Page 2 of 2

Re: Preamplifier and P/Z troubles

Posted: Tue Jul 07, 2020 7:58 pm
by ChristofferBraestrup
I will leave previous preamp design for reference, but preamp v.1.0 is very flawed.

I learned that the positive nice peaks i got was really the negative tail component of a, ugly bipolar pulse.

I've painstakingly assembled as much knowledge about nuclear preamplifiers for scintillation spectroscopy I could, and have now produced a preamplifier that is pole-zero tweakable until it gives a perfect gaussian peak.

See attached.

Here are a few points I've learned about preamplifiers, that wasn't easy to find in the litterature:

- Detector capacitance should be as low as possible
- Time constant (RxC) of the preamplifier should be LONGER than the scintillator time constant
->This is why counting preamplifiers give so poor resolution; low time constants are prefered to achieve higher count rates
->NaI(Tl) has a notoriously long decay time
- Standard scintillation preamps (like the Ortec 113) uses 50 ┬Ás as time constant
- The parallel capacitance across detector output subtracts from the signal going through the preamp capacitor
-> This is why longer cables before the preamp attenuates the signal (RG58 and RG174 both has a capacitance of ~100 pf/m)
- The math behind pole-zero cancellation circuits is pretty tough, trial and error is adviced.
- A NUCLEAR PULSER IS ABSOLUTELY CRITICAL for all this work.
-> I spent 4 months building different not very great preamplifiers before caving and getting one, and perfected my preamp 5 days after.

So there you have it, a low-cost functioning charge sensitive preamplifier for NaI(Tl) detectors, which is also compatible with standard NIM equipment.

Please let me know if you build one, I'd love to see if my work is reproducible :)

PMT CSA 2.jpg
preampv2.jpg