• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 47 subscribers
  • Views 150 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.

OPA855: OPA855 TIA: Non-linear saturation behavior (sag/dip) under severe input overdrive with 1ns pulses

Florian BONNET-GALAND
Prodigy 10 points

Part Number: OPA855
Other Parts Discussed in Thread: OPA695

Hi TI Support Team,

I am currently working on an optical receiver design and observing a very specific, non-linear saturation behavior on my signal chain. I would like to get your insights on whether this is a known characteristic of the OPA855's internal architecture when subjected to extreme input overdrive.

System Overview:

  • Sensor: Avalanche Photodiode (APD) receiving 1 ns optical pulses from a laser.

  • 1st Stage (TIA): OPA855 with a feedback resistor Rf = 4 kOhms.

  • 2nd Stage: OPA695 configured in an inverting amplifier setup with a gain of 2.

  • Measurement: The attached graph shows the peak output voltage measured at the end of the chain (after the OPA695) as a function of the peak optical power of the laser.

Observed Behavior:

As shown in the attached plot, as the optical power increases, the system exhibits a "double saturation" profile with an intermediate voltage drop:

58c97994-f0eb-4dbd-92b3-612fb2283daa.png

  1. Linear Region & 1st Saturation: The output rises linearly until it hits a first saturation point at roughly 2.5V.

  2. The "Sag" / Voltage Drop: As optical power continues to increase, the saturated output voltage suddenly drops by approximately 0.4V (down to ~2.1V).

  3. 2nd Saturation (Hard Clamping): At even higher optical powers, the voltage recovers and hard-clamps back up to 2.5V+.

Questions:

  1. Can you confirm if this specific "sag" during deep input overdrive is a known behavior of the OPA855 caused by the activation of internal input protection diodes?

  2. Does the internal clamping mechanism temporarily alter the biasing of the output stage, leading to this voltage drop before feedforward takes over?

  3. Reliability: Are there any long-term reliability concerns for the OPA855 when repeatedly subjected to this deep overdrive state with 1 ns pulses?

  4. Do you recommend any external protection schemes (e.g., external clamping diodes on the input) to mitigate this without sacrificing the bandwidth required for 1 ns pulses?

Thank you in advance for your time and expertise!

Best regards,

Florian Bonnet--Galand  -  IRIDESENSE

  • 0
    TI__Intellectual 2620 points

    Hi Florian,

    Apologies for the delay of this response. The main way any protection diodes would be activating is if the inputs or outputs of the device are above the supply rails. What are your supplies set at? If they are at +/- 2.5V then the output of the amplifier shouldn't be able to reach +2.5V normally. If that is the case then seeing the output start to go above +2.5V in your data would cause the ESD protection diodes to conduct and could possibly cause that kind of sag as you overdrive the device more. 

    If your supplies are well above +2.5V then this might be something else. You might want to double check if there is a large differential voltage in-between the Vin+ and Vin- pins that could conduct these diodes as well. There are reliability concerns with conducting these diodes if this is the case.

    Unfortunately a lot of external clamping schemes do sacrifice bandwidth. 

    Thanks,

    Evan