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.

OPA855: OPA855 output has a DC bias

Part Number: OPA855

Hi everyone,

I am using OPA855 as a trans-impedance amplifier with the photoelectric conversion. But it bothers me a lot that the output of the trans-impedance amplifier has a DC bias. The schematic diagram isattached below. The details of the problem are as follows.

1.  In order to eliminate the influence of dark current, I removed the APD and just connected the amplifier input (left end of R60) to GND. There is several mV bias at the ouput of U15. After the second (U16) and third (U17) stage of amplification, the bias can increased to several hundreds mV. Why the first trans-impedance stage has such a bias? Is my schematic diagram incorrect?

2.  My amplifiers are powered by ±2.5V. But there is 50 mV difference between the postive voltage and the negative voltage. Does it matter?

3.  Will the input impedance of the inverting and noninverting input of OPA855 influence the output bias? Is the input impedance of the inverting same to the noninverting input?

4.  The circuit has oscillation at some frequencies. Is there any effective way to eliminate the oscillation?

Many thanks

Zhao

  • Hello Zhao,

    Addressing your problems below:

    1. The maximum input offset voltage is at around 1.5mV. After amplification, several mV of bias at the output is within the scope of the device.
    2. This isn't an issue. Note the supply ranges of the OPA855 ranges from 3.3 to 5.25V.
    3. Yes, your input impedance will affect the bias of the amplifier. Even if these are matched correctly, there is still the offset current of up to 1 uA. Along with a 5k impedance, you can get up to 5mV of additional input offset voltage. 
    4. Your first stage will definitely oscillate without an APD, and possibly with your APD connected as well. Have you simulated for the stability of the first stage? What is the input capacitance of your APD? Note that the OPA855 needs a gain of 7 to be stable, otherwise you must use noise shaping techniques with your input and feedback capacitors.

    Best,

    Hasan Babiker

  • Hi Hasan,

    Thanks for your reply.

    1.  It seems the bias voltage of the first stage is inevitable, but it does affect the final output. The second and third stage can totally increase the first stage bias by 50 times. Is there any way to suppress the bias?

    2.  Could you please explain why the first stage will oscillate without an APD? The capacitance of my APD is 0.5pF.

    Regards,

    Zhao

  • Hello Zhao,

    1. You can AC couple the last stage as shown here:  

    2. Actually it looks like you are getting 50 - 51 degrees of phase margin for your first stage, I've attached the stability analysis below:

    OPA855_TIA_Stability.TSC

    It would be best to design for greater phase margin if you are still seeing oscillations. Have your followed all the layout guidelines in the OPA855 datasheet? Specifically are the power and ground planes removed under the device?

    Best,

    Hasan Babiker