Motor charakterisation using NucleoF302R8 and IHM07/08

[leo.r]

Dear Ladies and Gentleman, 

as a research project to find a smart and easy way to characterize different brush-less DC-motors, we are working with your NucleoF302R8 and IHM07/08 circuits.

To begin with, we started using the "ST Motor Profiler" to profile different motors.

In doing so, we encounter several problems: 

1.) Motor does not rotate. 

Hardware: NucleoF302R8 / IHM08; Motor BO4560/12 ( peak current 50A)

Using this setup we manage to get electrical parameters only. As soon as the Profiler tries to spin the motor, the motor does not turn at all. All that happens is a "Klick" every 2-3s. On the oscilloscope one can see, that the IHM08 gives just a very short pulse on the motor, which is not enough to get it spinning. It repeats to pulse until the profiler timeouts. 

Question: Why does the profiler not get the Motor turning? (Motor Idle current 5A

        and what can we do to fix this problem? 

2.) Mechanical profiles does not complete. 

Hardware: NucleoF302R8 / IHM07; Motor A10-12S (5.3A) 

With this hardware, we are able to get the motor spinning. Still the profiler is not able to complete the mechanical characterization. The motors slowly is accelerated until 2500-3000RPM and then is powered off and stops spinning. This process repeats, until the profiler (again) timeouts. 

There seems to be no way to accelerate the motor to max. speed (configured with 25000UPM). The current drawn is also very low, its about 200-300mA, which is way less then 2.8A, the IHM07 board is capable of. 

Question: Why does the profiler not accelerate the motor to max speed to finish the mechanical profile? (there is no error/warning) 

3.) Motor profile randomly is completed. 

Hardware: NucleoF302R8 / IHM07; Motor BR2804-1700

Using the provided example, we are able to get a complete motor profile from BR2804 motor using the exact same hardware/software as before. It just does not work reliable. So sometimes the profile works perfectly and measures every parameter in the first try. Starting the exact same experiment just a second later then can sometimes lead to not complete the mechanical profile due to "overcurrent" faults. The Motor is accelerated until 16000UPM and then is slowling down slowly, until stopping with a randomly occuring overcurrent fault. This overcurrent cannot be seen on our powersupply, which provides about 200mA/11V DC.

Question: Why is the profiler not able to reliable complete the profile? 

        Where does the overcurrent fault come from, when there is no (obvious) overcurrent?

Having made this experience we have to guess, that the board is not able to "smartly" profile unknown motors, as it failed to complete a profile 2 out of 3 times. Even the tested and provided example does not work properly.

If the board is not able to characterise unknown motors, please inform us so we do not continue to waste more resources in finding a solution that cannot work.

If the Profiler is able to measure unknown motors, please give us information regarding the described problems and give us different examples of motors the board reliably can profile.

Thank you for your help,

Sincerely 

Leon

[FrancescaMC]

Hi Leo,

I have similar problems with the same boards.

Did you find a solution?

I am using Motor Profiler both v5.2.0 and 5.4.0 with F302R8+IHM08.

After setting the parameters of the BLDC motor and starting the profile, the motor runs but it is noisy and it vibrates very much.

I already test the same motor with other board and it was profiled without any problems.

Is maybe something linked to the hardware configuration?

I soldered J5 and J6 to have 3Sh mode, removed R181 and closed JP1 and JP2.

How did you set the IHM08 hardware to use the MP?

Thank you,

Francesca

[leo.r]

Hello Francesca,

No, we have not found a solution as we think a firmware issue is causing the overcurrent error. 1)

The Motor Profiler V5.2 stops the motor with overcurrent fault at a current of 1,4A, which is significantly lower than the 30A claimed by ST.

As we do not have access to the ST Firmware, we are not using IHM07 or IHM08 development boards anymore.

Our project will be continued using the open source project VESC, which offers nearly same features, open source code and hardware design as well as much higher motor current.

https://vesc-project.com/

However I was not able to test the VESC because I am not working on this project anymore.

I am using the exact same hardware configuration as the profiler suggests and it should be similar to your config.

Which Motor are you using? Which nominal current? I am quite curious to know, because my hardware was not able to drive motors with startup current higher 1,4A. However it was able to drive "small" bldc-motors quite smoothly, even with standard parameters.

As my personal opinion I would suggest using different hardware, as this is a dead project having no comunity and no customer support by ST (they closed 2 tickets from me without giving me the answers I wanted). Why are you working with ST Hardware?

Sorry for late answer,

Leon

[FrancescaMC]

Hi Leon,

Thank you for replying.

In the end, I was able to use Motor Profiler with IHM08. I tried to desolder different components, so I did not follow the instructions of the link http://localhost:5331/mp/bundle?ctrl=NUCLEO-F302R8&pwr=X-NUCLEO-IHM08M1%203Sh

Actually, the Motor Profiler code is based on FOC mode, so JP1 and JP2 need to be closed.

Then, following the notes of the board User Manual (on page 7/29) I desoldered also C3 C5 and C7.

Finally, also R181 has to be removed, because it is connected to the potentiometer and this does not allow to the Motor Profiler to control the speed.

Now, I have another problem always on this board.

The motor starts but then the voltage power supply decrease. This is really strange.

Maybe something in the code that causes short-circuits in the H-bridge.

The motor is a 24 Vdc with 0.35 A of nominal current in free running.

I am using this board because I need to test the driver integrated on the IHM08 for my project.

I find this board with the needed specs and with L6398 on it, so I decided to use it.

I will continue to fight this challenge, if you want, I will let you know.

Thank you again,

Francesca

[leo.r]

It would be interesting to know why you need to desolder these components and ST is not giving any instructions to do so. Good to hear its working now.

Did you try out IHM08 in default hardware configuration? Because it worked well for me to drive bldc-motors with similar current like yours.

I never realised problems with sudden voltage drops.

If you have a short-circuit in H-bridge you can probably measure the current to your power supply using a shunt and oscilloscope.

I did similar measurements to measure my overcurrent fault.

Do you need to profile motors with higher nominal current? Because in this case it would be interesting if you can reproduce the same over current error that I noticed.

It definitly is a challenge, so I am free to help you if I can.

I would appreciate if you keep me updated, just for my personal interest.

Thank you,

Leon

[FrancescaMC]

I asked myself the same thing about ST support and documentation.

Also it would be interesting to know why the default code generated with Motor Control Workbench is not working.

I tried also IHM08 in default configuration, but I need to have access to the FW and my motor is sensored with 3 Hall.

For the moment, I have only motors with similar current to this one.

I'll let you know,

Francesca

[Laurent Ca...]

The question has been moved from the "Motor Control Hardware" section to the "STM32 Motor Control" section (the question is about the STM32 MC SDK). 

Best regards