Blown CIM?

[evand4567]

Quick question: is it possible to blow a cim motor, and if so, how would you test it to see if its performance has degraded?

We have a couple cims that were stalled a number of times for a non-trivial amount of time. I'd like to verify their performance before using or replacing them.

[Ether]

Originally Posted by evand4567

Quick question: is it possible to blow a cim motor

Yes, and this is how:

We have a couple cims that were stalled a number of times for a non-trivial amount of time.

how would you test it to see if its performance has degraded?

Measure the free current. This is easy to do. All you need is a $4 Harbor Freight digital meter.

Measure the free speed. You could do this with a calibrated strobe light or a special point-and-shoot rpm measuring tool. Or an encoder or tachometer.

Measure the stall current. You could use that $4 Harbor Freight meter to measure the voltage drop across a length of wire of known resistance. Or use the PDB current monitor.

Measure the stall torque. Put a torque wrench on the CIM shaft?

[Ether]

If you have 2 known good CIMs and a 2-motor gearbox, you could try the following:

1)

If you have a suitable encoder, install it on the gearbox.

Install the two good CIMs into the gearbox (with no load on gearbox). Run CIM1 with full voltage, and measure the voltage output of CIM2 (unshorted leads). Record the CIM1 applied voltage voltage, the CIM2 output voltage, and the encoder RPM. Now repeat with CIM2 powered and measure CIM1. Record those measurements. This is your baseline.

2)

Replace CIM2 with CIM3 (suspected bad). Repeat #1 and post your measurements here. Replace CIM3 with CIM2, and CIM1 with CIM3. Repeat #1 and post those measurements here.

[Ether]

Originally Posted by Ether

If you have 2 known good CIMs and a 2-motor gearbox, you could try the following:

1)

If you have a suitable encoder, install it on the gearbox.

Install the two good CIMs into the gearbox (with no load on gearbox). Run CIM1 with full voltage, and measure the voltage output of CIM2 (unshorted leads). Record the CIM1 applied voltage voltage, the CIM2 output voltage, and the encoder RPM. Now repeat with CIM2 powered and measure CIM1. Record those measurements. This is your baseline.

2)

Replace CIM2 with CIM3 (suspected bad). Repeat #1 and post your measurements here. Replace CIM3 with CIM2, and CIM1 with CIM3. Repeat #1 and post those measurements here.

3) Install the 2 good CIMs (CIM1 and CIM2). Short the leads of CIM2. Apply full voltage to CIM1 and record the encoder RPM. Replace CIM1 with CIM3 (suspect motor) and repeat. Record the encoder RPM. Post your measurements here.

[GeeTwo]

Originally Posted by FrankJ

Another possibility would be to run it with a known load & compare its speed & current draw against a new motor.

Originally Posted by Ether

Easier said than done, if you don't have a dynamometer.

You can compare a questionable motor to a known good motor by using the same load for each test. You don't necessarily have to know what the numerical value of the load is.

For example, if you have a CIM-powered elevator lift you can measure the lift time for a good motor and a suspect motor (using the exact same throttle command each time), and compare the results.

If you have a CIM-powered lift lying around (and for some strange reason a lot of teams do right now), you can use it to "measure" the stall torque by finding the weight of a load that you can just barely lift. You can also use it to evaluate performance near the maximum power by lifting approximately half of the stall weight and clocking the lift time. I fully concur with Ether that the simpler tests would be to compare your suspect CIM(s) to a population of known good ones, rather than trying to work through all of the gear ratios and possible sources of friction.

Also measure free speed and curent, as Ether put in post #1. Unfortunately, I don't think you can get a good stall curent measurement with the PDB as a CIM has a typical stall current of 133A, and IIRC, the current meter saturates at about half of that. Even if the limit is high enough, be sure to use a motor controller, as powering a motor directly from the PDP can damage the current sensors. Especially when you do the stall measurements, try to use the same battery in the same state of charge and the same wires for all the CIMs you test.

[FrankJ]

Yes it possible to blow them. Harder to tell if they are partially degraded. Many teams use new motors in their competition bots for that reason.

One possibility would be to use a good ohm meter & slowly turn the motor and watch the coil resistance & compare it to a new motor. That will only show a badly degraded motor though. like one with open windings.

Another possibility would be to run it with a known load & compare its speed & current draw against a new motor.

[Michael Hill]

Originally Posted by FrankJ

Yes it possible to blow them. Harder to tell if they are partially degraded. Many teams use new motors in their competition bots for that reason.

One possibility would be to use a good ohm meter & slowly turn the motor and watch the coil resistance & compare it to a new motor. That will only show a badly degraded motor though. like one with open windings.

Another possibility would be to run it with a known load & compare its speed & current draw against a new motor.

I've never had much luck with using an ohmmeter on a CIM. There has always been too much relative variation to get much of a measurement. I generally get somewhere between 0.4 - 3 Ohms (this is from memory right now) with a new CIM. I'm sure it has to do with the windings on the CIM and the shaft wanting to settle into a finite number of positions because of the magnetic field. Perhaps an ESR or LCR meter might be a better tool to measure the impedance with magnetics. I was just using a Fluke 87V DMM.

[Ether]

Originally Posted by FrankJ

Another possibility would be to run it with a known load

Easier said than done, if you don't have a dynamometer.

& compare its speed & current draw against a new motor.

You can compare a questionable motor to a known good motor by using the same load for each test. You don't necessarily have to know what the numerical value of the load is.

For example, if you have a CIM-powered elevator lift you can measure the lift time for a good motor and a suspect motor (using the exact same throttle command each time), and compare the results.

[FrankJ]

Originally Posted by Ether

Easier said than done, if you don't have a dynamometer.

Perhaps consistent or the same would have been better then "known". I was trying to think of simple tests without a lot of instrumentation.

I agree that ohming the motor has limited usefulness. It will tell you if you have open windings or a really bad commutator. Shorted winding look about the same as good windings to a typical ohm meter. Once again things to do with minimal tools.

[evand4567]

Did some testing and they definitely are degraded. Thanks for the help.

[Ether]

Originally Posted by evand4567

Did some testing and they definitely are degraded. Thanks for the help.

Now you can return the favor.

Please post your test procedure and test results so others may learn from what you did.

[evand4567]

We tested the resistance with an ohmmeter and found the degraded ones to be between .3 and .5, while the new CIM varied between 5-15 depending on it's position which is strange and I don't really understand what was going on.

Now for the unscientific part: we direct drove a wheel on and found a significant reduction in RPM noticeable to the naked eye. The sound motor sounded significantly slower as well.

That was enough for us to conclude the CIMs were bad, but if there's enough interest and I have the time, I can do some more specific and scientific testing of the motors.

[Ether]

Originally Posted by evand4567

We tested the resistance with an ohmmeter and found the degraded ones to be between .3 and .5

.3 and .5 what? ohms? milliohms?

while the new CIM varied between 5-15

5-15 what?

Now for the unscientific part: we direct drove a wheel on and found a significant reduction in RPM noticeable to the naked eye. The sound motor sounded significantly slower as well.

That's arguably more scientific than measuring resistance.

[Ether]

Originally Posted by evand4567

We tested the resistance with an ohmmeter and found the degraded ones to be between .3 and .5, while the new CIM varied between 5-15 depending on it's position which is strange and I don't really understand what was going on.

Can you please tell us what the .3, .5, and 5-15 numbers are?

[mman1506]

Measuring free current allowed us to identify 6 failed CIM motors In our 2014 drivetrain. While there was no noticeable physical damage or resistance the motors were drawing 240w (20a) with no load on them.