Time to acquire spectra

Discussions about radiation-related science
Post Reply
P2potter
Posts: 3
Joined: 25 Mar 2020, 13:07
Contact:

Time to acquire spectra

Post by P2potter » 25 Mar 2020, 13:25

As a recent buyer of a NaI probe and a GS-USB-PRO, I am trying to figure out the best way to acquire spectra of low activity sources. I have always assumed that as long a count time as possible is better from a signal to noise point of view, but at what point is there a diminishing return on the time involved?

Similarly, should I be aiming for approximate equal times for background and sample spectra, or is total count time the more important factor?

I am sure there is some maths involved here (above my head from a numbers standpoint but I’m sure I could grasp the concept), however is there a simple rule (e.g., acquisition time = time for sample cps 3x above background) that can be employed?

Thanks,

Phil

User avatar
Steven Sesselmann
Posts: 735
Joined: 27 Apr 2015, 11:40
Location: Sydney
Contact:

Re: Time to acquire spectra

Post by Steven Sesselmann » 25 Mar 2020, 17:03

Phil,

Good question..

Some of our members have been able to count very weak sources, check out a post by Karl Brewhens. viewtopic.php?f=5&t=261&p=1214#p1207

Adding even a small amount of lead shielding around your detector can significantly improve your signal to noise ratio, further you can apply background subtraction in PRA to make the signal stand out. Recordings of 12 - 24 hours are not un-usual, but not always necessary.

When you record the background you need enough counts to make the spectrum look smooth, how long this takes will depend on how many bins you employ.

I suggest starting with a 1 hour BG recording and see how it looks. The problem with very looooong recordings is temperature drift in the spectrum, which tends to widen the peaks.

Let us know how you go.

Steven

P2potter
Posts: 3
Joined: 25 Mar 2020, 13:07
Contact:

Re: Time to acquire spectra

Post by P2potter » 28 Mar 2020, 04:53

Steve,

Thanks. As an example of the data that I am getting, I am enclosing three background runs taken in my garage (the house has too much granite to generate nice spectra!).

The overlay of the 24hr data sets is as follows:

Red line - No shielding - total counts 5.1 x 10^6
Blue line - Detector under 1.5" thick lead shot bag - total counts 2.1 x 10^6
Gold line - Detector in lead castle (0.75" lead shot circle) sitting on 1.5" lead shot bag - total counts 1.4 x 10^6

As you can see, the peaks at 2.6MeV (I assume Th232 gamma) become more pronounced as the shielding increases. I presume that this is due to the reduction in the noise and relative increase in the signal. I also presume that the thorium is coming from the concrete of the garage floor. By contrast, I have included a spectrum from a Western Electric 346B tube housed under the lead shot bag (same conditions for the blue line above except 3hr collection time).

I think the data look reasonable, but still have not really figured out the best way to approach this from a truly scientific perspective.

Thanks,

Phil
Attachments
Radium 2.jpg
Radium 2.jpg (150.93 KiB) Viewed 87 times
Garage backgrounds composite small.jpg
Garage backgrounds composite small.jpg (150.55 KiB) Viewed 87 times

P2potter
Posts: 3
Joined: 25 Mar 2020, 13:07
Contact:

Re: Time to acquire spectra

Post by P2potter » 28 Mar 2020, 08:15

Steve,

Thinking about this could the doublet at ~2.6MeV be Th230 and Tl208? The latter has an emission energy at 2.614MeV and the former at 2.694MeV? Is it possible that I am detecting Tl208 decay in the Tl present in the NaI crystal?

Thanks,

Phil

User avatar
Steven Sesselmann
Posts: 735
Joined: 27 Apr 2015, 11:40
Location: Sydney
Contact:

Re: Time to acquire spectra

Post by Steven Sesselmann » 28 Mar 2020, 10:19

Phil,

Your spectra above look reasonable.

In regards to your second question, I have never heard of anyone seeing the trace amount of Tl in NaI(Tl), in any case I doubt you would be able to resolve 2614 and 2694 keV.

Steven

User avatar
Svilen
Posts: 108
Joined: 23 Sep 2016, 04:25
Location: Germany
Contact:

Re: Time to acquire spectra

Post by Svilen » 29 Mar 2020, 08:06

Hi Phil,

What detector do you use? Could this at 2.6KeV be some kind of artefact? Strange also that there is no peak from K40 visible in your BKG spectrum.

Greetings,
Svilen
Svilen

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest