Mixing Cims and Mini Cims

Hello,

My team is experimenting with different drivetrain and gearbox designs. We are thinking about using 1 cim and 1 mini cim per gearbox. While we know this will work, I’m wondering how the different speed of the mini cims will effect the slower speed of the cim.

Speedwise, there’s no issue. The miniCIM was specifically designed for FRC to be a drop-in replacement for a CIM. There is no problem mixing CIMs and miniCIMs in a gearbox.

The bigger problem you will run into is power. A miniCIM has approximately 2/3 the power of a CIM. Having only 3.3 CIMs of power in your drivetrain will cut into your top speed, your acceleration, and your pushing power. Nowadays very few teams use less than 4 CIMs of power in their drivetrain (excluding 2015). You will be at a distinct disadvantage if you try to short-change your drivetrain.

I’m interested in what benefits are driving you to look into 1 CIM 1 miniCIM drives.

If you have four separate gearboxes like mecanum or maybe swerve, it could be a viable option as you get more power without hitting the 6 CIM cap. However it’s very application-specific. You will lose a pushing match with 2 CIM and 2 miniCIM for sure, if that’s all you have in the drivetrain.

The most common application where teams mix miniCIMs and CIMs tends to be on 3 motor drivetrain gearboxs with 2 CIMs and 1 miniCIM per gearbox (with a total of 4 CIMs and 2 miniCIMs powering the drivetrain). Teams usually do this to have considerable power and acceleration that surpasses 4 CIM drivetrains while avoiding the potential brownout risks of running 6 CIMs.

229 used WCP single speed gearboxes with 2 CIMs and a mini-CIM, and we loved it! We had a few matches where the mini-CIMs had to be taken out and there was a noticeable difference in our acceleration. We had no issues with brownouts or fuses, but that may have been due to being an 80lb robot with only two other motors.

Wait, don’t the minicims spin at a faster RPM? Wouldn’t mixing the CIMs and MiniCIMs result in damaged motors and gears?

Not at all.

https://www.chiefdelphi.com/forums/showthread.php?p=1278900&highlight=Mini+CIM#post1278900

https://www.chiefdelphi.com/forums/showthread.php?t=156083

EDIT - Well, yes the MiniCIMs have a higher free speed. But read the threads, there’s a lot of good info in there. And I’d say Paul Copioli is a pretty reputable source on this.

Running a mini CIM with 2 CIMs causes the mini CIM run slightly hotter than the CIMs. This is because the mini CIM runs slightly faster so it’s working harder (because its trying to go faster but we’re not letting it) than the other 2. We ran one of those gear boxes last year and that is what we experienced. The heat wasn’t enough to worry about.

Are you not letting it in code or mechanical?

I don’t doubt your observation, but I think your purported cause is kind of dubious here, since the mini-CIM is not going to be “trying to go faster” across all speeds. Note that a Mini-CIM is significantly smaller than a CIM, and thus has less thermal mass…

At very low voltages, I think Mark’s explanation is valid, but once you get past around 10% voltage applied, that explanation breaks down. The observations and conclusions are the same though.

I agree it’s very application specific. There are so few cases where this is a good idea that if OP was in one of those cases they probably would have specified. I could possibly imagine it being used for a manipulator, but this post was specifically talking about drivetrains.

If you’re running swerve or mecanum drive and have 1 CIM and 1 miniCIM on each wheel, that’s 6.7 CIMs of power in your drivetrain. I really don’t see a need for all of that power; 2 miniCIMs each (i.e. 5.3 CIMs total power) will give you plenty of power for less weight. Also, if you’re actually taking advantage of all that power you’ll need to pay good attention to your total current draw because you’ll be even more prone to brownouts and main breaker trips than a 6 CIM tank drive. Unless you really feel like you need that extra power, I simply don’t see a need for a 1 CIM 1 miniCIM drivetrain gearbox.

For drivetrain gearboxes: no. See above explanations. However, technically there is a case where you are correct: if they are both connected to separate, mechanically-linked, non-backdriveable gearboxes.

The motors don’t damage each other in a normal setup because the faster motor pushes the slower motor a bit and the slower motor pulls on the faster one until they find an equilibrium. That will give you slightly different current draws in the two motors, but overall it’s not a problem.

However, if they are connected to separate, mechanically-linked, non-backdriveable gearboxes, they will not be able to find an equilibrium because the extra speed from the faster motor cannot pull on the slower one and vice versa. I honestly don’t know what would happen in this case, just that it will not be pretty and something will end up breaking. This is a very rare case for manipulators, and is non-existent in drivetrain gearboxes.

Yep. Good points. And if you actually had a CIM and a miniCIM on each gearbox in swerve, that’s 12 motors which leaves you very little to do much else. I guess the only real reason to do it would be to do 2 CIMs and a miniCIM or a CIM and 2 miniCIMs in a 3 CIM gearbox. That would would give you 5.3 CIM or 4.7 CIM of power while reducing weight. Overall it’s still a very niche case and probably wouldn’t be done much.

You’re right about the thermal mass, if all 3 motors are working equally, the mini CIM will get hotter because it’s physically smaller but in this case, the mini CIM will always be “trying to go faster” or another way to say it “works harder”. The Kv of the motors are different. No matter what voltage is applied to the motors the mini CIM will always attempt to run faster than the CIMs. This will “push” the CIMs at free speed with no friction. This does’t mean that the CIMs do nothing. All motors apply power to the common output just the mini CIM is doing more than his fair share. In this case, not enough to matter.

I understand why it works to run motors with mismatched speeds and I know that’s what Copioli recommends, but I’m not 100% clear on why he recommends it. If you use an 11t gear on a miniCIM and a 12t gear on a CIM, you get a 0.4% difference in speeds (according to motors.vex.com). If you use the same gear for both, you get a ~10% difference. I know it’s not necessary, but wouldn’t matching the speeds closer give you a more efficient system? Maybe someone can explain the logic that I’m missing here.

The motors are not running under zero load, though. According to Paul Copioli, the CIM and Mini-CIM match speeds at about 15% load. At higher loads, the CIM will be “pushing” the mini-CIM.

*Plot the 12 volt torque vs speed lines for CIM and mini on the same graph and it answers a lot of your questions.

At full throttle (12V) the CIM and mini torques are equal at ~4,440 rpm.

The CIM has higher torque at higher loads, but it never “pushes” the mini. The mini is always contributing torque.

*





This season, we used a 1 CIM 1 775 Pro per side on our drivetrain all season and got away with it. For our next event, we were able to redesign parts of our robot allowing us the weight to add more motors.
In 2010, we had one of the best pushing robots in FRC that year, running only a 1 CIM 1 AM planetary FP motor on each side due to weight constraints.

Here’s a spreadsheet to show the behavior of arbitrary combinations of motors given arbitrary torque loads. See the “notes” tab for the math.

Even if the motors are geared off a bit, they’ll both still contribute power. The trick is balancing the power contributions of the motors at a useful operating point (not necessarily free speed).

Ether,

Is it safe to say that at CIM + mini free speed, the mini is pushing the CIM. Between CIM + mini free speed and CIM free speed the mini is contributing more than it’s far share. Anything less than CIM free speed everyone’s all in. If so, I stand corrected.