- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
Original question:
TMS320F280049: Adding HWPWM TBPHSHR to DAB converter causes loss of waveform symmetry
Hello,
This is related to my previous question on HRPWM TBPHSHR. I am in updown count mode, HR autoconvert is on, HRPE is off, and switching at 200 khz. I am now able to get symmetric waveforms and the edges shift correctly. The problem I am seeing now is the tbphs and tbphshr shift the waveforms in opposite directions. That is, a tbphs to tbphs+1 will move the slave in one direction and a tbphshr =256 (minimum) to tbphshr = 64000 will move it in the opposite direction. This is shown in the figure below and I am setting/verifying the register values in the expression windows.
If all the HRPWM registers do is shift the waveform to the right (both edges for tbphshr) then I can see this behavior being correct and the math needing to be different depending on if you are leading or lagging. However, I didn't see this in the datasheet or online. This question (https://e2e.ti.com/support/microcontrollers/c2000-microcontrollers-group/c2000/f/c2000-microcontrollers-forum/858613/tms320f28379d-usage-information-on-trrem-register-for-periodic-hr-phase-update) says you need to subtract the HR phase for a negative phase shift, but that still wouldn't give the right value as you have to add +1 to tbphs first.
For example, if my understanding is correct and I wanted a leading phase shift of 5.75 I would need tbphs = 6 (not 5) and then set tbphshr = 64000-48000 (roughly) so the correct phase comes out to be 6-0.25 = 5.75. Is this the proper way to do a leading tbphshr phase shift?
Meanwhile a lagging phase shift of 5.75 would be tbphs =5 tbphshr = 0.75*(64000)?
Thanks
