I was browsing the motor curves again and came upon an issue that has always puzzled me. As far as I know, we are only allowed to provide motors with 12v of power. However, there is always a Banebots RS775-12 and Banebots RS775-18 listed. Maybe I just completely don’t understand it but which motor is attached to the PG71 and PG188 gearboxes sold by Andymark? Is it a RS775-18 being run at 12 v? I ask this because I’m looking to compare the PG71’s operated by 775 series motors and AM 9015ones and unfortunately Andymark has no data on the 9015 ones.
Banebots M7-RS775-18 at 12 volts Stall Current: 87 amps.
Banebots M7-RS775-12 at 12 volts Stall Current: 30 amps.
AndyMark 9015 Motor (am-0912) at 12 volts Stall Current: 63.8 amps
The motor stall current @ 12 volts is the same regardless of what gearbox it may be attached to.
The PG-71 and PG-188 use a am-2161 motor (RS775-5). The RS775-5 is a 12v motor, if the description on the case is to be believed.
I guess I still don’t get it. The power of a RS775-12 is in the range of 75 W while the 775-18 is 270 W. If the one on the PG71 is closer to a 775-12 then shouldn’t its power be around there?
I should probably explain more the issue we’re having. We were using a PG71 with an AM9015 to rotate and hold the position of our shooter (think motor hooked up directly to the side of it, rotating the whole shooter/hopper assembly). We had a PID loop running with a potentiometer directly to the jaguar but due to some mechanical stupidity managed to mangle the pot (rotated beyond its range). We switched to manual control and after about 10 minutes of use, the motor started smoking. I guess my first question should be to ask why do you guys think this happened.
I understand that it was our own bad design and mechanical disadvantage that caused the problem but at this point, our only solutions are workarounds. Our current thought is to put on a PG188 and let the increase in torque traded for speed that we didn’t need help us.
I guess the other question to ask is general motor theory: What goes on inside the motor current and heat wise when we run it at say 30% to hold a position? Does it depend on how much load is actually put on it? How would we calculate this based on the lever we’ve created?
Both motors depend on fan cooling. At zero speed you do not get any. Also the higher the voltage, the higher the stall current & the more heat from I^2R losses. So at 12 volts you rapidly cook the motor. At a lower voltage it will take longer. Running a PID loop you are running just enough voltage to hold position. Manually you are probably running more voltage (& current) than you need.
We has a similar issue with our Logomotion arm. We solve it by limiting the joy stick command to just enough to stall the arm. We had a turbo button to increase the command to actually raise the arm when needed.
Simon, there are a lot of things going on here, and I can see why it’s a bit easy to get lost in all of it.
When looking at motors, specifically those of the same ‘size’ or ‘class’ you’ll see that they can vary wildly. What you’re seeing with the 775-12 and the 775-18 also happens with various 500-series motors as well - just look at all of the different FP’s over the years, the BB 555, 540, AM 9015 etc and you’ll see what I’m talking about. Differences in the motor’s internals will cause differences in it’s characteristics some of which can make the motor more or less powerful. The most important thing to take away from this is than not all motors in the same size class are equal - if you really want to understand why exactly that is, you’d need to read into DC motors a bunch. (To be honest, I don’t know well enough to even attempt to explain this in depth.)
Now that we’ve gotten that out of the way, it sounds like your problem is a fairly common one. 300-series, 500-series motors and 700-series motors are all cooled through a fan connected to the motor’s output shaft. When you stall the motor that fan isn’t spinning. Now you’ve essentially removed all of the motor’s cooling (other than the case) it’s very easy for them to overheat and burn up, especially when loaded. One of the most important rules to remember when using 500-series motors (FP’s included) is that they shouldn’t be stalled ever - if possible. I believe the same is true for the 775-12 as well, but I’m not entirely sure.
The 775-18 is a bit of a different animal. It’s a motor designed to run at 18V that’s now being run at 12V. Because of this, it’s a bit better at handling excessive loading and stalling in some applications. In 2011, the team I was with used two 775-18’s on that robot’s shoulder joints and prior to the machine’s redesign, we’d routinely stall the 775’s for 15-20 seconds without issue. That being said, 775-18’s did have a problem with case shorting when they were first introduced, and since then the general consensus among many people in FRC is to avoid them like the plague.
Now that we’ve gotten all of that out of the way, here are a few things that you can try to make your current mechanism perform as intended without major modification
Passive Assist - Depending on the exact specifics of your shooter’s pivoting axis, you should be able to add some sort of assist without too much difficultly. This passive assist can be as simple as some surgical tubing mounted just right, or as complicated as a series of gas struts/springs/whatever. In any case, the assist should balance the shooter in such a way that it is ‘weightless’. Once you get the shooter to where it’s balanced fairly well, your motor has to do MUCH less work - ideally it’s just overpowering the assist - which should keep you from frying more AM 9015’s.
Active Assist/Position Hold - Again, depending on the exact specifics of your shooter, you can look into adding a mechanism to actively assist the motor and/or hold the shooter in position without having to power the motor. There are multiple methods of doing this: Disc Brakes, Pin Brakes, Worm Gears, etc will all do the trick.
OR, if you really want to, you can swap out the AM 9068/PG71 combination for a motor that’s a bit more robust. I know for a fact that the BAG motors take stalling in this application REALLY well (it’s a ‘CIM’ afterall) but, switching to one requires the use of a versaplanetary which may or may not be a pain in the butt.
Personally, I’d start with passive assist, as it’s usually the easiest thing to play with. Add surgical tubing/springs/rubber bands/etc until you get the desired performance and you’re set. By adding or removing assist, you can tune the mechanism to do some cool tricks like being ‘self opening’ or being perfectly balanced in a certain position. Also, if you get lucky, the PG71 might have enough internal resistance to where you only have to get close to the ‘perfectly balanced’ setup for the mechanism to stay in place.
I hope all of this is enough to get you pointed in the right direction. If you need anything else, feel free to shoot me a PM.
Thanks for all the help everyone, this is why I love the FIRST community.
One more question: If the Banebots and Fisher Price motors are fan cooled, how are CIM and BAG motors cooled?
The CIM, MiniCIM, and BAG motors are all sealed motors with a high thermal mass. All this metal helps to absorb the heat generated by the motor, in the same way a computer heat sink works.
The large thermal mass of the structures inside the motor, as Art mentioned, acts internally as a sort of large heat “sponge” or “diaper” absorbing the heat and preventing short periods of high current from producing unacceptably high temperatures within the motor.
Radiation, conduction, and convection eventually move that heat out of the motor:
Radiation: the outer case of the motor emits radiant heat (electromagnetic radiation; photons)
Conduction: heat is conducted away from the motor by the metal to which it is mounted.
Convection: the air around the motor gets heated by contact with the outer case of the motor, and the heated air rises away from the motor and is replaced with cooler air.
I spoke with Andymark and they put up the statistics for the motor inside the PG71 Gearmotor. Looking at them, they are rather puzzling:
If the Banebots RS775-5 is truly the motor inside the PG27,71,188 series gearmotors how can they have different stall currents? Either the numbers are wrong (likely the 775-5 as the gearmotor combo was empirically tested) or there is actually a different motor in the PG71 gearmotors.
For more info on why I’m asking, see the link to the Red Raptor in my stat; it’s the wrist joint.
The numbers for the am-2161 appear to be self-contradictory. The 174 max watts number is inconsistent with 5700 free rpm and 35 oz-in stall torque numbers.
Assuming a linear speed vs torque curve, the max power would be (5700/2)(35/2)/12/160.14198 ≈ 37 watts. So something’s not right.
And that 37 watts number can’t be right either, because if you run that thru a 71:1 gearbox you’re not going to get 44 watts output (as spec’d for the am-0914).