LAUNCHXL-F28377S: + BOOSTXL-DRV8301 Motor development kit and Current Spikes in no load operation.

Hi,

We can't reproduce this issue here. Can you try with compiler v6.4.6. Fyi, the SVGEN module is still the same, using the macro at \math_blocks\v4.3\svgen.h.

rgds,

ramesh

Hi Ramesh,

Thank you for responding and checking my issue with your hardware!!!

I am using: CCS-V-6.1.3.00033 and compiler used is: TI-V15.12.1.LTS.  I could not find suggested v6.4.6 compiler!  How to obtain it??

You are right, after simulating I find that SVGEN is all right !!

Best Regards,

Ljubisa 

Hi,
can you follow the video in this link to obtain an older version of compiler and see if you can find the version Ramesh mentioned?

[Edit] [Added below link]

http://processors.wiki.ti.com/index.php/Compiler_Updates

Best Regards
Santosh Athuru

Hi,
I updated compiler to older v6.4.6 and recompiled, but no improvement... Current spikes are still there in no load operation and in stand-still with no current applied.
Since you could not reproduce on your side, could it be, that I have faulty or noisy DRV8301 module?

Thank you very much.

Ljubisa

Hmm.. it is difficult to say DRV8301 is faulty.

Why don't you disconnect the motor and just do level 2, which is open loop. If you don't see the current spikes or noise issues, then the DRV is ok, and the current sense could be an issue. In that case, reduce the current sense gain of the DRV. Anyway, let me know how it goes.

rgds,

ramesh

Hi Ramesh,
I disconnected the motor with DRV_GAIN = 10 (16.5A base current) and run under Level-2, but still observing spikes, which can sometimes be +/- 0.14 x 16.5A on the display window. The amplitude is highest on B channel, about half as less on A channel, and they go in both directions + and -. They are of 1sampe duration (0.00005s).
Thank you very much,
Ljubisa

Hi Ljubisa,

Where do you see the current spikes? Is it in the ccs graph window?

Without the motor, why is there a current spike? Can you try with a different DRV board? Verify the dead band setting on the DRV.

rgds,

ramesh

Hi Ramesh,

You guessed correctly, I see them on CCS graph window.

I was also concerned about dead band and therefore increased it in software but could not get rid of spikes.

And you are right, if the motor is disconnected there should not be any such noise on current signal, but in my case there is.

Which makes me think whether DRV8301 is well configured on BOOSTXL module,  because I have another kit DRV8301_HC_EVM Rev D, which does not show

such noise.

I do not have different DRV board but if you know of one which is good on your side, why don' you send me to try here. I would definitely pay for it.

Thank you very much.

Best Regards,

Ljubisa

Hi Ljubisa,
Since you have a 'noise free' DRV EVM kit, you probably don't need anything from me I guess.

Btw, the original code sends out three PWMs to the DRV board and the dead band is taken care of by the EVM. But you can tweak the code to send out 6 PWMs with deadbands inserted. Make sure you get the right signals to the DRV connector before testing the inverter.

rgds,
ramesh

Hi Ramesh,
It appeared that I was not precise in explaining about having another EVM kit.

I have so called F2869 - DRV8301 -High-Current Kit from TI and that kit having the same DRV8301 does not produce any spikes.
So my point is that something must be wrong with my BOOSTXL-8301 module, therefore if you have one which does not show current spikes, will you send me and I will reimburse you.

Regarding 6PWM operation, I have done that but did not get better results no meter how long dead time I insert.
Thank you very much.

Best Regards,

Ljubisa