• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 82 subscribers
  • Views 1668 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Implementing Phase Shift Modulation between Four EPWMs: 2808 vs 28335

Amit Jain53087
Prodigy 130 points

Hello, 

  I am working on a power converter that uses four phase shifted 50% duty ratio signals, i.e. phase shift of three signals has to be controlled with respect to the first one.  I composed and successfully tested the required code for generating these signals on the F2808. Configuration of the EPWM modules is shown in the diagram below (up count only with CMPA=Period/2, Active high complementary outputs, phase loading enabled on EPWM2,3,4).  I need the hokey sync scheme below to ensure that the phase shifts remain within bounds. In the simplified version alphaP = alphaS =0 , which leads to :

        EPWM2A 180 out of phase with EPWM1A

        EPWM3A lagging EPWM1A by Phi_f

       EPWM4A 180 deg out of phase with EPWM3A

 

[Note: pi = Period/2; 3pi/2 = 3/4*Period etc]

Now, I am beginning to use the F28335 and the same code does not work as expected. I get two phase shifts correct (EPWM1,2,3A/B) while the third one (EPWM4AB) is off.  After significant trial and error with the EPWM settings I came to the following strange conclusion about the F28335 EPWM modules.  Setting the SYNCOSEL for EPWM2 to CTR=CMPB sets the SYNCIN of EPWM4 to SYNCO of EPWM1.   

So I reconfigured the EPWMs to the following scheme assuming that SYNCIN of EPWM4 is the same as SYNCO of EPWM1.

Now the output of EPW5 is what I expect it to be. Essentially, in the new scheme I am using EPWM4 to generate the phase shift I was expecting from EPWM3 in the old scheme and then I am using EPWM5 to generate the actual PWM signals for S3s and S4s. 

 

I find it hard to believe that the DSP actually behaves like this. What am I missing? Are there any changes in the EPWM modules going from 2808 to 28335?

Best Regards,

Amit

------------------------------------------------------------------------
Amit Kumar Jain, PhD
Senior Engineer, Power Electronics
Peregrine Power LLC
27350 SW 95th Avenue, Suite 3022
Wilsonville, OR 97070. 
------------------------------------------------------------------------

 

 

 

 

 

 

 

 

  • 0
    Expert 1070 points

    Amit, 

    On page 27 - 29 of SPRUG04A –October 2008–Revised July 2009, you may find an answer to the question. Notice that the daisy chain scheme (scheme 1) you described is the scheme "applies to the 280x, 2801x, 2802x, and 2803x devices". The scheme used by "all other devices" (I would assume 28335 included) is on page Scheme3 which has the chain 1-2-3, 1-4-5-6, 1-7-8-9.

     

    so i guess it would be easier for your application to use 1-4-5-6 or 1-7-8-9 instead of 1-2-3 + 1-4-5.

     

    hth

     

    gz 

     

    btw,  it is impressive to be able to figure out the chain sequence by experimenting it out.. 

  • 0 in reply to Leong
    Prodigy 130 points

    Leong, 

    Thanks a lot! 

    The 1-2-3, 1-4-5-6 scheme would certainly explain what I have observed. I was under the impression that F28335 are an upgrade from F2808 so the same code should generate the same PWM signals. 

    Regarding experimenting to figure out the sequence. It took me a really long time.

    I will try out your suggestions. 

    Best Regards, 

     

    Amit.