RPM closed loop calibration

[trypt]

Hello,

I've been reading a bit around and still trying to get my head around how these calibrations work, but still don't get it.

Could some be so kind to explain how the currentlimit and rpmtovoltage works with simple words and how they releate to each other? And what the CLxTERM and FF1TERM parameters really means

and how do you read this graph?



is duty cycle = throttle in simple terms?

for optimal results is it best calibrate every engine and prop (on the same copter) or is the difference so small that its not worth the time?

and last question will the Autoquad FC work with closed loop RPM mode on the ESC32s? I cant find any info about this, but most pppl here seems to be using RPM mode on their ESC32s?

thanks

[bn999]

The FF parameters are used to calculate the feed forward term for the closed loop RPM PID controller, and are only needed if you want to use that mode. The CL parameters represent a 5th order polinomial fit of the current flow resulting from the voltage presented to the motor at any given RPM.

The graph plots the calculated over the measured current from your calibration run. The two sets of lines should match up pretty well at higher RPMs. The above example looks very good. If you post the FFTERM graph and state the motor and prop makes and models, I'll add them to the database.

You only want to perform these calculations for any given motor/prop combo so there is nothing to be gained by repeating it if your results are good.

The public attitude controller for AQ does not take advantage of the closed loop modes of ESC32. I've only read of one person speak of utilizing these modes but I do not know the details.

[trypt]

Thanks for the reply! The picture was just from an example, but I will post my result when I have done the calibration.

[trypt]

I have started to do some calibrations and got some more questions.

Does the -a (amp) parameter of the esc32cal program set the actual MAX_CURRENT of the ESC32? So that I don't have to change the MAX_CURRENT parameter as well when running calibrations?

Im guessing the ADVANCE and SWITCH_FREQ paramters have impact when running esc32cal calibrations right?

What is causing the strange noise on the engine when it suddenly stops to spin and the ESC disarms? Is this because of the current limiter function of the ESC? Or is it timing (advance) issue? Or is it because the engine overloads and get to many amps? When this happend to my engine it draw 25amp (its rated for 27A for 180s by Tiger) I think the DISARM_CODE was 5.

[bn999]

The -a parameter only sets esc32Cal's testing limit. You are left to decide what the parameter should be and set it in the ESC. All other parameters are left unchanged and effect the performance of the ESC.

DISARM_CODE 5 is defined as REASON_CLI, which means that the user disarmed through the CLI. So, that was probably not the right code.

esc32Cal removes all of the normal throttle shaping and current limits in order to probe how far the motor can be pushed so that a performance profile can be developed. With some motors, the result is a loss of zero crossing sync due to high switching noise and other distortions in the feedback waveforms. It is the user's job to set the -a parameter low enough so that this does not happen. It can be dangerous to the motor if you dump too much current in a single step as does esc32Cal.

Changing the switching frequency or timing advance can help or hurt the situation. You just have to keep experimenting until you find the best combination of parameters. Since you can damage motors, always start conservatively and work your way up.

[axelnied]

[bn999]

It's been a while, so it is not fresh in my mine. Take a look at the esc32 source and see how its being used.