• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 47 subscribers
  • Views 154 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.

Original question:

OPA838: OPA838IDBV

OPA838: 91MHz oscillation using OPA838

derek murison
Prodigy 30 points
Part Number: OPA838
Other Parts Discussed in Thread: TINA-TI, OPA836

Tool/software:

Hi, my 2-stage non-inverting amplifier with BW 4MHz has some oscillation of 91MHz. 1st & 2nd stages are set to x10LIN gain. Single supply with 1.85V operating bias voltage. The 91 MHz amplitude is around 130mV. Basic amplification of input waveforms 400kHz-3MHz works. Using a decompensated amp for the 1st time (OPA838).

Tina-TI simulates everything ok. It does have a rising response at 100MHz:

Ideas will be appreciated. 

Thanks

  • 0
    Prodigy 30 points

    Layout

    Solid GND plane below. R8 & C8 added by rework. Maybe feedback cap C14 could be closer to amp?

  • 0 in reply to derek murison
    TI__Mastermind 29515 points

    Hello Derek,

      Thank you for the detailed overview! The spike in simulation does show around 90-100MHz which does indicate oscillation at that frequency.

      As a quick test, would you be able to try the same measurement but with C13 and C14 removed as well as C2 and C4 replaced with 0 ohm resistors? 

    Thank you,
    Sima

  • 0 in reply to Sima Jalaleddine
    Prodigy 30 points

    Sima, thanks for the reply. I have found that C13 & C14 have an impedance of around 100R @ 91 MHz. At this frequency the ac gain is unity which decompensated amps don't like causing instability. Removing C13 & C14 stop the oscillation but greatly widens the BW which I don't want because I need to attenuate any noise above 4MHz. I have reverted to using OPA836 with the same C14 & C15 to give me the desired BW. It works well now. C2 & C4 are required to allow the mid rail operating bias being a single supply design.

    Thanks
    Derek