Gamma Spectrometry Forum

I just got myself a HackRF radio, working on getting it to work. My plan was to try to use it for magnetic resonance spectroscopy. But now the thought occurred to me: could one use such a thing for getting the fast pulses from the PMT? Because it is a pity to put them through low-pass filters so that they can be handled by a sound card.

If one could use a tv-dongle as an input instead, the bandwidth would be two or three orders of magnitude larger. Those are less than 20 USD, the HackRF is 270 USD, or maybe one would need an Ettus USRP for about 800 USD (with an RF-daughter board that goes down to DC).

This is a bit of a note to self. Maybe someone here has already thought about it, maybe even done some work?

Pieter,

Thanks for sharing your idea, I had not come across the hackRF device before, it certainly looks interesting, and I see the software is open source.

I doubt that a higher sampling rate will improve resolution, as we are already at the limit's of NaI(Tl) performance, but one could push the boundaries of count rate a little further. Keep in mind that the decay rate for the NaI(Tl) pulse is relatively slow anyway, and by the time it comes out of the PMT it's at least 50 µs. but yes there are many scintillators which have faster decay times and would be suitable for really fast count rates.

Please keep us posted on the project..

Steven

ref: http://scintillator.lbl.gov
ref: https://youtu.be/Ju6o4Jucmhs

Interesting device, sure offers alot of possibilities.

As far as limitations are concerned, I agree with Steven that the resolution is not the issue; count rate is.

From my experience, count rates above 1500 cps can introduce strange effects into the spectrum. As we amateurs (fortunately) generally deal with low activity sources, this could be a problem for those who want to use sound card based spectroscopy for more advanced scientific analysis where count rates are higher

With that said, I think the biggest limitation to what we currently can do is not hardware related, but software related, in terms of making accurate quantitative analysis with the appropriate corrections.

Simply getting a spectrum and look at it is fairly straightforward...

Pieter,

I was just watching this video on the hackRF https://youtu.be/4Lgdtr7ylNY and Michael Ossman mentioned one could tap directly into the baseband analog port with a signal +- 1V, that's pretty much in the range of a PMT signal output, and sample at 20 MHz.

Since I don't have one of these devices i don't know if one would be able to read the samples with available software, but maybe you can test this for me and report back.

Steven

This is an interesting idea but as was mentioned the use of a sound card and low pass filter is probably not a limitation to NaI system resolution. I have carefully observed the output of the PMT with a digital storage oscilloscope and have seen the variation in pulse amplitude around the voltage corresponding to 662 keV from a Cs137 source. This variation is inherent to the detector and is not due to sample rate, however if the sample rate is too low the peaks will appear to spread lowering resolution.

That being said, there is no harm in experimenting.

Neil