[FTC]: Initial Modern Robotics Motor Controller tests.
 

My FTC teams are always focused on pushing Autonomous capabilities on our robots, so knowing how the motor controller (and accompanying software) works with encoders is something I always need to evaluate each season.

With the new hardware, this year is no exception.

I just ran some basic tests with the two most common FTC motors, to see how well new controller implements the three standard control modes on an unloaded motor. These are: open loop, closed loop and Run to Position.

The results were...(in order) Ick, Cool, D'oh!

I've posted my Summary doc (with data) here.
http://www.chiefdelphi.com/media/papers/3157

Suffice it to say, I'm disappointed in the "Run to Position" mode.

Phil.

Re: [FTC]: Initial Modern Robotics Motor Controller tests.
 

Thanks so much for posting this. I'm going to share it with my team.

I think I found a typo in section 2.0. You wrote "Mode “A” (RUN_WITHOUT_ENCODERS) produced the least predictable and consistent results. The MATRIX motor has an initial dead-band," but I think you meant "Tetrix motor". This confused me for a brief second because there is a new legal ModernRobotics / Matrix motor.

The most disappointing part was the clipping of speed to 78% with the new controller and NeveRest motors. I was looking forward to the new controllers because I thought this was sure to be fixed.

Re: [FTC]: Initial Modern Robotics Motor Controller tests.
 

Me too.

I'll fix that typo, thanks.

Phil.

Re: [FTC]: Initial Modern Robotics Motor Controller tests.
 

The "clipping" is only in "Run with constant speed" mode and it is done for the very specific purpose of allowing the function to have headroom to adjust the motor speed in this mode. In this mode power (not speed) is limited to around 80%. There is nothing to be fixed. Running the motors in other modes will allow power setting up to 100%. See the posting on the FIRST FTC forum for an explanation of how the motor controllers really work.

Re: [FTC]: Initial Modern Robotics Motor Controller tests.
 

OK, thanks, Steve. I was worried about the clipping at 78% because I thought it was caused by the Tetrix encoders having a different count-per-revolution than NeveRest encoders. I'm glad to hear that's not the case.