I'm designning a board including three channels of light pulse detection, each channel contains TIA and FDA: enter image description here

During test, I found large pulse on one channel will cause some crosstalk on other two channles. Adding metal sheilding between each channel can suppress the crosstalk but still need to improve. (refer to enter link description here)

I want to seperate the three channels as much as possbile but wondering what is the reasonable way. I imagined one scheme to seperate the power and GND on this board. enter image description here Does it make sense? But if I add metal sheilding between different channels and connected the sheilding to GND, the separation of GND seems useless.

Who can share some ideas and detailed notice on handling the similar problems?

BR, Thanks!

  • 3
    \$\begingroup\$ What if you removed the light sensors from 2 channels? Do you still get crosstalk? \$\endgroup\$
    – Andy aka
    Commented Jun 25, 2023 at 9:46
  • 1
    \$\begingroup\$ I recently built a compact, 4 channel transimpedance amplifier board and found that crosstalk was completely dominated by scattered light. Your overall design looks to have the electronics well filtered, so I would look at optical crosstalk. I doubt separate LDOs are needed. \$\endgroup\$ Commented Jun 25, 2023 at 15:03
  • \$\begingroup\$ @Andyaka Yes, even I removed the APD sensor from the 2 channels there's still crosstalk. I'm pretty sure the crosstalk consists of both radiation and conduction. It's obvious when large current flows into the channel A TIA. \$\endgroup\$
    – FNJU
    Commented Jun 26, 2023 at 1:18
  • \$\begingroup\$ @user1850479 As talked in the comment above, it's not caused by scattered light. The mojor part of the crosstalk shall be caused by the radition since the sheilding can suppress it. Minor part is related to the saturation of the TIA, as I observed. So I want to try a scheme which can seperate the three channels... \$\endgroup\$
    – FNJU
    Commented Jun 26, 2023 at 1:22
  • \$\begingroup\$ And the most essential reason, I think, is that the gain resistor is too large, since I need to amplify very small light pulse current. This makes it very sensitive. \$\endgroup\$
    – FNJU
    Commented Jun 26, 2023 at 1:30


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.