• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 48 subscribers
  • Views 1704 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.

OPA 656 Oscillation

wnefwkn dlvne
Prodigy 40 points

Other Parts Discussed in Thread: OPA656

I am using OPA656 in non-inverting configuration, with a gain ~= 32. If the input is connected to a BNC fixture (a bnc cable to a signal source, or a bnc cable with a bnc short), the amplifier oscillates with a frequency of 25MHz. Any suggestions what might be going on?

Thanks,

Mohit

  • 0
    TI__Genius 12631 points
    Hello, Its possible that there is excess capacitance on the inverting node or the output node that is leading to instability. Another issue could be insufficient power supply bypass capacitance. Can you please send me the schematic, layout and if possible a photograph of the setup?
    -Samir
  • 0 in reply to Samir Cherian
    Prodigy 40 points

    Hi, thanks for your reply. My OPA656 setup is a simple dc-coupled non-inverting configuration. This is in a metal box which has a BNC connector for amp input and a BNC connector for output.  I fixed the OPA656 oscillation problem by moving the input BNC connector to the amplifier proto-board (from the box). Sorry, I do not have a picture to share right now. But it seems that there was some problem if the input was connected through the BNC mounted on the metal box (faraday shield box). (I suspect it was related to ground, but it could also be the distance of the input to BNC).

    However, I have another question. This is simple dc coupled non-inverting setup (R1=50Ohms, R2=1500Ohms, G = 1+R2/R1). If I connect the non-inverting input to ground through R3=90kOhm, the amplifier oscillates. Do you know what might be going on?

  • 0 in reply to wnefwkn dlvne
    TI__Genius 12631 points

    Hello,

    There could be board parasitics, coupling from the amplifier output to the non-inverting pin, which is now very high impedance and is prone to pick up spurious stuff in this configuration. Can you please either change the 90 kOhm resistor to something a lot smaller, say < 1kOhm and see if the problem goes away. Alternately put a 1uF cap. in parallel with the 90kOhm resistor to create a low impedance path from the non-inverting terminal to GND at high-frequencies.

    -Samir

  • 0 in reply to Samir Cherian
    Prodigy 40 points
    Hello,

    Thanks for the reply. Yes, the oscillations went away with smaller resistors (even 40kOhms). Our goal is to make high frequency measurements from medium to high impedance sensors. Therefore, we don't want to modify the frequency reponse much by adding capacitors, and understanding the limits of the layout and of OPA656 are important to us. Is there a way to test whether the coupling is between non-inverting input and the output?

    -Mohit
  • 0 in reply to wnefwkn dlvne
    TI__Genius 12631 points
    Mohit, Can you eliminate the 90kOhm shunt resistor to GND completely and just add a small (say 50-Ohm) series resistor between the sensor output and amplifier non-inverting input?
    -Samir
  • 0 in reply to Samir Cherian
    Prodigy 40 points

    Samir,

    The 90kOhms resistor is connected between non-inverting input and ground. This is just to characterize my board. For your reference, I have attached a picture of my board. It is  a dc coupled non-inverting amplifier. I am trying to figure, what is causing oscillations with 90kOhm input reistor (the oscillations go away with 40kOhms input resistor). Hope you can access the PDF file that has 2 pages with 2 images.opa656_board.pdf

    Thanks,

    Mohit

  • 0 in reply to wnefwkn dlvne
    TI__Genius 12631 points

    Hello Mohit,

    Bread-boarding and socketing a high-speed part is a no-no. Please see section 11 of the datasheet for layout recommendations. An unpopulated evaluation module can be purchased from TI or local distributors for $5. Please see below. Using leaded resistors and bypass capacitors is also not recommended since they carry too much parasitic inductance.