• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 580 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.

TPS62130A: Phase margin in SLVA465A

Hirokazu Takahashi
Expert 8420 points
Part Number: TPS62130A
Other Parts Discussed in Thread: TPS62130, TPS62135, TPS61230,

Dear support team

I was able to find a bode diagram of TPS62130 at the application note of SLVA465A.

http://www.ti.com/lit/an/slva465a/slva465a.pdf


In case of Figure 3, is it correct that the phase margin is 90deg?
How should we think about the gain margin?

Best Regards,
Hirokazu Takahashi

  • 0
    TI__Guru**** 150760 points

    Hi Takahashi-san,

    Yes, the phase margin is typically high.

    With the high loop bandwidths near the switching frequency, it gets difficult to get measurable data to get the gain margin.

    As well, please see TPS62135/6 as newer, smaller, more accurate devices than TPS62130.

  • 0 in reply to Chris Glaser
    TI__Expert 8420 points

    Hello Chris-san

    Thanks for reaching out.
    I have two question regarding phase and gain margin.

    Q1. Is it possible that we don't care about the gain margin?　Do you think that there is no concern such as oscillation?
    Q2. Do you have any required value of the phase margin as guide line? Is it enough to be more 45 deg?

    BTW, the customer has proceeded to adopt TPS61230 already. So they can't replace it to other products.

    Best Regards,
    Hirokazu Takahashi

  • 0 in reply to Hirokazu Takahashi
    TI__Guru**** 150760 points

    Hi Takahashi-san,

    No, I have no guidelines on gain margin.  I do not use this value.

    Your customer may have a phase margin requirement.  If not, 30 degrees (corresponding to 3 rings in the transient response) is what we used for this device here: http://www.ti.com/lit/an/slva463a/slva463a.pdf

  • 0 in reply to Chris Glaser
    TI__Expert 8420 points

    Hello Chris-san

    Thanks for your support.
    Let me confirm about your comment "No, I have no guidelines on gain margin. I do not use this value."
    Does it mean that we won't need to care about gain margin for feedback stability?

    Best Regards,
    Hirokazu Takahashi

  • 0 in reply to Hirokazu Takahashi
    TI__Guru**** 150760 points

    Hi Takahashi-san,

    TI does not recommend any specific phase or gain margin values.  This is for our customers to do in their company's design requirements.  10 db is a typical value that I have seen.

    Here is a paper that discusses phase and gain margin: 

    This one shows a real example on slide 15 and 16: https://www.ti.com/seclit/ml/slup385/slup385.pdf

    Speaking for myself, I rarely look at the gain margin.  It is also difficult to measure, since the phase 0 crossing occurs at a very high frequency with most TPS62xxx devices.  These switch above 1 MHz and have very high crossover frequencies (and thus have even higher phase crossing frequencies).  Some of the measurement equipment that we have used in the past is not very accurate at frequencies near or above 1 MHz (where the phase crossing might occur).

  • 0 in reply to Chris Glaser
    TI__Expert 8420 points

    Hello Chris-san

    Thanks for your support.
    I could find below description in the document that you attached on the previous post.

    "Sufficient phase margin is required to prevent oscillations. A phase margin of 45 degrees or greater is the design goal. A gain margin of –6 dB is the minimum, while –10 dB is considered good."

    But it is difficult to mearsure gain magin at TPS62130A as your mention. How can we make sure to prevent oscillations?

    Best Regards,
    Hirokazu Takahashi

  • 0 in reply to Hirokazu Takahashi
    TI__Mastermind 36801 points

    Hello Takahashi-san,

    If there is a difficulty in measuring the gain margin, check if there is a duty cycle imbalance in the entire Vin and load range refer to slide 15 and 16: https://www.ti.com/seclit/ml/slup385/slup385.pdf). This way the customer will be able to gauge if the converter has enough gain margin. Besides, a good way to measure the converter's loop stability is check the transient load response for any oscillation during load attack and release.

    Best regards,

    Excel