC to a large heat sink may extend the time capacity at a given amperage. A well designed ESC will have 'current limiting' circuitry that will reduce the current as the temperature rises to prevent thermal failure.
The highest believable and substantiated amperage rating that I know of for an 'off the shelf' ESC is 320 amps (cold) 240 amps (hot) for the 4QD-300.
Note that an ESC may not have to handle the maximum (stall) amperage rating of a specific motor. If the motor is geared such that it provides enough torque to spin the wheels before the motor stalls, the maximum amperage actually used will be reduced. The
Team Tentacle Torque & Amp-Hour Calculator can estimate maximum amperage consumption for specific motors in specific applications.
Q: Aaron, in response to your expensive and heavy relays: [link to R/C 10 amp on/off switch deleted] [endorsement of specific website deleted]
A: Mark J here: the R.C switch for which you sent a link is not suitable for forward/off/reverse control of a motor. That type of control requires a pair of double pole double throw (DPDT) relays wired into a 'H-bridge configuration' -- the switch in question is a single pole single throw (SPST) switch with a fixed trigger point which cannot be used in an H-bridge.
Aaron's comment about heavy and expensive relays refers to high-amperage relays required to handle hundreds of amps, but suitable smaller mechanical H-bridges are also relatively heavy and offer little savings compared to ESCs of similar amperage ratings.
A: Mark J here: my previous statement applies to the use of hobby brushless motors in combat robot drivelines -- they work well in properly designed spinning weapon applications.
The first problem with using hobby brushless motors for robot drivetrains is finding a suitable reversing brushless speed controller. A combat robot that can't back up is entirely useless.
The second problem is the nature of the loads placed on drivetrain motors in robotic combat. Hobby brushless 'airplane' motors pack a great deal of power into small packages. As long as these motors are used for power output in the higher RPM range at which they are designed to operate their amperage consumption stays within reasonable bounds and everything is fine. In R/C cars and similar applications, these motors only rarely and briefly operate in the lower RPM range.
In combat robot drivetrain applications motors are often 'bogged down' in pushing matches and in rapidly alternating forward / reverse maneuvers -- with a resulting sharp increase in amperage consumption. More amperage produces more heat. These motors are so small compared to their amperage consumption under these conditions that they simply cannot dissipate enough heat and they 'melt'.
Q: Hey, the guy with the armor and brushless motor questions again. I recently picked up two speed controls (for brushless motors) that have reverse, forward and braking. $37 and they do 100 Amps continuous, 580 amp bursts (which I think should cover the "high" amperage for pushing). I'm using two motors with a combined 2000 Watt power. I'm pretty sure they will get extra-hot, however I'll have fans and also the match is not very long. If worse comes to worse I'll buy extras.
As for the operating in the high RPMs, I'm pretty sure I'll have enough power to spin the wheels the entire time so it'll stay up in the high RPMs :).
The motors I'm using are 320kV, so I'll only need a 3.4:1 gear reduction to run 15 ft/s at 14.8V (4s lipo). The problem I'm running into is that it is hard to find pulleys for large pitch timing belts where I can get a 3.4:1 gear reduction and still have the taller gear stay under my 1.75" robot height. Any suggestions for my 3.4:1 gear ratio? I could do a double gear reduction, however I'd like to minimize the number of belts. Currently the only pitch that works is MXL (2.03mm) which I'm pretty sure will not be able to handle the torque created by my brushless monsters (at 1/4" belt width).
My entire drive system has to fit within (inner dimensions) a 1.5" extruded square aluminum tube. Any suggestions would be nice.
A: I have several comments:
Q: [Received before above answer was published] Okay, about the brushless motors, again (sorry). I'll measure the resistance when I get them but I'm estimating around 30mOhm (from similar motors, these didn't have a rating!). This results in ~ 500 amps at a stall, within the capabilities of my controller but I doubt within the capability of the motor. As the motor heats up it will not behave ohmicly, correct? It's resistance will increase so this number will actually be much much lower.
I'm sure I could drive within these constraints (IE attempt to not stall it for more than brief moments) however this is not practical. I think I would be better off going with a higher-resistance motor (which is usually proportional to motor size) so that stall currents are not so large, and even then the motor should be able to handle it.
In fact my earlier statement about the wheels spinning could also be true... I may have enough torque to where I can, in fact, not stall the motor (for more than brief moments) and this set up will actually work fine.
I know you prefer questions that require brief answers, but I'm finding it hard to find information on brushless motors being stalled (because no one uses them in these applications). Thanks again..
A: No apology needed. I know how much thought goes into designing a robot. You're just getting a little ahead of yourself (and me).
You've learned one of the reasons why I recommend using brand name brushless motors: they come with full specifications. The Chinese knock-offs are inexpensive, but you have no idea what you're getting. Mounting one of these in your sportflier airplane to give it a try is one thing, but designing a combat robot around a pair of them is another matter.
Heat will raise the terminal resistance of the motor, but not by a great deal. The motor impedance raises linearly as the motor gains RPM due to backward-flowing electromotive force (EMF) that resists the current flow. That's why it's so important to keep the RPM up in the high range and to avoid stalling the motor.
As noted in the preceding comments: given the power of these motors, the weight of the robot, and the gear reduction chosen, there's no way that you're going to be able to stall the motors in this robot short of welding the axles to the chassis. You can expect to use only around 10% of the stall amperage (and torque -- they are linearly related) of the motors before the wheels spin freely.
You've stumbled across one solution to using brushless motors in combat robots: overkill. Selecting smaller, lower-output motors would appear to make sense, but their lower power also means that it's easier to bog them down and drop them into the amp consumption danger zone. No, if you really want to use hobby brushless motors in a combat robot I think you want to stay well up in the 'overkill' zone.
There is a very good reason why no one uses brushless motors in applications where they stall -- they melt! And quickly!! I think that's all the information you need to know about stalling them.
Q: Of course I do not believe 580 Amp ratings, but like you mentioned I'm banking on overkill (in every aspect). You mentioned using lower-powered brushed motors, however after a lot of research I found that most builders use overpriced victor speed controllers @ $200 each.
I think I'd be better off sticking with overkill at $140 (2 motors 2 speed controllers) than going with under-kill at $400+. I'm also fairly confident that at 3.2lbs (motors, ESCs batteries and wires) this will be the lightest 2000 watt drive system with 5Ah of 16V that I can find (probably lighter than brushed systems at lower power too).
As for the soft start I'm sure I can mess with the programming to fix that, if it happens to be a problem. What are your thoughts on the weight/cost/effect when comparing to a lower powered brushed system?
A: Victor speed controllers are not overpriced. They are designed and built specifically for combat robotics. Builders use Victor controllers because they will deliver every last amp promised in their specifications and do not fail in the middle of a match. You will discover the true value of this after you have played around with those hobby ESCs for a while.
The motors and controllers you have proposed do NOT constitute a 2000 watt combat robot drive system. The components are undocumented and unproven, the specifications are fictional, and if output should ever approach 2000 watts multiple components will fail from thermal overload. The drive MAY work in your proposed featherweight only because the traction available to a 30-pound robot will limit the effective output to around 350 watts.
My opinion continues to be that hobby brushless motors intended for model aircraft use are unsuitable for combat robot drivetrains for the reasons previously given. Perhaps you will prove me wrong.
A: I've answered this question previously as part of related questions, but I'll take this opportunity to give a complete answer. The answer is yes, multiple brushed motors may be controlled by one channel of an Electronic Speed Controller. There are a few conditions:
A: Mark J. here: what you're asking for is called multi-quadrant control -- see this article on brushless motor control for an explanation. Multi-quadrant control is a fairly common feature on high-capacity ESCs for brushed motors, but I don't know of any hobby brushless motor controllers that include a generator mode.
You can find references to industrial four quadrant brushless controllers by searching for: brushless quadrant motor controller.
A: IFI does not quote peak current for the Victor 884. I'd hesitate to push it past 100 amps, and that only for a blink. The IFI controllers are fan cooled, have no current limiting circuit, and are very easy to fry if overloaded.
Q: Do the ifi victors 884 have a BEC?
A: Nope - no battery eliminator circuit.
A: Mark J. here: I'll give it a shot. First, a couple of clarifications:
The amount of rotational force that a motor producing 100 N-cm of torque can exert is dependent on the distance (radius) from the center of the shaft to the point where the force is exerted. If a 100 cm long bar is attached perpendicular to the motor shaft, the force exerted at the end of the bar will be 1 newton (100 N-cm ÷ 100 cm = 1 newton). The force at the end of a 10 cm bar attached to the shaft would be 10 newtons (100 N-cm ÷ 10 cm = 10 newtons). If a wheel is attached to the motor shaft, the radius of the wheel is used to calculate the force available.
This available force is not directly related to the mass of objects the motor can 'pull' along a surface, it relates to the weight of objects the motor could lift. A rope wrapped around a 10 cm radius wheel on a 100 N-cm torque motor could exactly offset a mass exerting a 10 newton gravitational force on the other end of the rope -- the mass would neither rise nor descend. A smaller mass could be lifted by the motor, and a larger mass would overcome the force of the motor and descend.
A: Mark J. here: the RS-550 motor is overkill for a 3kg 'bot. Paired with the BaneBots P60 26:1 gearbox and 4" wheels, a pair of the motors would give a top speed near 9 MPH (plenty!) and still be able to break the wheels free when pushing using only 5% of its peak available torque. Huge overkill - better suited to a featherweight pusher.
The BaneBots RS-385 gearmotor would be a better match. It's lighter, costs less, will run great on a 2-cell LiPoly pack, and a pair will still deliver that 9 MPH top speed with 4" wheels and will break the wheels free when pushing just at peak horsepower output. I'd go with that!
Q: then what about RS-540 motor?
A: Let's not play 'then what about this motor?' You have my recommendation.
Given the very limited information you've provided, the BaneBots RS-385 motor is a great fit. Select the gear ratio to suit your wheel diameter.
If you'd like to evaluate other motors, try the
Team Tentacle Torque & Amp-Hour Calculator
-- it's what I use.
If you won't be happy without overkill, go for it.
Q: What motor other than the mabuchi motor will you recommend for a 3kg sumo robot? My requirement is a robot with unbeatable pushing power, and speed. the robot runs on 2 in wheels.
A: As previously noted, we do not compete in sumo and are not privy to the secrets that experienced sumo teams have for generating 'push'. However, we do know some things that won't work.
Excessive motor power is not the answer to unbeatable push. A wheel-driven vehicle can only produce a certain amount of 'push' before the wheels break free and start spinning. Power beyond that point simply spins the wheels faster without providing additional 'push'. What do you have against Mabuchi motors?
For best pushing power you will want the entire weight of the robot to be supported on powered wheels - that usually means 4-wheel drive. Are you designing a 4-wheel drive robot?
There is no point in gearing your motors for speed that the robot cannot attain within the confines of a small arena. I suspect you're really asking for quickness, but a very quick robot can be trouble in a small arena. I cannot recommend gearing without knowing the arena size. How large is your sumo arena (dohyo)?
Q: The ring is 154cm in diameter, i am running my robot on six 2in wheels and it is r/c controlled.
A: OK, a pair of BaneBots 36mm 20:1 RS-385 gearmotors [no longer available] driving 2" wheels will propel a 3 kg 'bot to a top speed of 4.7 MPH in 4.3 feet, crossing the ring in less than 1 second. They will be able to break loose all six wheels when pushing while only using about 1/4th of their available torque. You'll have your hands full trying to control that quickness and speed in such a small ring -- you may end up dialing back the throttle response to make it more controllable.
What type of drivetrain will you be using to power six wheels? You're not planning on six motors, are you?
Q: No, i will use 2 motors instead. drivetrain will depend on the motor. I need 2 with a combined 1/4hp motor. it is best if it comes with gearbox.
A: Why so much power? There is simply no way to use that much power in an arena 5 feet across -- it would be like trying to drive a rally car in your living room. You'd be more of a danger to yourself than to your opponent. You aren't the first builder to decide that massive power is the answer to the sumo challenge, but look around at the motors used by successful sumo competitors and you'll discover that much more moderate power is what wins.
I've given you my recommendation - twice. If you don't like it, feel free to use any motor set-up you please.
Q: Okay, I can forget about speed. But is there any motor with gearbox out there that fits my requirement of at least 100 rpm and produce alot of pushing power?
A: Why yes, there is! I don't know why I didn't think of this before: the BaneBots 36mm 20:1 RS-385 gearmotor spins more than 1000 RPM at 9 volts and has four times more torque than you can possibly use. MAYBE YOU SHOULD TRY THAT!
Thank you ! That makes my trip to the singapore robotic game 2011 much more easier.
A: Given no information about the design and performance needs of your robot, I can't even start to recommend a motor. See
Take a look at the
Team Tentacle Torque & Amp-Hour Calculator
-- it will allow you to evaluate the performance of a wide variety of robot propulsion motors. The BaneBots 36mm 16:1 RS-540 motor might be a good place to start.
A: Mark J. here: the 'working' current of a motor depends on the load placed on the motor and the efficiency of the motor in that load range. For drive motors, the
Team Tentacle Torque & Amp-Hour Calculator
can provide an estimate of working current based on the specific motors, voltage, gearing, wheel diameter, and robot weight. For weapon motors, the Team Run Amok Spinner Excel spreadsheet can provide a similar estimate based on the specific motor, voltage, gearing, moment of inertia, and weapon usage.
A: The spindle motor from a hard drive is a brushless permanent magnet DC motor. Back before purpose-built brushless motors were available for model aircraft, builders were modifying brushless hard drive and CD ROM drive motors for hobby purposes. The computer drive motors were designed for long life and stable output, not the type of high output applications hobbyists required, but they managed enough success to prompt the development of the wide range of hobby brushless motors now available.
You will need a suitable brushless motor controller to operate your salvaged spindle motor, and you will not have much guidance on the acceptable voltage level, or the output RPM and torque. I can safely say that the motor would be suitable only for insect-class applications and would not perform as well as a similarly sized brushless hobby motor. I wouldn't bother with it.
A: Right. I want to know that for some words put together you can make readable question.
A: Neither option is particularly good. A re-bent tooth has been weakened and is likely to fail, while a missing tooth is just asking for trouble. I can't recommend going into combat with a weakened gearbox -- the safe thing to do is replace it.
Q: can I get a new gear or gear case or should I just buy the whole thing?
A: You could ask around on one of the forums to see who has a Copal gearbox left over from a blown motor, or you can buy a new 'whole thing' and have a spare motor ready in your parts drawer.
A: There are two versions of the Park 370 outrunner. Using the brushless stall torque calculator from the Team Run Amok Spinner Excel spreadsheet with the following data from the Hobby Zone website for the 1360 Kv model EFLM1205:
A: You can match the three wires from the ESC to the three wires from the motor any way you like. If the motor turns the wrong direction for your application, swap any two of the wires and the direction will reverse.
A: Brushed motors should be 'broken in' to allow the countour of the brushes to wear down to match the curve of the commutator. This will provide maximum contact area between the brushes and commutator which will minimize heat build-up, arcing, and potential damage when run at full voltage under load. The basic procedure is the same for all brushed motors:
A: Let's start by getting the numbers right. The Pololu 20D44L weighs 1.7 ounces, just about 3/4 the weight of the B16 at 2.3 ounces -- not half.
There are a lot of factors that go into gearmotor design: design voltage, expected longevity, efficiency, required strength, etc. Let's compare the two at their best voltage. At 12 volts the B16 produces 12% more stall torque, 35% more RPM, and more than twice the total output power of the Pololu at 6 volts. I haven't used the Pololu motor, so cannot comment on its durability.
A: I have no idea. Automotive starter motors are generally inappropriate for use in combat robots. They are designed for high loads for very short time periods. Starters are usually 'series wound' motors that have a different torque curve than permanent magnet DC motors, which makes weapon design computation more difficult. Builders have tried to use starter motors for weapon power, but I don't recall any that were particularly successful. I recommend against their use.
A: Mark J. here: your approach is well thought out and will work -- if your ESC is actually limiting current the way you think it is. Many 'current limiting' ESCs don't directly limit current, they just dial back the PWM in response to the heat level being generated by the FETs. The 'programmable' current limit may just be setting the response to a thermistor on the heat sink. When cool, these ESCs will provide more than the limit setting; when good and hot, less. Check with the manufacturer to see how the current limit works with your ESC.
A: Mark J. here: Team Whyachi would have been my recommendation, and Odyssey would have been my backup suggestion to NiCads. You're on the right track -- best luck.
I'm the one trying to do an upgrade for power soccer, and I haven't been able to find a suitable gearbox for the Magmotor motors anywhere. Team Whyachi emailed me back saying that 16:1 was the maximum they could do.
A: Mark J. here: first, I think you mean 'higher' torque constant. Lower armature resistance from less and larger wire will result in greater current flow at a given voltage and greater stall torque. The torque constant equation is:
Greater stall torque at a given voltage will raise the torque constant for a motor.
Back to your question about the drawback: greater power in a small package yeilds reduced service life. Current produces heat, and heat causes reduced life in the brushes and commutator and raises the potential for winding failure. Industrial motors are expected to have a service life of hundreds or thousands of hours with minimal maintenance, whereas a combat robot motor will do well to have a life measured in tens of hours before major maintenance or replacement. Your power soccer application will not be so rough on the motors as robot combat, but you can expect to replace brushes and renew the commutator surface more often for a 'hot' armature wind than you would with a conservatively wound industrial motor.
Sorry to hear that Team Whyachi cannot provide a suitable custom gearbox. I have no other contacts.
A: You've got to control the drive motors somehow -- you certainly can't just plug them into the receiver! The FingerTech tinyESC weighs only 3 grams and should work well in a fairyweight robot.
A: Only if the motor is spending time at stall, which a brushless motor shouldn't do! Hobby brushless motors are designed for aircraft use where they never stall and are rarely bogged down at low RPM. They generally come with specifications for maximum amperage and how long they can survive at that current level. For longevity you should limit the current to that maximum recommended level, either thru the ESC or thru the design and use of the weapon.
A: I would recommend not going cheap on critical robot elements. You know how well the Axi works, you know it is an exact replacement, and you don't know anything about the off-brand motors except that they are cheap. Save yourself frustration and disappointment -- buy a new Axi.
Q: Hi Aaron, I have done some more research about direct replacements for the Axi 2217/9D and found a KD A22-20L Outrunner. I know you said that going cheap is bad, but I've already killed 3 Axi's, and they are just not worth it for 90 bucks apiece.
This one doesn't show the internal resistance, so how can I figure out the stall torque? Would it be better than the Axi? BTW, the kv and length are better than the Axi for my purposes Thanks.
A: Mark J. here: if a part is failing, going to a cheaper part is NOT the solution. Why do you believe that a $14 chinese knock-off is going to be better than an Axi? It isn't. The Axi can handle 34 amps for a full minute. The knock-off dies after 15 seconds at 25 amps. If you've blown three Axis you're going to eat these like popcorn. The knock-off is also heavier, spins slower, and is less current efficient. Without an internal resistance number you can't calculate stall torque, but I'd bet that the Axi is better.
First, take a look at your design to see what is contributing to the weapon motor failures. Consider going to a less radical wire count -- the Axi 2217/16 or /12 would be more reliable, but would provide somewhat less power. Power comes at the price of reduced longevity.
A: Mark J. here: do you remember where the Spinner Spreadsheet 'said something' about this? Right there is a calculator for estimating stall torque of brushless motors. Fill in the light blue cells with voltage, rpm/volt, and internal resistance values and the spreadsheet will spit out the stall torque in bright red font to the right under the friendly label 'Stall Torque'.
If you prefer to do it longhand, the formula is:
A: According to The Builders Database antweight 'Gilbert' uses Maxon gearmotors.
A: Mark J. here: the claimed stall current for the B16 is 2.6 amps at 14.8 volts. I don't have a B16 in the shop to test and verify, and I'm concerned about the validity of that number because the same motor is rated at a considerably higher current consumption when attached to other gearboxes (B62, B104, B231). A specific motor should have the same stall current irrespective of the gearbox to which it is attached.
The Ant 100 ESC is rated 2 amps continuous, 5 amps peak. If the stall current figures for the B16 are correct, you should be fine with the Ant 100 -- but I don't trust those numbers. I would strongly recommend independently determining the stall current of your drive motors. Instructions for determining the stall current using the 'D-cell' method are in this archive - search for 'D-cell'.
The Tentacle Combat calculator suggests that I could have equivalent acceleration with lower amperage requirements if I were to switch to a pair of Magmotor A28-400 motors. The batteries presently last for over four hours of use, so we obviously don't draw 160 amps all the time. Is there any good way to verify that the Magmotor motors will provide better battery range than the NPC-T74s? It will require a lot of expense and fabrication, so I would like to do as much homework up front as possible.
A: Mark J. here: the Magmotor motors are much more efficient than the NPCs. The NPC-T74 requires about 70 amps at 24 volts to produce 1.4 horsepower. The Magmotor A28-400 will produce the same horsepower at the same voltage while consuming only 51 amps. Based on continuous output, switching to the Mags would extend battery life by about 37%. Your actual battery life may be different as I suspect you're not out there placing a constant load on the motors, but you will benefit from the greater efficiency.
HOWEVER...
The purchase of the Magmotor motors and fabrication of gearboxes represents a lot of hours and dollars. You might consider going to an alternate battery technology. Optima deep cycle batteries are well designed but all lead-acid batteries perform poorly under very high loads, delivering only a fraction of their rated amp-hour capacity. The amp-hour rating is based on a small constant drain that will exhaust the battery after 10 hours. High pulse loads can cut this rating WAY down. Other battery types offer much better performance under these conditions. A NiCad pack, for example, can deliver more than twice the run time of an identically rated lead-acid battery under very high loads. This sounds a lot easier than replacing your motors.
Q: Mark, thanks very much for the reply regarding NPC-T74 versus Magmotor motor efficiency. I had initially been looking at forms of lithium batteries, but the required capacity is well beyond the legal limit for air travel. I was in touch with Powerstream, and I was told that all of the large format NiMH battery manufacturers are out of business. Do you have any suggestions in locating or building a NiCad based pack in the 40+ AH range?
A: Get in touch with the good people at Robotic Power Solutions. Explain what you're doing and I'm sure they can build something for you.
A: It depends on what you're trying to do with that heavyweight. The 6 horsepower that a pair of A28-150 Magmotor motors can produce is certainly plenty to maneuver a weapon-bearing heavyweight quickly and efficiently around an arena. If you're building a thundering ram-brick to hurl itself as a blunt-force projectile and smash your opponent, you'll need more power.
You can calculate performance for a wide range of drivetrains, weightclasses, and designs at the
Team Tentacle Torque & Amp-Hour Calculator .
A: Stall torque is 309 g/cm @ 7.2 volts. Stall amperage is 28 amps.
A: The four mounting holes are tapped for 8-32 machine screws.
A: Mark J. here: stall current is only a factor if you're going to stall the motors. Run your drivetrain design thru the
Team Tentacle Torque & Amp-Hour Calculator
and see how many amps the motors will pull to spin the wheels.
An example 2-wheel drive hobbyweight running two Johnsons ('small') with 4" wheels and a 12:1 gear reduction at 12 volts will pull less than 17 amps per side pushing at full throttle against an immoveable object. That would be well within the 20 amp continuous rating of the Scorpion XXL.
'Thermal Protection' is a good feature to have as a back-up, but you don't want to rely on it on a regular basis. If you overload the ESC badly it may 'smoke' before the thermal protection even notices that something is wrong.
Q: Thanks for the help! I am using the motors with the team whyachi t-boxes and on the team whyachi page it says the ratio for the t-boxes is: 11.52:1 what exactly does that mean? can I round it to 12:1 on the team tentacle calculator? Thanks
A: An 11.52:1 gear ratio means that the motor must rotate 11.52 times to get the output shaft to rotate once. The output shaft will have 11.52 times the torque of the motor and the output shaft will spin 11.52 times slower than the motor.
The Team Tentacle Torque Calculator will accept decimal gear ratios, so go ahead and enter 11.52. If you did need to round, I'd suggest rounding downward for a more conservative estimate of amperage use.
A: The Briggs & Stratton Etek is no longer made. Stocks ran dry a few years ago.
Q: does Brigs and stration manufacture the Etek-R Motor?
A: No, the Etek-R is made by 'Mars Electric'.
A: That particular motor has been unavailable for years. It was an 'inrunner' style brushless motor that ran at very high RPM. For a weapon drive, something like an Axi 4120/14 would produce comparable power at a lower voltage and at more manageable RPM.
A: To a certain extent, yes. Linear actuators are powered by electric motors: make the motor faster or more powerful and the actuator becomes faster or more powerful. Overvolting is a common way to accomplish this. Double the voltage and you will double both the speed and the lifting power of the actuator. You will also substantially shorten the life of the motor.
You can speed up an electric linear actuator but you can't make it break the laws of physics. Pneumatics are much better for producing large bursts of power.
A: See question #23 in the
A: The supply of D-Packs dried up a few years ago. They were really inexpensive (about $50) for a 2+ HP motor, but they were also inefficient amp-sucking death pigs. You get what you pay for. Save up and buy a pair of Magmotor motors instead.
A: Only if you want to go really, really slow. With 2.5" wheels at 6 volts the top speed is about 0.5 MPH. It would take more than 9 seconds to travel across an 8 foot arena!
Allow me to introduce you to the
Team Tentacle Torque & Amp-Hour Calculator. You can select a motor from their long list and set the other details of your design (weight, wheel diameter, number of motors, voltage) and it will calculate performance figures. Use it!
Q: I looked at the Team Tentacle Calculator but I could not make up my mind for my ant weight wedge. I want a motor $35 or under and under 1.3 oz. Fast but still controlable. Please recommend one. Thanks
A: Hard to recommend a motor without knowing anything about your design and little about your expectations. I will say that there have been a great many successful antweights built using the
A: I don't have any information on 'Unspecified Threat', but they certainly aren't using servos for their hammers! Far, far too little power.
If you're not finding gearmotors for your hobbyweight, you just ain't lookin'. Browse this archive for discussions on appropriate gearmotors. Cordless drill motors are popular.
A: Well, you've got the whole "overpowered" thing pegged.
Twin A28-150s is a huge waste of money ($598) and weight (7.8 pounds) in a hobbyweight. You'll only be able to use 2% of the torque capability of the motors before the wheels spin, the top speed will be severely limited by the small size of a typical sub-light arena, and full-power acceleration is going to throw the 'bot off in an unpredictable direction. Not a great idea.
A: Simple: run it in both directions and listen to it. It will run faster and have a higher pitch to the sound it gives off when running in the better-timed direction. Break in the motor before you try to figure out any timing advantage -- it may change after the brushes are fully seated. Sanyo 50:1 gearmotors are supposedly 'neutral' timed, so there isn't going to be much difference.
A: A standard two-channel ESC will control two motors to perform the same function as your control system. The advantage is that a hobby ESC will allow smooth variable speed control of the motors in both forward and reverse to give greater control than the combined receiver / motor controller in your RTR boat.
You would need to determine the voltage and amperage draw of your motors to select a suitable ESC to handle the power requirements. You'd also very likely need to replace the entire radio system of your boat with a hobby-grade system in order to be compatible with the signal requirements of the ESC. By the time you've done all of that, you'd likely have been better off to have chucked the Ready-to-Run boat and bought something more capable.
A: Mark J. here: all Permanent Magnet Direct Current (PMDC) motors have an inverse linear torque-RPM 'curve': they produce maximum torque at stall and zero torque at free-running RPM. If you know the stall torque and free-running RPM of a motor you can calculate the available torque at any speed.
Running at 24 volts the A28-150 Magmotor motor produces 1970 oz-in torque at stall (zero RPM) and zero torque at 6000 RPM (free running RPM). The torque formula for this motor is:
PMDC motors have a direct linear relationship between torque and amperage. For each amp the A28-150 draws it produces 5.1 oz-in torque.
A: Mark J. here: an Electronic Speed Controller (ESC) controls motor speed by switching battery power on and off thousands of times per second. The longer the ratio of switched 'on' time to switched 'off', the faster the motor spins and the more power it produces. At 'full power' the ESC passes (almost) the full voltage and current from the battery on to the motors.
By using the Adjustable Throttle Volume (ATV) feature on a computerized R/C transmitter you can adjust the radio so that the ESC will never receive more than a 'partial power' signal and will never deliver more than partial current/voltage to the motors, but there is no automatic voltage limiting.
The above explanation of how an ESC works is VERY abbreviated. If you're interested in details on ESC function see the 4QD-TEC: Reference Archive.
A: The idea behind breaking in a brushed motor is to seat the brushes against the commutator to prevent destructive arcing when the motor is run at full voltage (or overvolted) under load. This is usually done with a variable voltage supply, but there are other options:
Make sure the motors and gearboxes turn over smoothly before starting and monitor the motors during break-in for overheating or other signs of trouble.
A: I've got a couple of questions:
Why that motor? The 36 volt DeWalt with gearbox weighs a little over 3 pounds and produces about 1 HP at 36 volts. The old-style 18 volt DeWalt with gearbox weighs a little over 1.5 pounds, produces 1.5 HP at 24 volts, and is a well proven combat winner.
Next, what do you plan on doing with a 12 pound robot with 2 or 3 horsepower? It will be very good at spinning its wheels and crashing into the far wall of the arena before you can react -- is that sorta what you have in mind?
A quick check at the
Team Tentacle Torque & Amp-Hour Calculator
shows that a hobbyweight with a 36 volt DeWalt and 4" wheels will spin it's wheels at about 5% of the torque potential of the motor. Give this a rethink.
Q: About the ramming at a high uncontrollable speed, that's kinda the point. I want to slam the other with as much KE as possible. And I also plan to run the drive with a chain a sprocket system, at a 25:1 gear ratio with five inch wheels.
A: Check your math: a DeWalt 36 volt motor geared 25:1 with 5" wheels will give a top speed of about 11 MPH - hardly high uncontrollable speeed. You've picked the wrong motor, the wrong gearing, and the wrong attack strategy. I have nothing against ram bricks, but the whole idea behind them is 'controllable' speed and accuracy.
Q: Well, what DO you recommend for a hobbyweight rammer?
A: I can't recommend specific motors without knowing more about the rest of your design and details on the arena in which you will be competing. I can tell you that trying to put down a lot of power with a two-wheel chassis will make keeping it pointing in a straight line very difficult. For a typical sublight arena I'd go for a four-wheel drive chassis with 500 to 600 watts total motor output.
My dad contacted Team LNW's Greg Schwartz and asked about the motors:
"We run Ryobi 9.6v drill motors in DD on 18.5v lipos. Had the g/b's modified by Gaussewave (Rob Purdy) so the output shaft is a 3" x 1/2" keyed shaft."
Greg noted that he plans to upgrade with motors and gearboxes from these 18 volt drills which feature hardened steel gears. Thanks, Greg!
Q: Thanks for the help!, I once tried to modify two 12 Volt Black and Decker motors but in the end it did not seem to work well. Maybe it was the drill motors themselves, because they were not cheap chinese drills? Maybe a pair of Harbor Freights would work better, or maybe BaneBots motors? [Daniel]
A: Plenty of hobbyweights are powered by BaneBots motors. They are simple to mount and have a large range of power and gear ratio options. Some builders really like them and some really don't. In a design like 'Defiantly Daft' with bearing support on the end of each drive shaft I think they'd be fine.
A: Shaft length is not included in a standard motor length measurement and is not included in the motor length given by the Robot Marketplace for the B16 motor.
The Robot Marketplace provides a great service to the robot community, but they are inconsistent in the way they report dimensions and performance figures. Mixing metric and English units or leaving key measurements out of drawings or tables is common.
A: Not even close. The Sabertooth 5 ESC can only provide 5 amps of continuous current -- strictly antweight/beetleweight class. The Sabertooth 25 ESC can provide 25 amps of continuous power, which is enough for most hobbyweights or a moderately powered featherweight. Read thru this archive for help in selecting a suitable ESC.
A: At 24 volts the Magmotor E-150 pulls about 64 amps at stall.
You can calculate the stall amperage of a PMDC motor from the Terminal Resistance and Voltage:
Most of the common robotics motors have their vital specs listed at the
Team Tentacle Torque & Amp-Hour Calculator
A: I prefer speed controllers that have power limiting features that keep the ESC from frying if overloaded. Neither Vantec or Victor ESCs have any power limiting.
A: The Sanyo motors are rated for 5 volts. I know that many builders run these motors on 3-cell LiPolys (11.1 volts), but when you're frying as many of them as quickly as you are it means that your design loads them too heavily for that much voltage. Check the gearmotors after mounting to make sure there is no binding or interference with free rotation.
I'm also betting that you didn't break in these motors by running them in at a lower voltage before turning up the juice. Any brushed motor that you're going to overvolt should first be run-in under no load at about half its rated voltage for at least ten minutes to properly seat the brushes and bushings.
A: Mark J. here: you're in luck -- the Bosch GBA 750 from 'Run Amok' is sitting on my workbench and I've wanted an excuse to open it up to see how badly it was cooked at Robotica. I put a micrometer on a couple of the wires leading to the commutator: 0.036". According to a handy wire gauge chart that's AWG 19.
A: No, the Team Delta DeWalt mount fits only the old style 18V gearbox.
A: Mark J. here: No need for separate controllers for the turret motors -- this is likely just a set-up incompatability between the controller and the transmitter.
Check the Sabertooth 2x10 option switches first:
Still no joy? Write back and let me know. There are more things we can try.
A: Mark J. here: generally, no. Speed controllers for hobby brushless motors monitor current flow to 'sense' the rotor position and determine when power must be switched to control rotation. Unless multiple motors connected to one ESC are mechanically locked in perfect synchronization the contoller would receive confusing information on rotor position and motor operation would be disrupted.
Electronic controllers for brushed DC motors do no position monitoring -- the commutator takes care of that -- so you can run as many motors off of one brushed ESC as you like, provided that you do not exceed the rated amperage of the controller.
A: A pair of 12 volt drill motors are enough to power a simple featherweight. I'm not certain of the RPM available from your specific Black and Decker gearbox (there are several versions), but a typical drill gearmotor supplies about 500 RPM. With six inch wheels that will give a top speed close to 9 MPH and plenty of torque. That's a good start.
A second question relating to the first: the B16 is a very popular 3lb motor. Faster than Copals, but less torque at more than twice the weight. You frequently advise to "look at what works for others", so no arguing with B16's success. Can you help me understand what makes the B16 superior to the Copal in the 3lb class?
You have great stuff here. Thanks.
A: Copal makes relatively small runs of motors to custom order. Most of what the robot community sees are production run leftovers, and the specifications are subject to change.
The Copal SH50 has the same motor and gears as the standard 6 volt Copal 50, but the construction of the bearing support/mounting plate is different. This adds a little weight and changes the mounting hole spacing. It is not a direct replacement and may not fit a chassis designed for the standard Copal. Hence, it is being sold at a discount.
As for comparing the Copal to the B16: the difference is strength. The Copal gearbox is adequate for the forces encountered in antweight competition, but the stresses of mass and weaponry multiplied by three is too much to ask out of their lightweight construction. The B16 gearbox is a planetary design in a fully enclosed metal housing. The bearing support is much better and the gears themselves are much more robust.
A: If there was an exact replacement for the RS-385 that was 40% lighter, why would anyone use the RS-385? No such motor -- look for your weight savings somewhere else.
The closest I can come is the brushless Axi 2204/54 Gold Line. It has the same mounting hole spacing as the RS-385 (16 mm), is about the same diameter, has a little less RPM, a bunch more torque, and weighs in at less than one ounce. The shaft is larger, and you'll need a brushless motor controller. Retail price is around $75, versus about $2 for the RS-385. It would work as a weapon motor, but I wouldn't use it for propulsion: if you bog it down it'll melt.
A: Terrible photo, but I'm pretty sure they're Mini-EV Warriors. You can buy them at Robot Marketplace [no longer available]. The spur-gear reduction drive units shown are custom.
A: Mark J. here: climbing a ribbon is a little different than combat robotics, but I think I can point you in the right direction.
If you exceed the maximum surge rating of the controller it will likely destroy the device. Exceeding the 30 second or continuous ratings will (for the RoboteQ and most other controllers) cause the controller to cut back the amperage output and reduce motor power.
The A28-150 Magmotor motor, like all permanent magnet direct current motors, consumes amperage in direct linear proportion to the torque it is producing. At zero (free running) torque, the motor draws only 3.5 amps. At maximum efficiency the motor produces one horsepower and draws 37 amps. At peak output the motor produces three horsepower and draws about 185 amps. At maximum (stall) torque, the A28-150 motor can draw close to 375 amps.
Once you have examined the torque demands your design will place on the motor, you can match up the amperage demands to a suitable controller. It does no particular harm to go 'overkill' here -- better to have a controller with too much capacity than too little.
A: Not a great idea. Commonly available stepper motors are a type of brushless motor designed to rotate in precise increments to position something accurately -- like the read/write head on a disc drive. They require special driver circuits and are not designed for high power output.
When brushless motors were first being used in model airplanes, some hobbyists found ways to convert specific stepper motors for aircraft applications. Now that brushless motors designed for the purpose are widely available there is no need to go thru the trouble. I'd suggest sticking with more conventional motors.
A: I don't have a weight chart for drill motors with attached gearboxes, but I'm pretty sure they're all over 8 ounces. Robot Marketplace has a wide range of gearmotors with weights specified in their descriptions.
A: The Speed 400 motor has different mounting hole spacing than the Whayachi gearbox, and the gear reduction is not correct for the Speed 400. The BaneBots 28mm gearmotors are 1.1" wide and high by 2.8 inches long.
A: The RS-385 motors are identical -- the difference is in the gearbox. The heavier 36mm gearbox has a larger shaft and gears. It can take higher loads and more punishment from heavier robots or designs that place more stress on the gearbox. This comes with a weight penalty: the 36mm geabox weighs more than twice as much as the 28mm gearbox.
A: The planetary gear PGHM-09 and PGHM-14 motors are no longer in production and are no longer available. Lynxmotion does still supply the spur gear GHM-16 and GHM-12 gearmotors with similar performance ratings.
A: I wouldn't. If you want to run the motors from a 12 volt source, you can do so safely by turning back the transmitter 'Throttle Volume' setting for the channels that control the ESC. See our Transmitter Programming Guide for more information.
A: Not sure what the longest shaft I ever saw was, but it was a whole lot less than 18 inches!
It would help if I knew what use you had for this motor. Having to guess at the application: you're going to need a suitable motor, a miniature shaft coupling, a length of shaft material, and at least one pillow block to support the long shaft. Do a web search for 'miniature shaft coupling' and 'precision shaft', or seach an industrial supply site like McMaster Carr. Robot Marketplace has miniature pillow blocks.
You might also be able to find suitable parts at a well-stocked R/C boat hobby shop. They have speciality parts to connect motors to long propeller shafts that might work for your application.
A: The Jetis are made in the Czech Republic and the translation of the manual to English is not so good. Yes, Jeti controllers auto detect the number of LiPo cells. Here is the set-up drill:
A: Combat robots don't do anything very slow and smooth, Mike. Abrupt and violent is more like it.
Go have a look at the motor selection at American Science & Surplus. At the moment they have a small "Tough Gear Motor" that runs on 6 to 12 volts DC motor and spins 3 to 7 RPM. Should be just what you need.
A: Mark J. here: Don't confuse the relationship between 'power' and 'torque'; power is the product of torque and speed. Sanyos have less torque than Copals, but spin more than twice as fast.
There is also some confusion over the true output power of the Copals. The published performance figures are unclear about the voltage at which the measurements were made and the RPM numbers don't add up correctly.
My best guess for specs at 12 volts:
A: No. The 'flip channel' available on some speed controllers is an input from a spare receiver channel that changes the on-board channel mixing to correct the directional response of a robot when the 'bot gets flipped upside down (inverted). There is no power output corresponding to the flip channel input, so no possibility of running a weapon from it.
A: I'm guessing that you planned to use the 42mm BaneBots gearmotors. That's not a bad choice, but kinda overkill for a BBIQ Mini. A pair of the Banebots 36mm 16:1 RS-540 gearmotors will provide good performance. With 3" wheels at 14.4 volts the motors will give a top speed near 11 MPH, reach top speed in 21 feet, and will break traction and spin the wheels at less than 11 amps. Weight is 11.6 ounces. Don't worry -- they'll run fine at 14.4 volts. Just make sure you run in the motors unloaded for several minutes at reduced voltage to properly seat the brushes before putting them under load at full voltage.
If you want more power you can replace the RS-540 with the RS-550 motor. It bolts right up to the 36mm gearbox and will give more speed and power, but I really think the RS-540 is plenty for a 15 pound robot.
If you're looking more 'heavy duty', the Piranha 36:1 Gearmotor [no longer available] could work for you. The gearbox is very well constructed and has a 1/2" output shaft. With 6" wheels the top speed and max amp draw are comparable to the BaneBots RS-540 gearmotor but acceleration is better, reaching top speed in 14 feet. Weight is 1.5 pounds, price is about $94, and they are in stock.
A: Tough one, Mike. The motor could be almost anything. I don't recognize that number, but many DeWalt motors have six-digit numbers stamped on the case, and the size is about right. Browse thru the pictures and see if anything matches up.
Q: The speed controller is small and is for the insect classes. It has two receiver plugs and a black screw down terminal on the side. The circuit board is green and has two mounting holes. There is a black plastic thing sticking up on the board and another cylinder shaped thing that is half silver, half black. The letters and numbers on the cylinder say: 47 VFC 3T1. It looks like the flip channel plug was removed. Merci.
A: Yea, that's the Barello Ant 100. The color/style of the screw connectors changed with availability. The Barello speed controllers are no longer in production and Inertia Labs now sells them only with their full antweight kit, but you may be able to find a few in stock somewhere.
A: Combat robots of any weightclass need some method of controlling the direction and power of the motors. It's possible to use relays with R/C interfaces to get forward/off/reverse motor operation, but for reliability and precision driving control you need speed controllers.
A: My first guess is that you didn't bother to read the Scorpion XL manual. You just wired it up and turned it on, didn't you? If you can't find the time to read the entire manual, at least read the Scorpion XL quick start guide with particular attention to the radio calibration step.
A: RobotShop was the last source I had -- they're gone now.
A: How far a motor can be overvolted depends in part on how much load is placed on it. If you set up the wheel diameter so the motor can spin the wheel before it bogs down below about half the free RPM of the motor, you will get much better results than if you let it get close to stall and start to pull high amp loads. The
Team Tentacle Torque & Amp-Hour Calculator
can help you select wheel sizes that will keep the amp load reasonable.
That said, I would be comfortable running the ML-50 around 18 volts -- but don't stall it!
A: Mark J. here: Jeti ESCs are made in the Czech Republic, and the manuals are a somewhat difficult translation. Assuming that you have the Jeti Spin 99 Opto, there are two options in changing the programming of the controller:
A: For drive power, two Magmotor E-150s will cost about $160 and come without gearboxes. You'll have to fabricate your own gear reduction and drivetrain. Overvolted to 36 volts (not recommended for drive power) two will deliver 2.25 horsepower - overkill for a big-weapon middleweight. Top speed with 6:1 gearing and 3" wheels is about 10 MPH in 2.2 seconds. The motors would weigh 7.8 pounds without the drivetrain.
Four BaneBots 42mm 27:1 RS550 gearmotors will cost about $200 and come ready-to-mount with gearboxes -- no additional expense or effort required to fabricate a drivetrain. At 14.4 volts (maximum recommended) four will deliver about 1.5 HP. Top speed with 3" wheels is about 8 MPH in 2.3 seconds. The motors would weigh 4.5 pounds including the drivetrain.
Going with the Magmotor motors is false economy. You'll spend a lot more money to get your drivetrain assembled than the difference in cost. You'd also add a lot of weight. I'd keep it simple and run the BaneBots gearmotors.
A: Massive horsepower overkill. They would be suitable for a middleweight rammer or wedge, giving a top speed near 20 MPH. With a large weapon, that much speed would make the robot as dangerous to itself as to the opponent.
A: Speed controller selection depends on more than the motor(s) you plan to control -- it also depends on how you plan to use them. If your application runs motors heavilly loaded at or near stall they will require a controller that can either source the enourmous amperage draw they will require or a controller that can limit amperage to protect itself and the motors under those circumstances.
The
Team Tentacle Torque & Amp-Hour Calculator
can help you figure out how many peak amps your motor set-up will require in a specific robot. Use that for a guide in selecting a controller.
A: Mark J. here: if you want to 'push the limits', you should first know what the limits are. Acceleration and pushing force are both limited by the weight pushing down on the wheels. When you add more power than needed to break traction and spin the wheels at peak horsepower, additional benefits are very small. You can gear up to trade that excess torque for greater speed, but controlling a lot of speed in a small arena can become a real handfull.
Old-school lightweight 'Dr. Inferno Junior' used four 18 volt DeWalt drill motors at 24 volts. That's 6 horsepower in a 60 pound 'bot. Run that set-up thru the
Team Tentacle Torque & Amp-Hour Calculator
with 4" wheels. You'll see that the motors are wildly under-utilized; the wheels break traction and spin at less than 1/6th the stall torque of the motors.
I'd say that there's no good reason to push lightweight power limit any further than it's already been pushed, but people seem to enjoy ignoring my advice. If you're determined to demonstrate real insanity, how about 4 of the 36 volt DeWalt hammerdrill motors? That would set a 'high water' mark that would stand for some time.
Q: How about three 16:1 RS775 BaneBots 42mm gearmotors per side with 4 inch wheels at 16.8 volts?
A: Six RS775s at 16.8 volts would give you just about the same power as four 18 volt DeWalts at 24 volts. The difference in gearing would make the BaneBots setup considerably slower (15 MPH vs. 22 MPH). It would certainly be powerful, but not uniquely so.
A: The number of wheels is not an issue -- the weight of the 'bot and the expected performance is what matters. A second or third wheel can be driven by chain or belt from the hub of the wheel directly attached to the gearmotor.
A pair of DeWalt 14.4v gearmotors can provide good performance in hobbyweight or featherweight combat robots no matter how many wheels you choose to power with them. You can check the performance of all popular combat robot motors with a variety of gear, voltage, and wheel combinations at the
Team Tentacle Torque & Amp-Hour Calculator.
A: Mark J. here: different ESCs handle overload in different manners. Some have current limiting circuitry that allows them to cut back loads exceeding their capacity. Others, like the Victor 883, can handle only very brief overloads before they fail.
If this is for a drive system, the
Team Tentacle Torque & Amp-Hour Calculator
will provide the expected amperage needed to spin the wheels based on a given drivetrain configuration. Keep that under the max amperage for your ESC and you should be OK unless you somehow lock up the drivetrain.
If this is for a weapon system, the calculation is more difficult. Two RS-550 motors in parallel would pull 170 amps at stall -- almost twice the 90 amp continuous rated capacity of the Victor 833 SC. You could feather the throttle at the start of the spin-up, but you really don't want to be doing that in combat.
My advice: don't scrimp on your speed controllers. It's less expensive to buy what you actually need than to keep replacing inadequate equipment.
A: The Axi model number is a simple code:
See comments on calculating the output of Axi motors in this section of the archive.
A: Depends on what you mean by 'safe'. Any overvolted motor will have a reduced lifespan, and that reduction increases very rapidly with voltage increase and with the amount of time spent pushing near stall. If you gear the RS-540 to keep it from bogging down you'll be alright at 14.4 volts. I'd keep a couple of replacement motors in my spares box, but you're gonna do that anyhow, right?
Alternately, you could upgrade to the RS-550 motor. It bolts up to the BaneBots gearbox and will run all day at 14.4 volts without trouble; I've seen them run at 18 volts in combat. You'll get increased torque and reduced stall amperage, but you'll loose some RPM and gain a few ounces of weight. I'd go with the RS-550 if you're building a speedy 'bot that could bog down when pushing, and I'd stick with the RS-540 geared down well for a pushy-bot.
A: The published performance specs are the same: speed, torque, max current, power. However, it is heavier by more than 7 pounds (28 versus 20.8 lbs.) so you could say it's not as good as the original.
A: I'm gonna have to guess at how long you've run the motors, what voltage you run them at, and what sort of abuse you've put them thru.
I suspect that you didn't break in the motor brushes properly before subjecting the motors to heavy load. New brushed motors should be 'run in' at about 1/4 speed for at least 20 minutes before turning up the voltage and the load. This allows the graphite motor brushes to contour to the radius of the commutator and maximize the contact area. Failure to do this can cause the brushes and commutator to overheat, deform, crack, or rapidly wear.
Remove the motor brushes and inspect them for signs of overheating or wear. Make sure they move freely in and out of the brush holder assembly without sticking. Minor defects can be corrected with a small file. Also inspect the commutator. A drop of WD-40 on a Q-tip can be used to clean the commutator. It may appear blue if it has been badly overheated.
Reassemble the motors and run them in for several minutes to re-seat the brushes. If this doesn't cure the problem, you've cooked the motors and they should be replaced. Many successful robots have run Harbor Freight drill gearmotors -- just make sure to break them in properly and they should work well for you.
A: No. Maximum motor current is the allowable current draw for a specific time interval. The Etec-R can be loaded down to draw 320 amps for 2 minutes before something ugly happens, or loaded to 150 amps for as long as you like. I'd estimate stall current to be over 800 amps at 48 volts.
A: Electric lifters require a lot of power and a very sturdy gearbox. I believe 'Wipe Out' uses a Magmotor power it's lifter -- possibly the A28-150.
A: It appears to be a good old fashioned Small Johnson motor. Proves that you don't have to run expensive brushless motors to be competitive.
A: The motor comes with two screws that fit holes in the gearbox faceplate. Assuming that you have a flat plate chassis, you can make a wide metal L-bracket that bends back under the gearbox and fastens down securely. Drill holes in the L-bracket to match those in the faceplate and fasten with the provided screws.
The far end of the motor should also be secured to the chassis. If the high-stress gearbox end is well mounted, a zip-tie thru the chassis and around the motor end will do to secure the other end in an insect-class 'bot.
A: I don't have a source of replacement parts for the Sanyo Micro gearmotor. You could spend a lot of time and effort trying to find a suitable gear -- I'd replace it and save the stripped motor for parts.
A: No. Their gearmotors are all very small, low-output units. Also, most are geared way too low to provide enough speed for a combat robot. You could use their GM18 mini metal gearmotors in a 150 gram fairyweight, but the peak power output is only about 1 watt.
Now, I have seen Solarbotics gearmotors used in autonomous ants and ants with really huge weapons that just need to be pushed around slowly. There are 'special circumstances' were you can get away with very low drive power, but most ants and beetles will require much more power to be competitive.
A: Mark J. here: you've been browsing British websites, haven't you? Petrol is what they call gasoline in the U.K. -- short for 'petrolium spirits'. A 'petrol engine' is your standard internal combustion engine (ICE).
A: Mark J. here: a thwackbot needs enormous spinning speed and lightning acceleration to be effective, and that translates into power. This is the one application where I'd go for total motor overkill and reach for the big gun: DeWalt. A pair of 14.4 volt DeWalt drill motors and gearboxes locked into high range and overjuiced to 18 volts would give you a screaming blur of a thwacker. Yes, you can use the 18 volt DeWalts pushed to 24 volts if you really want to, but there's overkill and then there's total madness.
You're correct that NERC and Robot Battles are not RFL affiliates, but they may very well use a similar 'Damage / Aggression' scoring that works against thwacks. I'd check on this before building.
A: Dead Metal, like all the early HouseBots, was powered by Bosch GPA 750 motors: 97 inch pounds of stall torque at 24 volts. The motors were geared way down; none of the early HouseBots were very fast.
I never saw 'Mr. Psycho' in person. I'm told that it was propelled by twin 5 HP Briggs and Stratton motors at 48 volts, but I don't know which specific motors those are.
A: Mark J. here: the Magmotor company makes dozens of different motors for a variety of purposes with a wide range of output power. The 21-230 series you found on EBay is a low-output 2.7 pound industrial motor that comes in several different windings (not specified on EBay) for use from 12 to 120 volts. Peak output is about 0.4 HP with perhaps 350 oz-in torque. The motor is too heavy for its output to be of much use on a combat robot.
You have to know what you're buying on EBay. Descriptions are often incomplete and sellers may not know much about what they're selling. Stick with motors you know.
A: A normal DC electric motor has a rotating armature of copper wire wound around an iron or steel 'core'. The core supports the copper windings and helps to direct the magnetic field, but it's heavy and adds to the bulk of the motor. The actual force generated to turn the motor is exerted against the copper windings themselves so, although convenient in manufacturing, the core is not required.
By embeding the windings in epoxy resin the steel core can be eliminated, resulting in a motor that is lighter and faster to respond. The down side is extra expense in manufacturing and less sustained power handling capacity due to the loss of the heat sinking and radiating effect of the armature core.
A: Way too little information. Any gearmotor will 'work' on a 4-bar lifter, but I can't even guess at the performance of the resulting lifter since you gave me no clue about the 4-bar geometry or the weightclass of the 'bot.
As a general comment the motor is very heavy for the power it puts out, and the worm drive gearbox is built for windshield wipers -- not heavy lifting. There are better choices for a lifter motor.
A: The NPC-41250 is quite durable and easy to mount, but it is also fairly expensive ($150+) and very heavy (7.5 pounds) compared to gearmotors with similar performance. Worse, the gear reduction is too great for use with reasonably sized wheels: with 6" wheels at 24 volts it tops out at a hair over 3 M.P.H. -- way too slow!
If you want to stick with an NPC setup, look at the 'Build Your Own Gearbox Kit' at the bottom of the NPC-41250 page linked above. You get 14 M.P.H. for the about the same cost and weight.
There are plenty of popular alternative gearmotors. BaneBots offers a wide range of geaboxes that mate to a wide range of motors. Other good choices would be the popular cordless drill gearmotors: 18v DeWalt, Harbor Freight, etc.
You can 'test drive' a wide selection of motor/gearing/wheel/weight combinations at the
Team Tentacle Torque & Amp-Hour Calculator.
Q: What batteries should I use for a pair of NPC 'Build Your Own Gearbox Kits' in a lightwight?
A: Here are the steps to estimate needed battery capacity:
Q: Would I only need one BattlePack PN-3000-24 with a pair of NPC gearbox kits?
A: The 'Battery Chooser' estimates the power consumption of all the motors entered into the calculator. I made an error when I first did the battery pack calculations for a pair of NPC BYO Gearbox kits. I thought the gear reduction was 5.5:1, but it's actually 4:1. That changes things a little.
With the correct gear ratio, the 'Battery Chooser' shows that a single BattlePack 3000-24 would be fine for two NPC BYO Gearbox kits in a lightweight with 4" or 5" wheels. The performance difference between wheel sizes is small -- I'd go with the 4" wheels in an arena up to 24 feet across; 5" if you fight in a really big arena or just wanna brag about top speed.
With 6" wheels, the 'bot could completely drain a BattlePack 3000-24 at the end of a tough 3 minute match. You do not want to run out of juice at the end of a match! A single BattlePack 3600-24 would be a better choice for 6" wheels.
A: National Power Chair (NPC) is a company based in Minnesota that rebuilds motors and gearboxes for electric wheelchairs. Their gearmotors became very popular for combat robots back in the BattleBots days. All of their motors are brushed, as are other industrial permanent magnet DC motors built for economy, reliability, and prolonged high torque output capacity.
Brushless motors are found in electronic and computer applications where tightly controlled speed and compact dimensions are needed. The model airplane industry adopted brushless motors because of their very high power to weight ratio. High performance brushless motors are designed for high speed / low load operation and are not useable in combat robot drive systems where they would experience prolonged high loads at low speeds. They rapidly fail under such conditions. Robot combat use of brushless motors is restricted to weapons.
A: No, sorry. The NPC kit is simple, durable, and easy to mount but is less powerful than some other gearmotor options of similar weight. In my opinion, it's still ample power for a lightweight.
A: Depends on how much load you put on the Harbor Freight motor. For use as a hobbyweight drive motor with reasonably sized wheels, it should be fine up to its maximum rated 12 volts -- 14.4 volts might be pushing it. Controlling a big spinner weapon is likely not advisable. Note that the 'Super Rooster' does have a bothersome delay before it will switch from forward into reverse.
Hobby R/C ESCs are rated for operation of R/C cars -- not combat robots that spend a lot of time pushing. If the motor is stalled, your Rooster isn't going to live long. Don't trust the amp figures given by hobby R/C ESC manufacturers; they are purely theoretical and for instantaneous loadings only.
A. The Black Max has been used in some terrifying robots, but it is much heavier (15+ pounds) and less efficient than some more recently available motors. The A28-400 AmpFlow weighs less than half as much and produces more horsepower at a similar amp draw, but I'd go with the still smaller and lighter A28-150 for a middleweight wedge. That would be plenty.
A. The text description of the Piranha gearmotor at Robot Marketplace only says is that the gearbox is, "strong, lightweight and reliable". It's too new to have a combat record -- I don't know of anyone who is using it. Like any new product, this one is 'try at your own risk'.
What you're really buying here is the gearbox. For $93.99 it comes with an inexpensive 550 Johnson motor attached, but any 540 or 550 size motor would bolt right up. There are plenty of other 550-size gearmotors available, most of them quite a bit less expensive than the Piranha. It does have a big 1/2" shaft and beefy dual bearings in a tough-looking case. Mounting should be easy. The 36:1 gear ratio is a little to high for my taste -- with the Johnson motor @ 18 volts and 3" tires, it would only give 7 MPH.
A: ESC stands for Electronic Speed Controller.
A: Mark J. here: the EV Warrior was an electric bicycle built back in the 90's. The company went bankrupt and thousands of the Bosch-built motors hit the surplus market at very reasonable prices. The motors were compact, weighed around 3.25 pounds, and pumped out about 1.5 HP when overvolted to 24 volts. A lot of combat robots made good use of the motors.
Unfortunately, the last source of these surplus motors dried up in 2005. You might get lucky and find a few on EBay, but buy a couple of spares -- overvolted EV Warrior motors were not known for longevity.
Q: If the 'EV Warrior' motor was so popular, why where they discontinued?
A: They were popular because they were dirt cheap -- as little as $15. They were only dirt cheap because a warehouse full of them was dumped on the surplus market after the electric bicycle company went bankrupt. When Bosch was making them they cost ten times as much and were not a bargain for robot builders.
A: Brushed motors use a set of physical sliding contacts (brushes and commutator) to control the direction of current flow through the armature windings and keep the motor spinning, so all the ESC has to do is control the speed by limiting current flow.
Brushless motors have no sliding contacts and rely on an 'intelligent' motor controller to perform the duties of both the commutator and the speed controller. A brushless motor will not even run without a suitable motor controller.
See also a previous post on this topic in this section.
A: Too slow! The
Team Tentacle Torque & Amp-Hour Calculator
shows a top speed less than 3 MPH for that configuration. It'd be OK with 6" wheels. Better yet, use the 900 RPM drill motors with 6" wheels.
A: BaneBots is, if anything, a little conservative with their voltage ratings. I've seen the RS-550 motors run at 18 volts in combat, so the RS-775 should take the voltage nicely.
Plugging the numbers into the
Team Tentacle Torque & Amp-Hour Calculator
indicates that 4 of these gearmotors would give a wild ride to a featherweight 'bot. Top speed comes out to about 14 MPH, which would come in 1.3 seconds or 18 feet. The tires would spin at less than 11 amps per motor (127.5 amps is stall current). Might be overkill.
I was using the Hitec Digital Robot Servo with an additional 4:1 spur gear drive before -- remember that it got hot at stalling when trying to hold things? I have a choice of a 5:1, 10:1 or 20:1 worm reduction for the final stage of the new motor. With size and weight in mind, what gearmotor would you consider to be the best choice to replace my servo?
A: Take a look at the BaneBots RF-370 20:1 25mm Spur Gearmotor. Coupled to a 20:1 worm drive, it will give better torque and speed than your 4:1 geared Hitec servo at the same 7.4 volts.
If you're willing to run at a lower voltage the BaneBots FF-180 30:1 25mm Spur Gearmotor is lighter and at it's maximum rating of 4.5 volts with the 20:1 worm drive would provide torque and speed similar to the 4:1 geared Hitec servo at 7.4 volts.
Q: Can the BaneBots FF-180 Motor be overvolted to 7.4 volts? Specs says that its operating voltage is from 3 to 4.5 volts.
A: The Mabuchi FF-180SH-2860 sold by BaneBots is a metal-brushed motor. Metal-brushed motors don't generally put up overvolting as well as carbon-brushed motors, particularly in applications where it will be even briefly stalled. I'd advise against trying to run it at 7.4 volts in your application.
Q: If I choose the 4.5V Banebot gearmotor for my gripper, what would be your suggestion for dropping the 7.4 volts from my LiPo battery to 4.5 volts? I tried to use a solid-state voltage regulator in a previous project, but I wasn't happy with the results. Should I try to regulate the voltage or get a different battery for the gripper?
A: Mark J. here: I don't know how you're controlling your gripper motor, and that makes a difference. The problem with mildly overvolting DC motors is the extra amperage that comes with the increased voltage -- not the extra voltage itself. Amperage creates heat and increased heat will destroy the metal brushes in the BaneBots FF-180 motor.
If you're using an R/C system with an Electronic Speed Controller (ESC) to control the BaneBots gearmotor you can limit the amperage to the motor either by setting the ESC to deliver less than full power, or by setting the Adjustable Throttle Volume (ATV) on your R/C transmitter to prevent 'full throttle'.
If the motor will be controlled by simple switching, adding additional resistance to the motor circuit will restrict the maximum amperage and will keep the brushes intact. How much extra resistance is needed to hold the stall amps constant at the new voltage?
The FF-180 motor may be run reliably at 7.4 volts if you add ( 7.4 volts - 4.5 volts ) ÷ 6.7 amps = 0.43 ohm additional resistance. You'll have the same torque that the motor would deliver at 4.5 volts and gain greater speed due to the increased voltage.
How do you add 0.43 ohm that will carry 6.7 amps? AWG 22 gauge wire will carry 7 amps of current and has a resistance of 16 ohms per 1000 feet. A loose coil of 27 feet of 22 gauge wire will not overheat and works out to 0.43 ohm.
Plugging in values for the Axi I get 2194 watts. Thats comparable to a Mini-Magmotor motor and only 1/6th the weight! Can that be right?
A: Mark J. here: Axi and other brushless hobby motors do put out a huge amount of power for their weight. They were designed for model airplanes that can't afford extra mass. They were also designed to spin along at high RPM and high efficiency and to never be bogged down into the rev range where they produce peak power. Their small mass means that they will fail very quickly from thermal overload if they are required to operate in a high-torque range for more than a very small percentage of their run cycle. We can just barely get away with using them for spinner motors.
Magmotor motors are designed for combat robot propulsion. Their extra size and mass enables them to perform for extended periods at near-stall speeds and in the max power range of their torque curve. All that extra mass is wasted if the motor spends most of it's time at low power output keeping a spinner weapon up to speed. Conversely, an Axi would fail quickly if used as a middleweight propulsion motor.
Also remember that my formula for estimating brushless motor output is theoretical. All sorts of things will work to decrease the actual performance in use: additional circuit resistance caused by the motor controller, inability of the battery to provide full voltage at large amp drain levels, power loss from the gear/belt drive, etc.
A: Mark J. here: they are not as different as you think. The shape of the torque, amperage, and power curves for brushed and brushless motors used in combat robotics are effectively identical. Both brushed and brushless are Permanent Magnet Direct Current (PMDC) motors -- the only substantial difference is that brushed motors use a set of physical sliding contacts to control the direction of current flow through the windings, while brushless motors use an electronic controller box for that purpose. The elimination of the sliding contacts reduces arcing and friction, making brushless motors more efficient.
The Team Run Amok Spinning Weapon Excel Spreadsheet works equally well for brushed and brushless motors.
A: How many volts are you willing to run? Up to 12 volts, I'd run the 2826/8. For 4 or 5 cell LiPolys, switch to the /10.
Q: So in order to get the most power out of a Axi gold 2826, I should use maximum number of batteries allowed with the highest number 2826? For example, the most powerful Axi 2826 combination would be 2826/12 with 18 NiCd cells?
A: Mark J. here: NO -- stop looking for short cuts and do the math.
On paper the 2826/10 (0.042 ohm) on either 16 NiCads (19.2 volts) or 5 LiPolys (18.5 volts) puts out more power than the 2826/12 (0.062 ohm) on 18 NiCads (21.6 volts). The number after the slash on the Axi model number refers to the number of turns of wire around each armature pole. The sequence for the Axi 2820 series just happened to work out with a larger number of wire turns providing greater power at max recommended voltage.
Q: How are you finding the terminal resistance of the different axi's? I could do the math if I knew how to find that.
A: Most sites that sell brushless motors provide the 'internal resistance' (same as 'terminal resistance') for each motor in the specifications. I get my specs straight from the Axi website.
A: Mark J. here: from other equations I've given here, adjusted for the specs usually provided for them:
Q: Say a hobbyweight has a spinning bar weapon with 3500 joules of energy. That means that the bar could be spun up by an Axi 2820 in about 3 seconds, since 1 joule is equal to 1 watt second. So once it spun up, how long could it spinning before it melted?
A: Your calculation of spin-up time is over-simplified. The output of a permanent magnet DC electric motor operating at constant voltage is dependent on the RPM of the motor, with peak output at 1/2 of the no-load RPM. The power consumption, torque, and total power produced by the motor all change with RPM as the weapon system spins up. A full analysis of weapon spin-up time is provided by our Team Run Amok Spinning Weapon Excel Spreadsheet.
Amperage creates heat. Spinning up the weapon requires high motor torque output and coresponding high amperage draw. Once the motor is up to a high-speed spin, the amperage draw is greatly reduced as very little power is required to maintain the weapon speed -- just enough to offset frictional drag. In this situation a properly geared Axi 2820 would spin along happily until the battery goes flat.
The problem of meltdown becomes a worry if the motor is used for robot propulsion and is used for extended periods in high torque/amperage pushing, acceleration, or rapid reversing.
A: Batteries produce Direct Current -- use a DC motor.
A: Like most R/C car hobby motors, very little performance information is provided for the Mamba Max. From what is given there isn't any way to tell how much power it puts out. It does claim to spin 82K RPM at 14.4 volts, which could be difficult to gear down to useable weapon speed.
The Mamba Max is an 'inrunner' style brushless motor. Outrunner 'rotating can' style brushless motors are much more popular for weapon motors because they produce much greater torque at lower RPM. Stick with a proven motor.
Q: Which Feigao motors are best for hobbyweight spinning weapons?
A: I haven't seen the Chinese-made brushless Feigao motors used in robot combat, so I can't comment on their suitability. The limited information they give for their 540Y-15/7 outrunner motor looks about right for a hobbyweight weapon, but why not stick with one of the proven brushless motors?
Q: Alright, are there any hobbyweight weapon motors you can recommend?
A: It would really help if I knew something about your weapon other than it goes on a hobbyweight. Full Body spinners, drums, bars, and disks of varying sizes all have different power requirements. We've said all this before on the site:
A: The
Team Tentacle Torque & Amp-Hour Calculator
has listings for 83 different combat robot motors, including the Harbor Freight 9.6, 12, 14.4, and 18 volt drill motor/gearbox combinations.
A: Mark J. here: I estimate the peak power output of an Axi 2814/10 at a little over 830 watts. A pair of those in a hobbyweight would give more than 140 watts of drive power per pound of robot -- about eight times what you could effectively use in a small arena. The extra power will just be lost in wheelspin, and keeping it pointed in a straight line would be a real challenge.
You could go with a smaller Axi, but the motors have tight limits on their maximum sustained amp draw: they melt if pushed too hard. An Axi makes for a great weapon motor, but they aren't ideal as drive motors.
A: Mark J. here: the primary market for brushless hobby motors are model aircraft, and model aircraft builders don't care about the specs we use -- model aircraft don't stall their motors! You can find formulas and definitions for all the motor specs aircraft guys use at the bottom of the Aveox technical page.
The Team Run Amok Spinning Weapon Excel Spreadsheet needs only two key motor specs: stall torque and no-load RPM. Brushless motors usually have an RPM per volt specification that lets you calculate the no-load RPM at your voltage. The stall torque can be calculated if you can find the 'terminal resistance' for the motor:
If you can't get the required info for the motor you're interested in, I have a couple of low-tech solutions:
A: Would you believe it doesn't make any difference? You've got three black wires with male connectors from the motor and three black wires with female connectors from the controller -- plug 'em in any way you like. If the motor spins backwards, switch any two of the connectors to reverse the spin direction.
A: Your RS-540 motor is fine, but you picked the wrong gearbox. Try re-running the calculations with the 16:1 gearbox, and leave the 'Gear Ratio' box on the
Team Tentacle Torque & Amp-Hour Calculator
at 1:1. That will give you better than 9 MPH and lightning acceleration with 3" wheels -- perfect for a small arena.
Q: If I decided to build a pushy bot/rammer instead of a wedgebot, would the RS-540 16:1s still work well?
A: See the previous article on propulsion power requirements. The RS-540s will give you plenty of power for anything you might want to do with a hobbyweight.
A: If you wire two identical motors in parallel to a single channel of an Electronic Speed Controller, the potential total amp draw from the ESC will double compared to a single motor. Each motor will have 100% of normal power output.
If you wire two identical motors in series to a single channel of an Electronic Speed Controller, the potential total amp draw from the ESC will be cut in half compared to a single motor. Each motor will have only 25% of normal power output.
See a related post on multiple motors.
A: Your dad treating the motor like a Buffalo Wing isn't the problem. Take a minute and read thru the instruction sheet for the ESC. If the battery voltage is too low, the BaneBots GWS ESC will not start-up and will make a continuous 'dong-beep' sound. Charge up your LiPoly (the ESC only works with LiPolys) and you'll be fine.
A: Mark J. here: the motors you will need depend on the style of 'bot you're planning to build. A 'bot with a massive weapon that just needs to push itself around the arena will need a lot less drive power than a ramming brick or wedge. It comes down to matching the power-to-weight ratio of your 'bot to the type of attack strategy you're planning.
Some maniac builders pack as much as 40 watts per pound into their 'bots, but there really are limits to how much power can be used in a small arena. I've built three champion robots on the 4 watts per pound formula, so don't tell me it can't be done.
The BaneBots 36mm planetary gearmotors are popular choices in the 12 pound hobbyweight class. The output shaft is a beefy 3/8" diameter. Three different motors are available with the 36mm gearbox: RS-385, RS-540, and RS-545. BaneBots does not provide output power numbers for these motors, but they do provide stall amps and voltage. You can get a rough estimate of output power with the following formula:
The formula gives the following outputs:
From the 'rule of thumb', a 12 pound wedge robot would require about 16 watts per pound × 12 pounds = 192 watts of power, so two of the RS-540 or RS-545 motors coupled to the 36mm gearboxes would give you plenty of push.
Which gear ratio will you need? See the Ask Aaron Optimum Gearing page.
For heavier robots, BaneBots offers a 42mm gearbox with a 1/2" output shaft attached to the RS-550 motor. The RS-550 puts out approximately 250 watts, so two of these would be adequate for a 30-pound featherweight. More power? It is possible to mount the RS-775 motor (370 watts) to the 42mm gearbox.
Want still more power? Use more gearmotors! Four RS-550s would be enough for a 60-pound lightweight wedge, and six RS-775s would power a 120-pound middleweight rammer.
A: Mark J. here: at stall, a Permanent Magnet Direct Current (PMDC) motor produces maximum torque and will consume:
A: How much drive power a combat robot needs depends on the design and attack strategy of the 'bot. A wedge or rammer will need a lot more drive power than a full-body spinner. Several top-ranked middleweight wedges run a pair of A28-150 "mini" Magmotor motors -- that's 6 horsepower worth of go-power! A middleweight 'bot with a large active weapon could get by with as little as 500 watts of drive power, but common practice calls for more like 1500 watts to keep the 'bot in the fight. One horsepower equals approximately 750 watts.
Wheel diameter isn't a factor in motor selection. For larger wheels, just increase the gear reduction to keep the torque high enough to avoid stalling the drivetrain. See the Ask Aaron Optimum Gearing page.
A: Mark J. here: Reliable robot motor speed controllers are expensive, but at around $260 the Robowars IBC controller isn't more expensive than comparable dual-channel controllers like the Vantec RDFR23, or Robot Solutions RS80D. Don't confuse the power ratings of robot controllers with the advertised ratings for R/C car controllers. The R/C car controller ratings are entirely hypothetical -- try to pull the full rated amperage thru an R/C car controller and you'll very quickly get an expensive puff of smoke. Remember: the most expensive speed controllers are the ones you overload and blow up!
The Robowars IBC controller is popular for hobbyweight and featherweight robots, but with a 50 amp-per-channel power rating I wouldn't try to build a middleweight around it.
A: Mark J. here: I don't think 'most popular' is a good way to select a weapon controller. The controller should be matched to the specific motor and weapon loading for your 'bot, not just to it's weight class. I've seen a very wide range of weapon motors and designs in hobbyweights. If you can tell me more about the weapon you plan to build, I can be of more help.
I'm planning to use two Mini EVs or HTIs to power a vertical disk weapon.
A full analysis of your weapon system power requirements and energy potential also requires the weight and diameter of your disk, and the speed reduction of your drive system. Assuming an 8" disk weighing 3 pounds and a 3:1 reduction, your twin-mini-EV powered weapon running at 12 volts would spin up to 3780 RPM in about 1.6 seconds with 530 Joules of energy, and will eventually reach over 1000 Joules. Killer for a hobbyweight!
If you have a no-slip drive system for the weapon (chain, gears) and switch power full-on, those twin Mini EVs will pull an average 117 amps during that 1.6 seconds, with a peak inrush of almost 170 amps. That's killer too, but not in a good way.
The easy solution is to use a hefty battery pack, a Victor 883 controller, and a belt drive. You'll be able to feed the power in quickly enough to keep a great spin-up time while avoiding that nasty sudden amp spike you'd get with a solenoid/contactor. A belt drive set-up with a little slippage will help keep you from melting down when the weapon stalls. The 833 is reversible to allow you to clear the weapon if it jams.
There are other options:
A: Mark J. here: mechanical switching control of brushed weapon motors has a number of drawbacks. Solenoid operated switches are called 'contactors', and a reliable contactor isn't much less expensive than a weapon ESC. If you get a cheap one, the contacts can weld shut under the high current loading encountered when suddenly dumping full voltage to a loaded weapon motor. That high current draw is also hard on your battery pack and may cause a voltage drop that can glitch other 'bot systems. A smooth transfer of current available with an ESC avoids these problems.
A contactor requires an R/C interface to trigger, like the Team Delta solid-state R/C 'D' switch. The R/C switch plugs into your receiver and controls a small current load to activate the coil of the contactor. The contactor then switches the high current load to the weapon motor. You can control as many weapon motors as you like with a single contactor -- as long as the total current draw of the motors does not exceed the capacity of the contactor. This also applies to ESCs.
Oh, and did I mention that contactors are heavy? They are not commonly used in high power sub-light 'bots. I'll say it again, 'cheap' and 'robot combat' don't mix.
A: It's not easy to compare performance figures for the brushless outrunner motors -- manufacturers often don't supply the type of numbers you really need for comparison. Also, there are different versions of the AXI 2208 (2208/34, 2208/26, 2208/20) that have very different performance figures.
Browse around the Aeromicro website. They have what specs are available for AXI, Park, Himax and other outrunner motors. The E-flite Park 370 Outrunner has seen good use in insect class 'bots and would be a good place to start.
A: Not a toggle switch -- some very early robots used an R/C servo to close microswitch contacts to turn motors on and off. It was unreliable, and suitable only for very slow robots. Also, some early R/C dune buggies used a servo to rotate a mechanical speed controller based on power resistors - both heavy and unreliable. If you really need cheap, you'll be better off to hack a radio/controller module out of an R/C toy or pull the low-power speed controller out of a low-end servo.
A: There are lots of motors you can use in your robot. You can find pictures, descriptions, and specifications for dozens of popular motors at: Robot Marketplace.
A: Mark J. here: check #16 in the
Speed controllers are the heart of your 'bot. They are key to the function and usability of the entire machine, and if they fail you're toast. To be useable in your 'bot, speed controllers must:
I don't recommend scrimping on robot electronics -- you can bet your opponent hasn't.
A: Inexpensive cordless drills are a very popular source of compact and powerful gearmotors. If you really want to 'salvage' something, permanent magnet DC gearmotors are used in automobile electric seats, windows, and windshield wipers. Your local auto salvage yard can be a good source.
Note that the power and output speed for gearmotors varies widely. You may have some difficulty finding something that meets your needs, so be ready to improvise!
A: Thwackbots are not popular in the insect weight classes. They need time and a clear area to spin up that just doesn't exist in a small insect class arena. If you really wanna try a beetleweight thwackbot, the B16 High Speed Gearmotor might be worthwhile.
A: Mark J. here: to calculate the spin-up time and energy for a specific rotary weapon system you need to know the stall torque and top speed of your electric motor, the shape and dimensions of the spinning weapon, the type of material you'll use to make the weapon, and the gear reduction between the motor and the weapon. With that information, you can use the Team Run Amok Spinning Weapon Spreadsheet to get the performance specs of the weapon.
It may be difficult or impossible to find accurate torque information for small hobby motors like the Park 730. If you run into this, my suggestion is to learn from the experience of other builders. Find a robot with a weapon set-up similar to the one you wish to build and see if it performs well. Based on what I've seen, a Park 370 outrunner should have no trouble spinning up a 6 ounce drum of reasonable proportions. I'd use a belt drive with a 2:1 or 3:1 reduction, depending on the diameter of the drum.
A: Bosch makes several motors used in combat robotics:
Q: How much Horsepower does a Bosch 750 motor produce at 24 volts?
A: Mark J. here: There's a simple formula to calculate the horsepower of a DC permanent magnet motor at different voltages:
The Bosch 750 makes 3.375 HP at 36 volts, so horsepower at 24 volts = (24 ÷ 36)2 X 3.375 = 1.5 horsepower.
A: Check the motors, mounts, and gearboxes at Team Delta.
A: Mark J. here: real specifications for R/C hobby motors are just not provided by the manufacturers. They offer all sorts of variation in magnets, armature winds, and can designs -- but they won't give real-world numbers on the performance differences. Without figures like amperage draw, torque curves, and real RPM it's a pure guessing game.
How did you decide that two 550 motors are the right number to power your FBS weapon without specs for the motors you want to use?? How are you going to select an ESC for the weapon if you don't know how much amperage the motors draw??
I'd stick with motors that provide real specifications, like the Mini EV [no longer available. If you go with a hobby 550, try to find a bot that's using something similar and use their experience. Otherwise, you're on your own!
A: Mark J. here: a 'machine wound' motor has the armature windings done quickly by a machine. The windings are not neatly and compactly arranged on the armature. A 'hand wound' motor is wound by a human who can take care with the wire and make the windings more regular and uniform. A hand-wound armature usually starts off better balanced and requires less drilling or grinding in the dynamic balancing process. The more uniform wire winds also result in a little more efficient use of available current. The overall difference is small and is of more interest to an R/C racer than a 'bot builder.
A: You can see a whole bunch of different drive and weapon Electronic Speed Controllers at Robot Marketplace - speed controllers page.
A: Mark J. here: Axi brushless outrunner motors are built for the model aircraft market and the specifications they list are designed to make it easy for an airplane builder to pick the best Axi for their purpose. It's not easy to extract robot type specs from the information provided.
The simple answer to your question is no. At the same voltage, the Axi 2208/20 will produce both more torque and more RPM than the Axi 2208/34. It will also pull almost three times the amperage at stall and will require a larger battery and an ESC with greater power handling capability.
A: Mark J. here: The manufacturers don't tell, and the R/C racers don't care. To find out, you'll have to buy one and do your own testing (or find someone who has). A procedure to measure the performance specs of a DC motor, including stall amperage, can be found here.
I dug thru my box of old R/C parts and found a 17 turn hobby R/C motor. A quick test using the 'D-cell method' shows 31 amps stall current at 7.5 volts. That seems a little low. A 19 turn motor would pull less stall current, but I can't say how much less.
A: Mark J. here: there are several methods for determining the maximum current a permanent magnet DC motor will draw at stall. I think the simplest for most builders is the 'D-cell' method. You will need:
The reading provided by the ammeter is the stall current at 1.5 volts. Current is proportional to voltage, so the stall current in operation would be this reading times (V ÷ 1.5) where V is the operating voltage of your 'bot. This all assumes that your 'bot battery can actually deliver the calculated amps, and that the resistance of your speed controller and wiring is negligible.
A: Mark J. here: an insect class FBS has a disadvantage -- it has to spin-up really fast because of the small arena it will fight in. That's why you don't see many beetle FBS designs. A pair of Speed 400s will give you about 150 watts of power -- too little to spin up a heavy shell before your opponent gets to you.
Q: If I still want to build a beetle class FBS, what kind of motors would be adequate?
A: Mark J. here: I've posted before about calculating power requirements for spinning weapons. 'Adequate' power depends on:
The cardinal rule of FBS design is: "There is no such thing as too much power." For a beetleweight friction-drive FBS, there many not even be such a thing as enough power. I'd consider another design.
A: I really don't think that there are worthwhile options to a good speed controller, but I have seen some other things tried:
You can use the Team Delta RCE225 dual ended switch. It plugs direct into your R/C receiver and will give forward/off/reverse control for a single motor (rated 24 amps) for $42.50. Works best on slow 'bots.
If you're really hurting for funds, I remember seeing some early 'bots that converted the variable speed trigger that came from the same cordless drill as their motor. They removed the trigger spring and used a servo to move the throttle. Another servo ran the converted forward/reverse switch. Clunky and hard to drive, but it did work.
A: Sorry -- I've never used either a Jeti or Phoenix. Try asking that question at the Delphi Antweight Forum.
A: You need to be very careful when interpreting motor amperage specifications. Brushless motor specs may list a "maximum loading" which is a recommended amperage over an extended period of time that is much lower than the peak amperage draw! Electronic Speed Controllers are usually rated for the peak amperage they can provide over a very short time period.
It doesn't hurt to use an ESC with a higher amperage rating than you need, and it may keep you from cooking the ESC. When in doubt, use a higher rated ESC.
A: Tech question, Mark J. here: In a brushless motor, each of the three leads is directly attached to different spots in the stationary field windings of the motor -- just as a commutator would be attached to the rotating field windings in a brushed motor.
Since brushless motors have no commutator, they require some electronic 'intelligence' to correctly assign the current flow to the field windings of the motor as the armature rotates. This intelligence is built into the motor controller matched to the brushless motor you have selected. The motor controller also acts as speed controller on hobby brushless motors.
A: Mark J. here -- You can turn your weapon motor on and off with a simple mechanical relay or contactor, but that puts a really big momentary drain on the battery. The drain can drop the voltage so low that your radio receiver cuts out. A speed controller allows you to feed power smoothly to the weapon and avoid the voltage drop. With some designs, it's also handy to be able to reverse the weapon direction if you're inverted or if the weapon gets fouled on something.
Also, the new 'brushless' motors require a controller to run at all. You can't just hook them up to a battery -- it won't work!
A: Technical question - Mark J. here: The wire colors are different, but the signal and polarity is the same on Futaba, JR, and Hitec radio systems -- they are fully interchangeable. Some Airtronics radio gear has the polarity reversed, so always check before mixing anything with Airtronics. More info on the different connector types can be found here.
A: Technical question - Mark J. here: A twin motor controller can handle two (or more) motors in parallel on each channel (four total) if the total current draw is within the amperage rating of the controller. The specific controller recommendation would depend on the total draw for a pair of the motors you plan to use.
If you want four motors each with independent control, then you'll have to use two, two-channel controllers.
A: If you want to go hard-core there are some tiny new brushless outrunner motors like the MSYS-300T that could make you 'King of the Fairies'. Only 20 grams, plus 2 grams for the controller!
On a lower budget, I think I'd try a warmed-up 130 size motor like the 'Hyper Dash 2'. They're light (17 grams), cheap, and put out crazy RPM at low voltage.
A: I'm getting tired of plugging Robot Marketplace, but how about trying their geared motor section?
A: They're the same size and they weigh about the same, but the Speed-300 spins faster, puts out about 3 times as much power, and consumes about three times as many amps. There's also a Speed T-280, and a Speed S-280 that each have their own specs.
A: The T-280 draws 5 amps at max power output, but much more at stall. If your application makes sure it never stalls because of high gear reduction or slip clutches, and your ESC can really handle 5 amps for as long as you plan to abuse the motor, then maybe - but I wouldn't.
A: All sorts of motors are found in R/C cars. Many toys have the 130 or 230 class motors that are popular in antweights. Larger scale hobby R/C racers have motors too heavy for most antweight designs, but they might work well for beetleweight weapons or hobbyweight drives.
A: The 130's are more than enough, but heavy. Take a look at the Sanyo NA5S motor/gearbox. Less than 0.3 ounce each!
A: 'Best' depends on your design needs. The 'outrunner' style (Axi, Park) motors are widely used because of their high torque and reasonable RPM. I like the Axi 'cause it's easier to mount.
A: You don't see many ant thwackbots! Thwackbots use their high turn-in-place speed to spin an attached hammer or axe up to dangerous velocity. In a small ant arena you'd better be able to spin up quickly. Remember to keep your track width as narrow as possible to maximize spin RPM. More power is better, but I think a pair of hot Speed-280s should be enough to get you some respect.
A: The 'Speed 280' motor is a good choice. It's inexpensive, light weight, and has enough power to be feared! The current trend in weapon motors is 'brushless' -- but these are expensive and require special controllers. Try a Speed 280 for your first spinner! They are used in R/C aircraft, so your local hobby shop may stock them. If not, check out the ant motor selection at The Robot Marketplace
A: Technical question - Mark J. here: NO! Most micro electronics have a very narrow range of operating voltage. Raising the voltage can fry them. Unless you're very sure you know what you're doing, keep the voltage to your electronics at the original specified levels.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
If the transmitter and ESC both have mixing turned on 
A: [Mark J.] Brushed ESCs make sense to me. Brushless ESCs 
A: Properly selecting, installing, and setting up VESCs is more like joining a cult than building a robot. There are constant releases of new firmware from the 'VESC Project' and the ability of the associated 'VESC Tool' to properly setup a specific controller for your application is variable. If the setup provided by the tool is causing performance issues in your robot you will decend into a very dark place where a great deal of trial and error may or may not provide a solution. It can drive you nuts. Many teams have difficulty in this area and resort to the services of a VESC whisperer.
I think I'd run a shorter motor. 
If your transmitter has channel mixing capability it is generally preferable to use that option and leave the DESC mixing turned off.
A: [Mark J.] The FPVKing ESC is sold under many names and may be ordered direct from China for less than $5. For that price you should not and can not expect high quality components or reasonable quality control. A fair percentage of these fail to work right out of the box, the instructions are gibberish, and I would not trust a single solder connection on the board in combat. Ask yourself if you want to invest considerable time, effort, and money to build a combat robot and travel to a tournament only to have a toy ESC fail and send you home.
Q: Hello, right now I’m having a a problem with one side of my robot. We have two drive motors connected to two of the same ESC’s- BlueRobotic Basic ESCs. I’m using an RF remote controller. One side of the robot runs fine. The other has a weird issue. When we toggle the joystick, forward or backwards slowly, even until max throttle, the motor turns the wheels just fine. However, when we try to toggle the joystick quickly, the motor starts sort of lagging. It goes back and forth but lags often. Especially when we switch between backwards and forwards on the toggle quickly.
A: [Mark J.] Once in a while a few crates of motors were found in a warehouse and liquidated to a surplus dealer. These motors may have originally sold for a couple hundred bucks apiece, but for whatever reason their market dried up and they were available dirt cheap until the supply ran out. Some proved useful for combat robots:
Q: I bought a robot someone built: twin 24v motors and two 24v Victor 883 ESCs. No transmitter or receiver, but supposedly plug and play to put in. I'm researching that and have found a 6-channel system that may work - just waiting on few answers from the manufacturer.
A: The device labled 'Removable Power Link' in the diagram is the master switch. In large robots this switch is usually a 'removable link' for safety purposes. A removable link is just a two-pole connector with one side shorted by a loop of wire -- pull the connector apart and the circuit is broken.
Q: How powerful are the brushed 
Me and my son are newbies to robot building and I have a very rudimentary question regarding weapon motor. I have always wondered how does the motor continue running even when the weapon attached to it has stopped spinning or rotating.
Q: Looking at robots like 'Sawblaze' and 'Skorpios', how are they controlling the end points of their arm movement? I know they aren't using servos at this scale, and I don't think they are simply using hard stops and letting the motor stall. Do they use limit switches? Are they building their own servo control system into their larger mechanism? Thanks. 
Any idea how I can make a BEC-disabled Scorpion work? [Social Media]
A: [Mark J.] First, attributing the success of 'Uppercut' entirely to their custom weapon motor is an error. There are many examples of winning robots with ineffective weapons, and many more examples of losing robots with awesome weaponry. A successful combat robot depends on all of the basic mechanical systems working well and reliably.
One weakness of the Scorpion Mini is its battery eliminator circuit (BEC). Does the hot chip look like the photo at right, and is it next to the 'BEC' label on the ESC board? That's the voltage regulator that reduces battery voltage to 5 volts to power the ESC and the receiver. Either the voltage regulator has gone bad or your receiver bus is drawing too much power. Take a look at this post in the 
For simplicity, the (orange) Leopard is mounted next to the (black) Magmotor with a timing belt drive joining their shafts. The Leopard spins about three times as fast as the Mag so the pulley sizes are adjusted to compensate. The Mag remains in its original mount position in the chassis, and its shaft drives the chain and sprocket drivetrain. The team reports that the power from the Leopard frequently strips the teeth from the timing belt connecting it to the Mag, so watch for modifications to that part of the drive.
A: [Mark J.] The process is very similar to mounting a brushed motor. Here's a bare-bones description based on info from builder Emmanuel Carrillo:
A: [Mark J.] I agree that it's an ESC issue, but your problem is not a standard failure mode for the Scorpion. From your description I'll guess that you have a broken PC board trace and/or a failed solder joint. Break out a magnifier and spend ten minutes examining the copper traces and solder junctions. Keep a lookout for stray scraps of wire that might be shorting something. Pay close attention to the motor connector block in question.
A: [Mark J.] I don't know where the practice of building a thick epoxy 'shell' around stator windings became popular. It does reduce stator cooling, and it does not prevent the lower layers of wire from shifting under impact load. A better method:
A: [Mark J.] The simple answer is 'no'. Larger diameter outrunner motors tend to spin slower and deliver more torque than smaller diameter outrunner, and longer motors tend to produce more torque than shorter motors -- but there are multiple elements in the internal design of the motor that have larger effects than do external measurements:
Q: as my design is pretty compact & we have size limitations out there.. i have at most 11 inches for all three wheels. so,do you think it'd be fast enough with 3.5 inch wheels.??
A: [Mark J.] So, you were out shopping for a beetle ESC and all the solutions were running about $50. Then you came across the Dual TB6612FNG for $4.95. What a savings! Did you at any point wonder why the Dual TB6612FNG was available at a 90% discount over the stuff other builders use for their beetles?
A: [Mark J.] How much extra salt should you shake onto your hamburger? Some need a little, some need a lot, some have too much already.
A: [Mark J.] There are a variety of Chinese gearmotors on eBay that give adequate performance in larger insect 'bots. The pictured style is available in several gear ratios and has standard motor mount hole spacing. This allows you to bolt up different motors, like the RS-300 series BaneBots or even small brushless outrunners.
A: [Mark J.] 
Q: Why are RageBridges such a sought after ESC? I get that it can run two motors at once, but the specs say it can only supply each side at 40A continuous. Isn't that rating only good for beetleweights and under?
I've probably missed some fundamental knowledge with the entire ESC shebang, so could you explain to me how it works, pros and cons, what it's good and not good for, and why exactly it seems to be so popular in the combot community? [Ontario, Canada]
Q: Hi Mark, what is an EDF brushless motor, and is it a good idea to use them on lighter bots to power spinning weapons? [Sydney, Australia]
A: [Mark J.] Yes, and yes.
'Net wisdom' recognizes that these cheap motors are generating a lot of electrical 'noise' that interferes with the speed controller electronics. The usual suggested solution is to add capacitors from each motor lead to the metal motor case in addition to the standard capacitor (often supplied with the ESC) across the motor leads. This treatment may suppress enough of the electrical noise to allow the 'bot to function, but it ignores the true cause of the problem.
Q: Weird Question Aaron - instead of rewinding brushless outrunners I was wondering if stacking the cores and outrunner magnet shells on one shaft and wiring each set of motors core in series would give an advantage over running them in parallel.
Q: what are the pros of using 2 motor to power a single shaft, like will bales use in hypershock after he "un-suckified it", instead of just using a motor or each wheel or one for 2 wheels? [Quebec, Canada]
Q: Can you explain what "brushless" means in a way that someone who hasn't studied engineering since high school can understand? [Logan, Ohio]
Q: I read your 
A: [Mark J.] 'Recent' is right -- it was uploaded just three days ago. Kevin Barker has gone a different direction than other teams with his robot and weapon drivetrains. Most builders are comfortable using brushless motors for weapon drive and stick to brushed motors for the robot drivetrain. Keven sticks to a proven brushed AmpFlow motor for the spinner weapon but has patched together a brushless robot drive from new inrunner brushless motors and old gearboxes from his parts bin.
Q: Dear Mark Joerger,
Q: what is the logic behind the new wedge of original sin? is it beneficial to have such wedge? [India]
Q: Hey Mark, I'm thinking about using the AmpFlow G43-500 motor to power my spinner weapon. What do you think? [Location Withheld]
Q: I have a question about metal case on bottom layer of Sabertooth 2x60 motor controller, can it be disassembled from PCB without damaging the controller? [India]
Q: hllo sir..
Q: hey mark .. any idea about botblitz esc? the 85amp version . and why do these esc look like brushless ones? are these combat suitable ? [India]
Q: Hey mark,
A: [Mark J.] We only answer robot questions at 'Ask Aaron', so I had to re-write your query just a bit...
A: [Mark J.] A very wide range of motors are used in cordless drills. Many drills use motors that resemble the BaneBots RS-550 motors (225 watt output), but they are made by differing manufacturers to differing voltage and power specifications. Performance varies greatly: power output runs anywhere from sub-100 watts to more than 900 watts.
Q: i want to build a robot which can play badminton .How to choose dc motor for base platform having four wheel mecanum wheel(100mm)of robot ,if robot having wieght of 20 kg? [India]
A: [Mark J.] Your BattleSwitch gets power for its own electronics from the receiver connection, but it does not pass that power thru to the screw terminals. The power it switches comes from an external battery source. If the coil voltage is the same as the weapon motor voltage, you can use the diagram in the very first question in our 
Q: Mark, Philadelphia here again.
Q: Hey, I am unable to [mount] my ampflow e 30-400 motor vertical on my base...those screw in the base of motor are [too] small to hold it...suggest some solution? [Tamil Nadu, India]
A: You're trying to do too much with just two channels. Control of a skid-steer robot drivetrain requires two control channels to provide independent control of motors on the left and right sides of the robot. This allows for full directional control:
Q: What do you think about the Equals Zero DeWutt gear motor? Do you think its a reasonable replacement for the Team Delta Powerdrive kit? [Oregon]
Q: Thanks a lot dude.... but the thing is Arenas in India are just 5m x 5m in size and don't need speed control for my drive and weapon motor. I am using two E30- 150 ampflow motors for driving and one 1 F30-150 for my weapon(drum), so what would you suggest me to use instead of victors for motor control. (please, as far as possible suggest me something that would be available in India also as I wish to make my bot completely wireless unlike other bots in India)
A: If you want to control the speed and direction of the motor, you will need an 
Also it now has a pronounced drift to the right going forward, as in one motor is running faster then the other. Did not do this with the original board.
A: 
A: Electronic Speed Controller selection depends on more than the motors. ESCs are rated by the amperage they can handle, and the peak amperage draw of specific motors will depend on wheel diameter, weight on the driven wheels, and voltage used (most builders overvolt drill motors).
A: We're big fans of 'scrap yard' parts, but maybe you should splurge and use motors for which you have at least basic performance figures?
A: Mark J. here: in most cases, the answer is 'no'. Although the input signal from a common receiver would be the same going to two speed controlers, each ESC has its own internal 'clock' that regulates the power output pulses. The two ESCs would sometimes cooperate, and sometimes be completely out of synch. This applies to both brushed and brushless motor controllers.
Q: If I'm only pulling about 30 amps continous from a drive motor, is it safe to remove the fan from the Victor 883s? They're rated for 60 amps continous. I'm not using the 884s because I'm running at 24v.
Q: Can you help me find the torque curves for the A28-150 Magmotor motor? Thanks.
A: Hi, Daniel. Team LNW's hobbyweight 'Defiantly Daft' (6 wins, 7 losses) is an able competitor, but it is not known for massive power. Why copy their motors?
A few basics: you're controlling speed on a single motor and I assume you will want to control it in both directions (forward/reverse), so you will be looking for a single-channel dual-direction permanent magnet DC controller. A key factor in motor controller selection is knowing how many amps of current the motor will be consuming, and for how long. For example, here are the ratings for a popular line of robot controllers:
A: 'The guy who made Toro' would be either Reason Bradley or Alexander Rose of Inertia Labs. They used and sold the IFI Robotics speed controllers. If those don't look right I'm gonna need a decent description. Alternately, you can contact Reason and Alexander thru the 
