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.

UCC28070-Q1: Multiphase Operation of two UCC28070-instability issues

Part Number: UCC28070-Q1
Other Parts Discussed in Thread: UCC28070

Hello,

I am currently working on a PFC with following specs:

  • 230VAC input
  • 425V output
  • P_total=8kW equally shared between two PFC-controllers (4 interleaved phases with 2kW each).
  • f_sw= 65kHz

In my first setup, I ran 4kW on a single controller smoothly. Load jumps went smoothly and the overall performance was really good without making any adjustments from the excel sheet provided.
I now want to go up to 8kW with two UCC28070 in multiphase operation from pg 18 of the datasheet and I am running into some issues.

My setup is the following:

  • 130kHz clock from a signal generator with 200ns pulse width, 180° phase shifted as recommended
  • adjusted value of R_RT with factor 1,1 due to external synchronization
  • adjusted value R_DMAX with factor D_SYNC due to external synchronization
  • CDR to V_REF1/2, RDM to clock
  • schematics follow pg. 18
  • for shared signals I always equally divided resistors and capacitor values and placed them close to the ICs
    (so eg. for V_SENSE, the lower resistor of the divider is now placed with (2xR and 1/2C at IC1 and IC2)

I was able to run 3 phases simulatenously, 4 give some instability issues. Theses occur in all phases simultanously, so the current sharing appears to be working fine. I have two theories and would like to know your thoughts or any other issues that you might be aware of.

1. What is the impact of external synchronization on current loop gain exactly? It is written on pg. 16 of datasheet:

It must
be noted that the PWM modulator gain is reduced by a factor equivalent to the scaled RRT due to a direct correlation between the PWM ramp current and RRT. Adjustments to the current loop gains should be made accordingly. 

So far I didnt make any changes to the current loop. Could you explain how exactly these changes are supposed to be made?

2. Does R_IMO need to be adjusted if working with two controllers? I accidently changed the value, and received very different behavior.

Would be very grateful for any ideas or advice

Thanks

Julia

 

  • hello Julia,

    for multi phase application, the SS/VAO/ IMO should be re calculated according to 1/n, here n is the controller number.

    in the page 18,4 phase interleaved, the Rimo should be 1/2 value compare with two phase interleaving.

    for current loop, from the datasheet, the application should increase the current loop bandwidth by factor 2, means you need increasing the current loop bandwidth to keep the same performance.

    let me know what you find, thanks.

  • Hi David,

    thanks for your reply.

    I am not exactly sure whether my strategy was well explained. My situation is the following: I used the excel sheet to calculate the values for P_out=4kW (which means in my case 1 UCC-IC with 2 interleaving phases).

    I am now using 2 UCC-ICs with 4kW each (still 2kW per phase), which gives me a total of P_out=8kW.
    I so far just copied the resistor and capacitor values from previous design, because each controller still needs to control the same power within their designated phases.




    From your comment, I understand and agree that VAO needs to be adjusted - because now I effectively have double the amout of phases - therefore my system will be able to react much faster concerning voltage control.

    You also mentioned SS-pin. I understand that point - but this is not giving me any issues currently.

    But RIMO pin - I do not understand. Since each controller still controls the same power (so current per phase is the same as previously). I do not understand why I need to change the value. Could you explain that to me? R_IMO from page 22 of datasheet is depending on

    • 1/n*I_in(pk) - n now 4 instead of 2, I_in,new(pk) now 2*I_in,old(pk) - so the same as previously
    • R_S - still the same
    • N_CT - still the same
    • I_IMO(max) - still the same

    Concerning the current loop bandwidth - this is now set to 1/10th of f_sw. Do I understand you correctly, that I now should aim for 1/5th of f_sw?


    I will have a look into voltage loop now, and maybe this will already help me.
    Thank you very much already

    Julia

  • Hi Julia,

    since  Imo pin on each UCC28070 is connected together, and I_imomax is a constant value. from your description, there are one Rimo on each UCC28070 controller, means that the two Rimo are in parallel. And in page 18 datasheet, there is only one Rimo for the two controllers working at four phases interleaving..

    My understanding of page 18  is single Rimo should be 1/2 value for two controllers in parallel. if you already place two Rimo on each resistor, then, the total value is already 1/2 .

  • Hi David,

    thank you very much for your answer.

    I actually received a stable operation point for my circuit yesterday by only adjusting VAO-pin like you recommended earlier. In order to do that, I calculated V_ripple for full load (8kW) and adjusted C_PV, R_ZV, C_ZV according to the formulas provided in the application notes.

    But I did not test at full load yet- because there are some adjustments I need to make to the set up. So the impact of R_IMO might not have been visible just yet. But I understand your point about R_IMO. Thanks for clarification. I will keep this in mind and update as well.

    So far, I wasn't able to follow your thoughts on the last part you recommended -  to change the current loop bandwidth. I would be very thankful if you could point me to the right direction. From my understanding the current loop does not change depending on the number of phases operating. Why do you recommend to increase the bandwidth?

    Best regards

    Julia

  • Hi Julia,

    for the current loop that  is based on my previous experience. if you have a chance to test the current loop, I would suggest to test it or evaluate the current loop based on your PFC performance like Harmonic and Ithd/PF.

  • Hi David,

    thanks for your reply. I probably won't be able to measure the current loop directly due to necessary equipment. But I'll keep the bandwidth in mind and analyse harmonics. For now I think this issue is solved and you can close the topic. I will open a new one, in case there are any other points, that come up.

    Thanks a lot for your support!

    Julia