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.

OPA209: phase shift and phase margin

Part Number: OPA209

Hi team

 

  1. In the datasheet of OPA209, the frequency response curve shows that the phase shift of fc is about 80 degrees. But in the parameter table, there is a column that says that the phase margin is also about 80 degrees.
  2.  According to my understanding, the phase margin should be equal to 180 degrees minus the phase shift of fc. Why is the phase margin directly equal to the phase shift of fc in the datasheet of OPA209?

        

    

Best Regards

Wesley Huang

  • Hi Wesley,

    the phase margin is the phase shift of OPAmp at the frequency where the gain equals 0dB. So 80° phase margin is correct.

    Kai

  • Hi Wesley,

    There is a lot of information about op amp phase margin on the web. Here's a link to a TI applications report that presents phase margin on the first page:

    https://www.ti.com/lit/an/slyt087/slyt087.pdf

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • Hi Thomas,

    Thanks for your sharing.

    According to the document you share to me, the plot of the OPA209 is made by phase shift so the phase margin of OPA209 should be 180 - 80= 100 degree. Do I understand correctly?

    Best Regards

    Wesley Huang

  • This had apparently come up before, I had this little analysis in my files, 

    0830.OPA209 BW investigation using 2011 TINA model in the V11 library.docx

    And, for more detail on gainbandwidth product confusion, I had written this articl, 

    www.planetanalog.com/.../

  • And, back then I had not been adding that extra Cload noted in the specs. Not sure why that is there on these parts - emulating probe, parasitic board C,l or trying to extend gain of 1 closed loop BW by dropping phase margin - in any case,that does drop the simulated phase margin to 75deg with 18MHz LG=0dB

    Going back to that F-3dB extension plot at 75deg phase margin, says we should see about a 1.35*18Mhz = 24Mhz gain of 1 closed loop if we add that Cload, 

    Actually get 28Mhz in simulation, oddly, there is no SSBW plot in the datasheet, so this is all we have to go on, 

  • Hello Wesley,

    The OPA209 gain/phase across plot frequency as shown in the Figure 9 image that you included indicates that the gain roll-off has some subtle deviations from the ideal -20 dB/dec (first-order). This is most evident in the phase response between about 300 kHz and 3 MHz where the trace is deviating about some from that of an ideal first-order phase response. That tells there are additional poles and zeros influencing the overall gain/phase response. It results in the close, but not exact -20 dB/dec roll off rate at all points along the gain curve. Michael's very informative Planet Analog article provides information on the non-idealities that can contribute to small unity-gain bandwidth cross point differences. Clearly, from the phase variations from ideal the OPA209 designer carefully manipulated the internal stages to maximize the op amp's bandwidth and phase margin. That is a common practice.

    Our Precision Amplifier datasheets indicate a phase margin based on the phase remaining before when the open-loop gain curve passes through 0 dB (1 V/V).

    Since the OPA209 datasheet curves, including Figure 9, represent Typical Characteristics, it is best to also refer to the Electrical Characteristics table for the phase margin. That table lists the Phase margin (Φm,) as 80°, for the RL = 10 kΩ, CL = 25 pF. That is the typical phase margin I would use for an OPA209.

    Regards, Thomas

    Precision Amplifiers Applications Engineering