LAUNCHXL-F28379D: Issue with Matlab-generated code controlling the ePWMs

Francois Charlot

Hello,

My customer has started with Matlab-based control code for a digital power system based on F28379D. I have their .slx source code that I could send separately along with some screenshots.

The issue is the PWM outputs do not behave as expected at power up. Upon a hot reset (USB power cycle), the phase relation of the PWM outputs is not as expected; and upon a hot reset (CCS start button), the PWMs may start with a positive or a negative edge.

Is there an application note or coding tips that describe how to properly initialize the PWMs so the above does not happen? Alternatively, I may provide source code for analysis and screenshots. Thank you.

Best regards,
François.

over 7 years ago

0 Yanming Luo over 7 years ago

TI__Guru** 116920 points

You might have a look at the following link about solutions for C2000 MCU’s.
e2e.ti.com/.../484814

And the following link to the Mathworks ePWM documentation page.
www.mathworks.com/.../c280xc2802xc2803xc2805xc2806xc2833xc2834xf28m3xf2807xf2837xdf2837xsf28004xepwm.html

Or you might post the configuration code for the ePWM module, we will assign the appropriate expert to answer this question.

0 Devin Cottier over 7 years ago in reply to Yanming Luo

TI__Guru 60865 points

Hi François,

Were you able to get help from Mathworks for this issue? Do you have any questions about the ePWM module iteslf (and not the simulink code controlling it in this case)?

We'd expect it to be possible to get a consistent power-up polarity of the module.

C2000™︎ microcontrollers

C2000 microcontrollers forum

LAUNCHXL-F28379D: Issue with Matlab-generated code controlling the ePWMs