ADA4807-1 had output oscillation on Non-inverting Topology

Hi,

I had designed a non-inverting amplifier by ADA4807-1, following picture is the simulation result of LTspice , 

I used actual IC to verify on the PCB, but the output waveform of ADA4807-1 is 13.26MHz Oscillation waveform. My schematic of PCB is the same as LTspice. I use 100nF X7R & 6.8uF X5R on ADA4807-1 bypass capacitor. Following is the Oscilloscope waveform. CH1 is ADA4807-1 output, CH2 is ADA4807-1 input.

   

Why dose the ADA4807-1 oscillation? How do I solve this issue? Please help me clarify this issue, Thank a lot.

  • +1
    •  Analog Employees 
    on Jan 15, 2021 9:50 PM 1 month ago

    Hi Klex Lin,

    ADA4807-1 is a 180MHz, 1mA supply current device. You should make sure you're doing a high frequency layout board (ground plane, short leads, etc.) to avoid creating excessive parasitics that would cause oscillations. Also, I recommend you lower your feedback from 15k to avoid creating phase shifts from this large resistor and any parasitic capacitances you may have. You can lower RF to 2k (and RG along with it) to see.

    It may help to eliminate the 6k in series with your non-inverting input. There is significant input bias current (uA) that won't work well with such large resistor. You may also want to load the output more heavily (say 500ohm) to see if that helps. Also, when probing with a scope, make sure you have some resistive isolation to the ADA4807-1 output pin (to avoid scope capacitance giving you issues). May want to use the scope 50ohm terminated input as a way to measure instead of a scope probe.

    Regards,

    Hooman

  • 0
    •  Analog Employees 
    on Jan 16, 2021 8:55 PM 1 month ago in reply to Hooman

    Hi Klex Lin,

    Also, please make sure that if the generator that you've shown (V2) is at a distance from the ADA4807-1 stage and there is an interconnect (coax) cable involved, that the generator impedance (50ohm for most lab generators) is terminated in 50ohm near the ADA4807-1 stage before going to the non-inverting input. If you have a coax interconnect and you're not terminating it near the stage, that could be an issue in your setup.

    Regards,

    Hooman

  • Hi Hooman,

    I change RF to 1.5K ohm (and other R along with RF) and clear path issue between source generator & non-inverting input. The oscillation issue of ADA4807-1 had been solved. I really appreciate your support.

    Best Regards 
    Klex Lin

  • 0
    •  Analog Employees 
    on Jan 26, 2021 7:31 AM 1 month ago in reply to Klex Lin

    Hi Klex Lin,

    That's great that a lower RF resistor resolved your oscillation issue with ADA4807-1!

    We all need some good news these days.

    Regards,

    Hooman