Problem detecting K-40

[sgt_bear]

Hello

I have a working detector for Spectrum, and i made some spectrum with it - so its generally in working condition.

Since i have my new shield, the background is reduced much, and i've seen a prominent peak at 1000keV, i first thought this was K-40, but Steve told me in another post the Peak should more be a hump.

Detector is a Scionix 38B57 running at 850V

So i bought pure Potassium Cloride, and put in the shielding with the detector. I'm getting around 10cps now - instead of 6 without the sample.

Sadly, there is no peak at the K-40 region - but there should be nearly half of all cps of K-40.

I have attached two pictures, one without the sample, and one with.

With Sample:

Without sample

To validate if the Potassium is legit, i used a geiger counter which has gone up from 50 to 400cpm.

I also attached the Detector to my Ludlum Model 3 (950V)
Background: 3.2 k CPM
With Sample: 4k CPM

To find out if the software is the problem, i tried PRA and the Image looks relatively the same.



Anyone has a guess what could be the problem

[lodovico]

Hello,
you have to run the measure for as long as you can, for example 2 or 3 hours. First take the background for three hours, you memorize it, and after that you take a measure with the sample (as much as you can, 100 g for instance), in the same chart, to have evidence of the difference. Of course the detector has to be calibrated, it seems that the peak there is but you have to stretch a little bit the energy range.
L

[luuk]

Hi Lodovico,

The detector that you are using the Scionix 38x57 is it modified
or do you use it as it was original?
The point is that this detector is none linear used as it is.

I am not familiar with the software you use but the spectrum you showed is strange I don’t see any activity in the high energy region, like it is cut off.
You should always see some signal there coming from the environment.
So maybe one of your settings is not correct.

Luuk

[sgt_bear]

I bought this detector from ebay seller 'iRad'. He says "refurbished with new 120meg dynode chain" so i guess the detector is modified. But since i think the seller is very experienced i dont think thats the problem. There is also an image in the auction with Co-60, perfectly detecting with good resolution.. so if co-60 works do should K-40..

The settings i used are similar to the one in the auction and other. But since this problem occurs on Theremino and PRA i dont think it can be settings?.

For the detection circuit i use the theremino circuit but with a soundcard chip on board....

Sadly i have no Co-60 to try

[luuk]

Hi Jonathan,

Sorry, I called you Lodovico before!

If the detector is modified by Tom from Irad I know for sure that it is done in a good and professional way so there is not the problem.

My guess is that there is something wrong in the adjustment of the thermino software, unfortunately I have no experience with it but the lack of signal in the high region of the spectrum looks like there is a cut off some were adjusted, do you have an other type of source to test if it is working good, like a thorium mantle?
Luuk

[sgt_bear]

Hi luuk, no problem :)

Yes i have a lot of samples to know its working, see my post:
viewtopic.php?f=5&t=264

But you still can see there that prominent peak, even with background substraction.

Sadly i have no high-energy sample.

I tried PRA Software, and the spectrum looks the same.
Currently my only idea is the sound card chip, i will open my detector-driver tomorrow and solder a audio-cable to the pmt output and try with another soundcard...

Regards Jonathan

[iRad]

Hi Jonathan - Something does not look right about the spectrum, and I would love to see one with a higher count rate. Is it possible to see a spectrum with other common sources, especially Cs137 if you can find one. If you can find a source to give you in the 250 to 350 cps rate, that would be good. With only 5 to 15 cps, any noise will tend to make large spikes, and that tells me there is probably something wrong somewhere. Background alone should be in the >50 cps, unless the detector is heavily shielded. Also, make sure AGC is off on audio input. Leaving AGC on is the most common problem, and you will not be able to get a good spectrum with it on. Best regards, Tom

[sgt_bear]

Hi Tom, yes this was all inside my shield. Outside the shield i have around 40-70cps, but then the peak increases. My only source i have would be Am-241, i have not any other active sources. I will check tomorrow with another sound card to see if thats the problem...

Regards Jonathan

[iRad]

Ok, that make more sense. With such a small count rate (only about 5 pcs), any noise could show as a large spike in comparison. Cheers, Tom

[Sesselmann]

Jonathan,

I think you should try PRA, it shows you the spectrum without any convolution, it could well be that you are seeing K40 as a massive peak if Theremino is doing something strange to it.

Alternatively if you have a faulty cable or if there is a high voltage arc in your system it could produce a false peak which drowns out everything else, but I doubt this is the problem, try PRA.