This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hello,
I am running an ac-dc converter, which uses a full bridge inverter and full bridge rectifier. All these four half bridges (two inverter half bridges and two rectifier half bridges) in the converter need to be operated at 50% duty cycle, and the phase shifts between them should be modulated according to the line cycle voltage to achieve the PFC operation of the converter. I am using ePWMs 1 - 4 for this purpose and loading the TBPHS value of these ePWMs from the lookup table, which stores phase shift values with respect to line voltage. The below figure shows the switch node voltage waveforms of all the half bridges (1 through 4), which are driven by ePWMs 1-4, respectively.
Phase shifts of these half-bridges 2 to 4 vary in the following ranges with respect to half-bridge-1:
Half-bridge 2: 80 degrees to 180 degrees
Half-bridge 3: 60 degrees to 265 degrees
Half-bridge 4: 220 degrees to 270 degrees
Half-bridges 1,2 and 4 are switching properly throughout as expected, but half-bridge 3 is switching with a duty of more than 50% in some regions over a line cycle, and sometimes phase-shift of this half-bridge is also not as expected. Every time before such weird behavior in the duty of half-bridge 3, I also observed the period of half-bridge 3 being longer than the expected value for one cycle, as shown below.
Any comments or suggestions to mitigate this issue are highly appreciated.
Thanks.
Hi Dheeraj,
Every time before such weird behavior in the duty of half-bridge 3, I also observed the period of half-bridge 3 being longer than the expected value for one cycle,
It could be possible that when the phase is changed an action qualifier event is missed. What are the counter compare value/s used for EPWM3? If you are setting the phase shift value to something larger than the counter compare value then you need to check for this in your application and change the value of CMPX to be TBPHS + 1, just for that cycle, to ensure you don't miss an event tied to that compare value.
Best Regards,
Marlyn
Hi Marlyn,
Thanks for your reply. My CMPA value for EPWM3 is 166 (i.e., 50% of the TBPRD). I understand that when the TBPHS value is larger than the CMPX value, compare skip events occur. I already took care of that in my code. But, the waveforms of EPWM3 don't look like skip events, duty is around 75% (which is not what I want), and the period is surprisingly more for one cycle. I am not able to understand what is causing this strange behavior.
Dheeraj,
Is EPWM3 setup any differently than the other epwm modules or is it just the phase that is being applied?
Best Regards,
Marlyn