Amazing mailag episode! Thanks, Dave! I really liked the wafer production steps and the Single Photon Counter. I wish there would be someone like Dave doing optoelectronics (lasers, photon detectors, electro- and acoustooptics etc.)... I find these things very interesting (and happend to be doing a PhD in). Maybe there is, anyone have any recommendations?
Or maybe Dave could do it? <: He seems very modest in saying he's not an expert, but he's doing it very well!
Here's a few ideas on regarding the Photon Counter and the signal in the first 2 us:
1) The signal is some sort of real optical backscattering, 2 us is only 300 m roundtrip for light, maybe there's something reflecting it back. Although the abrupt end of the saturaton makes this explanation unlikely.
2) Trigger-to-laser emission jitter? You say there's a 600 ns delay between the trigger rising edge and the laser emission and you show a long persistence trace. Is it the same pulse-to-pulse? Maybe 600 ns is just the minimum delay and it varies in the range of 2 us. I'm assuming your laser is Q-switched and these can have pretty much random trigger-to-emission delay. I believe it gets worse with increasing pulse energy.
3) EMI from the laser Q-switch ring-down? (Again assuming your laser is Q-switched). I've had major interference problems with a table-top mJ-level Nd:YAG Q-switched system and my detection electronics. After the laser emission the detectors would pick up us transients which would get worse the closer they were to the laser head and the Q-swtich PSU. Even some meters away I would get unacceptble EMI. And it's all with shielded BNC cables, grounded cases etc. It did get better with double-shielded cables though and erm... some tin foil wrapping here and there. Maybe it's the Q-switch EMI pickup that is triggering your detector?
4) Finally, if the detector has a gate input, why not use a fast PIN diode with a TTL pulse shaper and use the laser pulse itslef to gate out the initial detector response? (You may need to get creative to have the gate pulse arrive before the laser pulse). If the problem persists, you'll know it has nothing to do with the laser pulse and most likely is purely electronic.
And yes, the output for these detectors should be loaded with 50 Ohms. <: