• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 38 subscribers
  • Views 436 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.

Phase Margin plots in SwitcherPro

Alexander Grimes
Prodigy 20 points
Other Parts Discussed in Thread: TPS40200

Hi,

I've been looking at the plots for total gain and total phase with SwitcherPro, and the phase plot isn't making sense to me.

I've always read that phase margin is the difference between the phase measured at the point where gain is at unity and -180 degrees.

Now when I look at SwitcherPro's plots, it looks like the program added 180 degrees to all measurements of the "true" phase plot, changing the definition of phase margin to be the difference between the phase measured at the point where gain is at unity and 0 degrees.

Is that what the program is doing? It just threw me off a bit when I saw it, and I want to make sure that's what's going on...because I didn't see that in any documentation, and the plot states "phase" instead of "phase + 180d."

Related to this, I thought gain margin was the amount of gain that could be added to the gain measured when the phase crosses -180, and therefore would be a positive number, and not the measured gain at that point as SwitcherPro indicates.

But perhaps I'm reading these plots wrong?

  • 0
    TI__Expert 3875 points

    Hi Alexander,

    As you noted, it is common practice to associate a phase of 0 with negative feedback and a stable response. A phase of 180/-180 typically corresponds to positive feedback and instability in a conventional operational amplifier system where the output is fed back to the input through the inverting input which adds an additional –180 degree of phase. So, when calculating the phase margin for an op-amp (as you noted), one typically subtracts the phase from -180 at the unity gain frequency.

    The same control theory applies to SMPS designs. The FB pin is typically the inverting input of the error amplifier. The plot in SwitcherPro shows the response, including the 180 phase shift in the error amplifier. So, the phase plot is shifted by 180 degrees. In SwitcherPro a phase of 180/-180 corresponds to "good" negative feedback and a phase of 0 corresponds to "bad" positive feedback. In this scheme, the phase margin is simply the phase at unity gain frequency. This convention is also used on TI data sheets. For example, please take a page 29 of the TPS40200 data sheet. As in SwitcherPro, the plot start at around 140 deg and does down to 0.

    Likewise, I have seen gain margin defined both ways. In many references a positive gain margin denotes a stable system. In SwitcherPro, a negative gain margin denotes a stable system. This is also visible from the Bode plot. When the phase crosses 0, the gain is usually well below 0dB.

    Hope this helps.

    Regards,
    Michael

    Michael Krasnicki
    Analog eLab
    Texas Instruments