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.

OPA2197: Gain vs frequency question

Part Number: OPA2197

I have used 2 opa2197(one chip) in differential amplifier configuration. One's output is connected to the other's input. My closed loop gain is(should be) ~3 for both of them. I have a signal of +-0.34V sinusoidal at the input of the first amplifier. Amplifiers' supply is +-9V. I want the input signal to be 400KHz. At low frequencies like below 50KHz everything seems to be normal. But as frequency goes up to ~280KHz the output gets larger (i.e. gain is increasing); after that frequency the output is getting smaller (i.e. gain is decreasing). What is the reason of this? What can I do?

  • By reducing feedback resistor values, the response greatly improved, nearly OK.
  • Mustafa,
    In order to assist you, it would be much better to draw the circuit instead of describing it. In the low close-loop gains, you may run into so-called gain peaking where the overall gain increases with frequency before it is attenuated by the effective bandwidth of the circuit configuration. If the gain peaking is greater than few dB, this may lead to instability of the circuit.
    Also, in case of using large value resistors, the main culprit behind the peaking might be the interaction of the op amp input capacitance with the input resistor, which creates a zero in the close-loop gain transfer function. At dc and low frequencies, the close-loop gain is defined by the ratio of feedback and input resistor but at the higher frequencies, the gain increases because it is dominated by the ratio of the feedback resistor to the parallel combination of the input capacitance and the input resistor, G(f)=RF/(Rin||Cin). Decreasing the value of the resistors, pushes the zero to higher frequency - beyond the effective bandwidth of the configuration - and for most part eliminates the gain peaking. An alternative approach, would be to add a feedback capacitor across the feedback resistor in such a way that Rin*Cin=RF*CF – this effectively cancels the zero caused by the input capacitance.