• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 84 subscribers
  • Views 548 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TMS320F280039C: SFRA and Compensation Designer using issue

Cherry Zhou
Mastermind 22235 points
Part Number: TMS320F280039C
Other Parts Discussed in Thread: SFRA

Hi team,

The customer has the following questions when using SFRA and Compensation Designer to test the feedback factor of the current circuit:

1) Using the BuckBoost circuit, when scanning with SFRA, is scanning in Buck of open/closed loop, equivoltage, or Boost mode? 

2) Is ISR Freq the switching frequency of PWM in Compensation Designer? 

3) With 2P2Z compensation, the customer finds that after compensation, Folg_cf, Gain Margin, Phase Margin are NA, but it shows stable Loop. But the output of the actual test closed loop is stable: 

With KDC turned up, Folg_cf and gain margin, Phase margin have values and are also shown as stable Loop. But the output of the actual test closed loop is unstable. Generally speaking, if there is value, it should be more stable than NA, is that right? 

4) Why does Folg_cf, gain margin, and Phase margin all show a stable Loop when they are NA? 

5) The output of the test closed loop is stable without load and the compensation curve is as follows:

However, when 1A is loaded, the output will jump erratically, causing the power supply to also enter CC mode. Is it related to compensation? 

6) Is Plant Mode referred to as closed loop in SFRA? 

Could you help check this case? Thanks.

Best Regards,

Cherry

  • 0
    TI__Prodigy 410 points

    it's re-assigned to the right expert to respond back to you. 

  • 0
    TI__Genius 15935 points

    Hi Cherry,

    Cherry Zhou said:
    1) Using the BuckBoost circuit, when scanning with SFRA, is scanning in Buck of open/closed loop, equivoltage, or Boost mode? 

    The question is a little bit vague. The SFRA ifself is independent of which mode you would like to measure. This depends on which condition customer would like to test the bandwidth.

    Cherry Zhou said:
    2) Is ISR Freq the switching frequency of PWM in Compensation Designer? 

    No, it is different. The ISR frequency is the control loop freq. It is different from switching frequency.

    Cherry Zhou said:
    3) With 2P2Z compensation, the customer finds that after compensation, Folg_cf, Gain Margin, Phase Margin are NA, but it shows stable Loop. But the output of the actual test closed loop is stable: 

    Are they using our reference design? or they implement the SFRA by themselves inside their own project? This usually happens when the curve is not typical. It may mean you have a very large buffer to make the system stable.

    Cherry Zhou said:
    6) Is Plant Mode referred to as closed loop in SFRA? 

    The question is not clear for me. when running closed loop, SFRA is measuring the "open loop transfer function". When running open loop. SFRA is measuring plant model. 

  • 0 in reply to Chen Jiang ti
    TI__Mastermind 22235 points

    Hi,

    Thank you for the support.

    Chen Jiang ti said:
    The question is a little bit vague. The SFRA ifself is independent of which mode you would like to measure. This depends on which condition customer would like to test the bandwidth.

    So it  just need to let the program take action? If scanning under Buck, and the feedback was stable, would it be stable in Buck, equal voltage, and Boost? 

    Chen Jiang ti said:
    No, it is different. The ISR frequency is the control loop freq. It is different from switching frequency.

    May I know how to determine the value of the ISR frequency? 

    Chen Jiang ti said:
    Are they using our reference design? or they implement the SFRA by themselves inside their own project? This usually happens when the curve is not typical. It may mean you have a very large buffer to make the system stable.

    Yes they are using our design, but they're trying to adjust the parameters so that Fog_cf, gain margin, and Phase margin are within a range of values, not NA, and show a stable loop, but the actual output is unstable. Can stable Loop be used as an indicator of stability? 

    Chen Jiang ti said:
    The question is not clear for me. when running closed loop, SFRA is measuring the "open loop transfer function". When running open loop. SFRA is measuring plant model. 

    When using the SFRA, Open Loop/Plant Mode Select needs to select Open Loop if they want to measure a closed loop,  if they want to measure an open loop, just select Plant, is that correct?

    Thanks and regards,

    Cherry

  • 0 in reply to Cherry Zhou
    TI__Genius 15935 points

    Hi Cherry,

    Cherry Zhou said:
    So it  just need to let the program take action? If scanning under Buck, and the feedback was stable, would it be stable in Buck, equal voltage, and Boost? 

    Yes, most probably if your control function is the same since SFRA is integrated with your control function. 

    Cherry Zhou said:
    May I know how to determine the value of the ISR frequency? 

    This depends on the control method and topology. For Buckboost, I think you can use the same freq as switching freq if possible to achieve the best performance. 

    Cherry Zhou said:
    Yes they are using our design, but they're trying to adjust the parameters so that Fog_cf, gain margin, and Phase margin are within a range of values, not NA, and show a stable loop, but the actual output is unstable. Can stable Loop be used as an indicator of stability? 

    yes, if it shows stable loop, this can be a indicator of stability because this is based on the basic control theory.

    Cherry Zhou said:
    When using the SFRA, Open Loop/Plant Mode Select needs to select Open Loop if they want to measure a closed loop,  if they want to measure an open loop, just select Plant, is that correct?

    if they want to measure the open loop transfer function, they need to select open loop and running under closed loop operation. If they want to measure the plant transfer function, they need to select plant and running under open loop operation.

    Regards,

    Chen