• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 49 subscribers
  • Views 2987 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.

Check stability of Differential Amplifier

Chris Smith
Intellectual 825 points
Other Parts Discussed in Thread: OPA365

Hi, I have a circuit that I am using for driving a anti-aliasing RC filter for the input to an ADC.  Our signal conditioning uses the TLC082AI in a differential amplifier configuration as sown in SLOA039A Figure 16.  This circuit has worked well for us as it provides gain, offset, a differential measurement, and good common mode rejection.  My question is related to increasing the capacitor in the anti-aliasing filter to give us the full precision on the ADC as well as increasing the bandwidth.  I'm looking at going from a RC of 50 ohm and 47nF to 1.5 ohm and 400nF.  Another improvement is to move to an amp with higher GBW and less offset voltage such as OPA365.

So my concern is checking the stability of the circuit with the larger load cap.  I'd like to measure phase margin but am a little puzzled at how to properly break the loop and measure open loop gain.  I've broken one loop with the transformer and resistor (K1 and R11), but there are two other loops.  Loopgain is then Vresp/Vfdbk.  Can you suggest a good method for verifying stability or perhaps another way entirely?  The only other thing I've been able to come up with is checking the circuit overshoot during a full scale step or perhaps keeping the peaking of the closed loop transfer function below 3dB but these seem like more crude checks which may not be full-proof.

Thanks,

Chris

  • 0
    TI__Mastermind 25805 points

    Hello Chris,

    Both methods are viable options for checking the stability of the loop.  For the step input, however, I would recommend applying a small signal step as opposed to full-scale.  You can then correlate the overshoot to phase margin.  Considering breaking the loop with a transformer, you may be interested in using a gain-phase analyzer with signal injector.  One option is the Omicron Bode-100 paired with a Picotest J210xA signal injector.  Finally, you may find the stability article series by Tim Green useful.  It is located on http://www.en-genius.net.  Here is a link to part 10 of the series from which you can also download parts 1-9.

     

     

  • 0 in reply to Pete%20Semig
    Intellectual 825 points

    Pete, thanks for the information.  There is some good stuff in that series of articles.  I guess my main concern is that I can't break the circuit down to a single loop as is done in the article.  Or to put it another way there is no one place to break the circuit and measure open loop response.  For instance current through R6 does influence voltage at negative pin of the op-amp.  So opening one loop still leaves that one closed.  As for measuring overshoot I'm concerned that since the overall circuit may not be second order that those assumptions don't hold.  Is it really safe to only analyze the primary feedback loop on the final op-amp?  I don't see any mention of multi-feedback circuits except for the non-iverting noise gain example in part 6 and that is somewhat different than what I have.

    Thanks for your help,

    Chris

  • 0 in reply to Pete%20Semig
    Prodigy 10 points

    hello pete,

       What you have said is 100%  right