strange trend of pwm

andrea canestrari

Other Parts Discussed in Thread: DRV8301

goodmorning everyone

I'm using matlab / simulink to generate code to drive a module pwm, but I have a strange pattern of response that I read in the output with my oscilloscope.
I have a tooth 10 of us who do not know where to be born, in the module configuration pwm I disabled dead band, and the maximum bandwidth of the predriver is 500ns.
any of you know what 'is the problem and the solution to be adopted?
thanks

over 11 years ago

Gautam Iyer over 11 years ago

Guru 192875 points

Hi Andrea,

That's odd! What hardware are you using? Any chance of it to produce such kind of a glitch?

Regards,
Gautam

Brett Larimore over 11 years ago in reply to Gautam Iyer

TI__Guru 57255 points

Hi Andrea,

I agree with Gautam.

You appear to be viewing the PWM signal after a ~12V FET driver. Can you examine what the PWM looks like before the driver?

Thanks,
Brett

andrea canestrari over 11 years ago in reply to Brett Larimore

Intellectual 420 points

hello
I also thought it was a problem of transistors, but if I measure the signal pwm upstream of the transistor I get the same trend, in particular the first yellow is the output of the transistor with a maximum voltage of 12V (VBAT), the second is module output pwm microcontroller f28069 have the same trend 10 us, I emphasize once again I have not configured any dead band in the form of pwm simulink

if I increase the frequency of the PWM are now at 2.5 kHz, if the change to 25 kHz this trend worsens, what the problem is, how can I solve it?

Gautam Iyer over 11 years ago in reply to andrea canestrari

Guru 192875 points

Can you check at the pin? ie 3.3 V logic

Brett Larimore over 11 years ago in reply to Gautam Iyer

TI__Guru 57255 points

Hi Andrea,

As an experiment, what happens if you remove the transistor from the board? If you do this, does the PWM waveform coming from the C2000 MCU look clean then?

Thank you,
Brett

andrea canestrari over 11 years ago in reply to Brett Larimore

Intellectual 420 points

hi brett,

I use the test card DRV8301-HC-C2-KIT with ISO f28069, I think it was certainly tested by TI that it functions without encounter any difficulties, at least I hope.

but then what sense does it remove the transistor, in the middle between the micro 28069 and mosfet there is the pre-driver drv8301

thanks all

Brett Larimore over 11 years ago in reply to andrea canestrari

TI__Guru 57255 points

I meant to remove the driver. The intention of doing this is to try and figure out the PWM coming out of the C2000 is bad or if it's only bad when combined with whatever is loading it.. Currently, I cannot tell if your issue is with the C2000 or the DRV chip.

I wasn't aware of the hardware you were using previously.

Gautam Iyer over 11 years ago in reply to andrea canestrari

Guru 192875 points

Hi Andrea,

Core motive behind all this is to check whether mcu is the culprit or your driver circuitry. Hence we need to know the quality of PWM signal from controller by isolating the hardware connected to it.

Regards,
Gautam

C2000™︎ microcontrollers

C2000 microcontrollers forum

strange trend of pwm