How to generate profiler firmware for 6.2

[Jedijump]

Hi Zied,
After waiting patiently for MCSDK 6.2.0, I was overjoyed when it was finally released.
However, I could not get the motor profiler to communicate with my board. It seems to require a proper firmware to be downloaded before it can run. It does not allow me to select the correct boards even after the uart connection is done (although it instructed to "try to configure the board"). see attached screen grab.
[VERSION]: MCSDK 6.2.0
[TOOL]: Motor PROFILER v1.2.4
[DETAILS]: Not able to choose any control nor power board
[EXPECTED BEHAVIOR]: I could usually be able to select the boards and enter into "Start Profile"

I saw from other posting, your suggestion to download .hex into the board first and so I had tried to use  NUCLEO-F302R8#X-NUCLEO-IHM08M1 3Sh#(6.1.2).hex from the MCSDK 6.2.0 folder. Once connected, it shows the boards, but I want to use CTM009v1, not IHM08M1. So, that is a problem. See attached 2nd screen grab.

Thirdly, I had also used the Workbench 6.2.0 to create a new project, select the right boards but unable to use my motor because no motor file is created yet (failed at Motor profiler).
Please help, I need to get this going but it has been delayed due to these configuration issues.
Thank you.

Regards,
Jedijump
 

[Zied b.]

Hi,
I managed to reproduce the problem, I will investigate to find the cause of the problem and I will let you know if I find a solution to provide you.

Regards

[Zied b.]

Hi @Jedijump ,

After analyzing your problem, we found a bug which means that the Hall Profiling part does not work with Boards other than G4. While waiting for the next Release which will provide a Fix to this problem, here is the new Lib that you must replace in MotorControl\libMP as well as the mc_math.c and mc_math.h files that you must replace in Inc and Src folders. 

Please tell me if everything is ok for you.
Regards

[Jedijump]

Hi Zied,
Thank you for your patch. I have tried and it works! .. managed to complete the Hall profiling successfully without hanging halfway.
I had to copy the mc_math.c and mc_math.h after the STMCubeIDE code generation. Is that correct? Or is there a way to copy (or overwrite) earlier in the code generation process so that I do not have to repeatedly overwrite each time ?
It is ok for now. I am thankful to proceed further.
Thank you.

[Zied b.]

Hi @Jedijump ,
This is very good news, what you did is correct. MCSDK 6.2.1 will correct this bug. In the meantime you can use this patch.
You can also replace the FTL files, this will save you from copying and pasting the .c and .h each time, you must put the .ftl files in the location C:\Users\yourComputer\STM32Cube\Repository\MCSDK_v6.2.0 -Full\MotorControl\templates and your lib files that I provided to you above here: C:\Users\YourComputer\STM32Cube\Repository\MotorControlDev\libMP

Have a good day

[Jedijump]

Hi Zied,

Thank you so much for the ftl_file patch. It will save me from repeatedly copying and pasting at each code generation. Have a nice day.

Regards

[APerr.3]

Hi @Zied b. 

I have the same symptom with Nucleo F746 and IHM07M1.

Hall profiling go nearly to 100% for the first direction then back to 78% and board disconnect. The motor spin in the other direction. 

Your patch solve the problem too. Thanks again

Thanks.

Arnaud