• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 48 subscribers
  • Views 669 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.

opa657 oscillation

Bruce Bauman
Prodigy 10 points
Other Parts Discussed in Thread: OPA657

I am using the OPA657 in transimpedance mode with a 22pF photodiode. I have the anode of the photodiode inputting into the inverting input, non-inverting goes to ground. Supply is ± 5V, decoupled with 0.1uF caps. No matter what feedback circuit I try, the output has a very large, about 3-4V, oscillation. I have tried various combinations of  feedback components, feedback resistors from 10K ohm to 100M ohm, feedback capacitors from 0.2pF to 22pF. Nothing stops the oscillation. Any suggestions? I would like to have the circuit work with a 10M ohm feedback. Is the anode input from the photodiode a problem? I notice the test circuits shown on the OPA657 data sheet have the photodiode's cathode going to the inverting input. If possible I need to have it work with the anode inputting.

  • 0
    TI__Expert 3555 points

    Bruce,

    First of all, it doesn’t matter that your photo diode connection is reversed as compared to the data sheet. With your configuration, the output voltage will always be negative (instead of positive from the data sheet configuration). If you include Vb in your circuit, you need to make this voltage positive. This voltage source is configured in such a way to reverse bias photo diode. When you do this, the capacitor’s parasitic capacitance is reduced.

    Making sure that this bias voltage is implemented correctly would be your first task as you solve this problem.

    If this doesn’t get you to a stable circuit, we need to look at the ac response of your circuit. I believe that your problem is solvable, but I need to know what the value of the parasitic capacitance of your photo diode (Cphoto). With that number, we can determine the high frequency gain (1 + Cin/Cf), where Cin is equal to Cphoto + Cdiff + Ccm. Cdiff is equal to the op amp’s differential input capacitance (0.7 pF) and Ccm is equal to the op amp’s common mode input capacitance (4.5 pF). Convert this gain to dB with the formula 20*log(1+Cin/Cf).

    Once you determine the high frequency gain of your circuit, you can evaluate where that gain intersects the open loop gain curve of the OPA657 (see page 7 in data sheet). That interest point will give you a clue about your stability as long as you look at the phase margin at the interest frequency.

  • 0
    TI__Expert 3385 points

    Hi Bruce,

    Adding to the previous answer, I had a couple of queries regarding your application and photodiode that will help us solve your issue.

    • What is the shunt resistance of the photodiode that is sinking current current from into the transimpedance amplifier? An idea of the dark current may also help us derive this value.
    • What is the bandwidth that you are targeting for this application? In the present configuration you will be able to achieve <100kHz bandwidth.
    • What is the board parasitic induced at the inverting pin input?

    As mentioned by hooman in this post http://e2e.ti.com/support/amplifiers/high_speed_amplifiers/f/10/p/333642/1172344.aspx#1172344 the stability of the OPA657 needs to be evaluated with respect to the open loop gain and the noise gain. The feedback capacipor should be small enough to maintain the minimum gain reguire for opamp stability while making sure the zero caused by the Cin is nuetralized.

    Presently for a bipolar circuit, assuming a shunt resistance much larger than the feedback resistance a feedback capacitance of 250 fF should be enough to stabilize the circuit. 

    Regards

    Pavan K Shetty

    Analog applications associate, HSP amplifiers