I have been reading more into the gain of a photodetector and how the system works. We are picking up extremely small signals and need to use a detector with a gain of around 1.0 × 10^7 but preferably higher. We are not looking to use PMTs. We are then also looking at a photodetector which can pick up modulation frequencies on the order of 500MHz - 2GHz so the detector needs to be very fast.

There are a few questions within this:

  1. Are there specific types of detectors which are best for this applications, such as Avalanche photodiodes or others?
  2. With detectors that have lower gain, how much is it possible to amplify them, can you take a detector which has a gain of around 1.0 × 10^5 and increase it to around 1.0 × 10^8 within hurting other areas such as Dark Noise?
  3. If you know of detectors which can do this can you point me in the right direction?
  4. Are GE Avalanche photodiodes more sensitive than Si and InGaAs?
  5. Are there other methods which can improve a photodetector which cannot achieve it currently?

Extra notes: - The wavelength that we are using in between 750-900nm

Thanks for the help!

  • \$\begingroup\$ Are you sure you need all that gain in the detector itself, you can't use an electrical signal amplifier after the detector? \$10^7\$ is an awful lot to ask from an APD, AFAIK. \$\endgroup\$ – The Photon Mar 13 at 17:22
  • 1
    \$\begingroup\$ And an APD is (again AFAIK) the only detector with internal gain that's likely to get you near 2 GHz bandwidth. \$\endgroup\$ – The Photon Mar 13 at 17:32
  • 1
    \$\begingroup\$ What's the level of flux when it arrives at the detector? (Photons per second per mm^2 or something similar.) And do you have detector size limitations due to the optical situation? Can you talk more about the surrounding situation? (It seems like you are being a bit tight-fisted, right now.) \$\endgroup\$ – jonk Mar 13 at 17:37

To answer your questions:

  1. Gain is not really relevant, you can cascade as many amplifiers as you want. What matters is NEP, that sets when your SNR is limited by detector noise vs. shot noise. What NEP do you need?
  2. Sure, amplify as much as your want. It does not help SNR.
  3. Specs are not well defined yet. Can't specify if it is possible.
  4. APDs have lower NEP, but introduce multiplicative shot noise. Hence, if you can be shot noise limited with a standard TIA photodetector, it will perform better than the APD even though the APD has a much lower NEP.
  5. Probably, but you need to explain what you are using.
| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.