• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 47 subscribers
  • Views 1267 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.

OPA228 power filter noise problem

Vlad Sergeyev
Prodigy 120 points
Other Parts Discussed in Thread: OPA2228, OPA2227

Hi,

We use power filter based on OPA2228 to power the high sensitivity transimpedance amplifier. The filter is shown in the picture below. Since the transimpedance amplifier has the input stage based on discrete transistors, PSRR is low. Any noise of the OPA2228 filter is directly propagated to the output of the amplifier. The input and power voltage of the power filter ±11V; the current supplied to the amplifier is 8-10 mA. Total capacitive load of the filter is 0.2 uF on each rail.

During testing the excessive RMS noise (tens of  uV) was found on the output of the transimpedance amplifier. The reduction of R13, R15 from 33R to 5R improved (reduced) RMS noise by 40%. Shorting R13, R15 causes the power filter stability problem. The derived conclusion is that excessive noise comes from the power filter based on OPA2228.

PSpice simulation does not show any of these effects.  Since the device is going to mass production, there is still uncertainty about the validity of the improvement, schematic correctness, and the choice of the opamp for the filter. Could you please advise?

  • 0
    TI__Guru* 93105 points

    Hello Vlad,

    My colleague, Tim Green, and I have been analyzing your circuit and believe the noise anomaly you are observing is actually due to circuit instability - oscillation. The present circuit configurations have insufficient phase margin. This is aggravated by the fact that the OPA2228 is not unity gain stable. It is specified for a closed loop gain of 5 V/V, or greater. In the current configuration the gains are less than 5 V/V for each stage.

    Our suggestion is your replace the OPA2228 with its unity gain stable counterpart the OPA2227. That will improve the phase margin, but the circuit will still only be marginally stable at best. We then suggest that you add some external compensation to the circuit to improve the phase margin of each stage. We analyzed several different approaches and found that adding a series 2 ohm resistor, plus 150 nF capacitor from the output pins to ground, adds a pole and zero to the response that reduces the bandwidth the stages and substantially increases the phase margin. The series RC network would be added from the output pin of each stage to ground, before the 33 Ohm series resistors.

    Let me know what you find.

    Regards, Thomas

    PA - Linear Applications Engineering