• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 47 subscribers
  • Views 720 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

OPA857 high frequency oscillation

Gerard Visser
Prodigy 165 points
Other Parts Discussed in Thread: OPA857
I am working to design in OPA857 for a transimpedance amplifier application. (Not for a photodiode but rather for an "ion chamber", but the requirements are similar. Detector capacitance is ~2.5 pF and the signal is 10-30 uA and we need somewhat over 80 MHz bandwidth.)
I obtained the evaluation board (rev B) and made some simple tests. In our application we'll use the single-ended output so I configured the eval board to connect that (pin 8) through a 422 Ohm resistor + 47 nF coupling capacitor to the SMA connector J6 on the board. I have no input connected (J2 and R4 open) and the test input is disabled (TEST_SD is jumpered to GND, TEST_IN is floating (R3 removed).
On the output I find a strong oscillation at about 3.7 GHz !! See the following scope photo. The oscillation frequency depends slightly on the capacitance attached at OPA857 input pin. Ref1 shows the case with the 1.5pF capacitor (as eval board is received). Ch1 shows the case of removing that capacitor, only board parasitics for the external capacitance.
scope shot
Anyone have an idea what is going on, why is this oscillating? What can I do? I don't think just filtering it out is the right solution...
  • 0
    TI__Genius 12631 points
    Hello Gerard,
    This is a known issue with the OPA857 and we are working on new Silicon to fix it. We hope to have the new silicon ready for release some time in October this year.
    -Samir
  • 0 in reply to Samir Cherian
    Prodigy 165 points

    Hello Samir,

      Thank you for confirming it. I was afraid I had somehow broken it or was confused. I look forward to the new device.

      Is there a specific ordering code to ensure getting the new version through distribution channels (e.g. Mouser or DIgikey)?

      I guess the existing device is still useable with lowpass filtering of the output. Most users probably haven't noticed the issue, since I didn't see any mention here yet. But, do you have any comment whether this high frequency oscillation might cause trouble with signals within the intended passband up ~130 MHz. I suppose nonlinearities could mix this oscillation with the input signal. Is it safe to ignore the possibility? To be specific, do you expect I can measure a 15 uA pulse signal with 0.1% resolution, given the 15 nA rms noise level (with no input) as given in the datasheet.

      Thanks,

          Gerard

  • 0 in reply to Gerard Visser
    TI__Genius 12631 points

    Hello Gerard,

     When the new deivce is released existing customers will receive a change notification from TI and all old inventory will be flushed so you will only receive new material.

    As you mentioned the old device can be used with an LPF which is usually the case since the subsequent stages may contain an ADC which would require anti-alias filtering. I personally havent seen much of an issue with mixing that causes excess noise in the region of the amplifers BW. The 15nA rms noise level spec in the datasheet is for a limited BW up to 90MHz, so with a 1st or 2nd order low pass filter that should be achievable. This device tends to be utilized in OTDR like applications, which utilize a lot of averaging which further reduces the noise. Note that a large percentage of the output noise from the amplifier is actually caused by the 20KOhm feedback resistor.

    -Samir