PMT Test Questions

[Tom Dressel]

I now have everything needed to make a neutron producing fusor except the BC-720 scintillation block.

Maybe I an just too anal but, before I spend the money on the scintillator, I would like to know that all the components of the detector at least work.

I was thinking of mounting a small LED in the window of the PMT, inside a light tight cap, and drive the LED with my pulse generator. Use the pulse generator to trigger the ocilloscope, watch the LED voltage on one channel of the scope and watch the PMT to preamp to SCA to rate meter to counter output on the other channel of the scope. Would this give me some idea of how the components are working, or would the signal be unrecognizable after going through all the NIM components?

Does this seem to be a reasonable thing to do?

Is it possible that a small, fast pulsed (~50 nano sec.) LED output would saturate the PMT? If so, I could place a filter between the LED and the PMT, like a plastic disk painted black, with a few scratches in the paint to let a small amount of light through.

Tom Dressel

[guest]

Try driving the LED at a low current for testing. Normally they (the LEDs) want at least 10-20mA for full output. I'd start at ~ 1mA and work down. If you use a blue LED, you'll be near the sweet spot of a PMT with a bialkali photocathode. Be careful with reverse voltage on the blue LEDs - they are more sensitive than the garden variety GaAsP/GaP varieties. You might also want to try placing the LED at a distance and maybe use a diffuser so that you exercise the whole photocathode area (nice but not essential) A piece of polyethylene from a milk jug would make a fine diffuser and do some attenuating as well. I might want to try something similar myself, as I'm putting together a temporary cardboard tube setup to test PMT base noise with and without scintillator. I have the preamp assembled, and it works well. I recently finished simulating a discrete preamp/integrator/pulse stretcher setup (nice results), and have just finished the board layout. This will eventually be married to a discriminator board and be placed inside the PMT housing along with a HV supply, the intent being an integrated detector that only needs a scaler to complete the measuring system.

[Tom Dressel]

I was going to use a green LED bit I will look for a blue LED tomorrow. My pulse generator has control of the output voltage so I should be able to keep the current low as you have suggested.

Tom Dressel

[DaveC]

Using an LED is an excellent way to test the PMT. Any color at all will work. Use a square wave to drive it. You can probably do it right off the signal generator. I would suggest a 100 or so ohm resistor for current limiting. If the signal generator has a 50 ohm output, that would give the best rise time.

Depending on the preamp you have for the PMT, you should rise times in the micro second range or shorter. A good TTL square wave will have transitions below about 50 nS.

The LED rise time, if memory serves me correctly can be shorter than the fall time. But usually, the PMT's amplifier circuit is a lot slower and thus sets the individual pulse resolution limit.


Dave Cooper