Pulse height shift with pulse rate?

[Mike S]

What would cause a pulse height shift on a detector with a change in pulse rate?

I'm not 100% sure this is a detector issue, but I haven't seen this with other detectors. I don't think this detector is bad; it is from ebay, but it looks unused. I just think there's a phenomenon going on that I don't know about yet.

I bought one of the cute little Ludlum 44-62 detectors. It has a 0.5" x 1" NaI crystal and a 0.5" PMT in a package not much larger than a large marker. It seems to work OK, but I do have to run it at below about 650V to avoid excessive counts (I have not run a plateau curve yet).

What is confusing me is when I try to run a spectrum, the pulse height appears to be affected by the pulse rate. I kept chasing my tail trying to do an energy calibration because it would shift every time I changed sources. Then I tried keeping the exact same setup and software settings but just moved the source. I got the effect of shifting the spectrum significantly. Take a look at the attached screen shot for a comparison of a Cs-137 source at 350 pulses per second vs. it moved farther away and giving closer to 60 pps. Note that the energy calibration is off because, well, what I just said.

What could cause this shift? I'm assuming it has to do with the small crystal somehow.

Thanks in advance!

Mike S.

[Sesselmann]

Mike,

Interesting observation, I have never noticed it, are you seeing the same thing in PRA?

How much difference in pulse rate are we talking about?

The limitation of sound card spectrometry seems to be around 2000 - 2500 cps., so it's generally a good idea to stay below this.

In any case it would be interesting to see if it is a problem on both software platforms.

Steven

[Mike S]

The count rates were 60 and 350 cps.

Yes, the issue is there with PRA also.

I tried going up 100V and the shift is less prominent, but still present.

I'm thinking the problem is noise from a bad SHV-BNC cable (not from GS). I think at low count rates the noise has a larger impact and somehow shifts the output. That's my working theory for now. I'm already working on resolving the cable issue since that needs to be eliminated anyway.

[luuk]

Hi Mike,
Interesting effect you see there.
Normal you would expect a peak shift e.g. when the vd chain is too high in value, then the peak can shift ad different counter rates, but normal when the counter-rate is higher the peak will get lower, because the vd cannot deliver enough current to the pmt at higher counter-rates
Your effect is just the other way around that is special, never seen it before.
Luuk

[Peter-1]

I can also make a contribution to this topic. I also recorded different pulse frequencies for my system. But I also have to say that my high voltage is very stable and the resistances between the dynodes of the PMT are very low. So the lines do not change, but the intensity is influenced. It is likely that the signals are coupled via several capacitors. With a fast pulse sequence, the zero line is shifted and could be the cause.

2020_06_11-Pechblende-CPS-b.jpg

2020_06_10-Pulszahl-CPS-B.jpg

Peter

[Mike S]

I can also make a contribution to this topic. I also recorded different pulse frequencies for my system. But I also have to say that my high voltage is very stable and the resistances between the dynodes of the PMT are very low. So the lines do not change, but the intensity is influenced. It is likely that the signals are coupled via several capacitors. With a fast pulse sequence, the zero line is shifted and could be the cause.

Peter

That's what I'm thinking (capacitors in the circuit shifting the voltage), but the schematic in the manual is clearly wrong, and the voltage divider board layout in the manual doesn't match what's in my detector either (I took a peek, but I'm not sure if I can actually get it and the PMT out of the housing). I keep meaning to contact Ludlum and see if they can share the correct schematic.

Mike S.

[Peter-1]

Hello Mike,
I have another circuit directly on the PMT. So I'm independent of the capacity of the cable. Of course, this circuit needs more power. The scintilator head comes from Siemens.

Szintillator-06.jpg

Peter

[Sesselmann]

Mike et.al.,

I just did a quick test with the GS-USB-PRO and one of my standards GS-1515 NAI detectors using a Cs137 source placed at 15, 10, 5, and 1 cm. with count rates varying from 221 cps, to 1710 cps. and see no significant drift. The slight drift we see is more likely due to temperature changes, or that the background slope is more prominent at lower count rates.

CPS - Mean arb.u
221 - 25.310
284 - 25.296
516 - 25.494
1710 - 25.417

15cm.png

10cm.png

5cm.png

1cm.png

Steven

[Mike S]

Mike et.al.,

I just did a quick test with the GS-USB-PRO and one of my standards GS-1515 NAI detectors using a Cs137 source placed at 15, 10, 5, and 1 cm. with count rates varying from 221 cps, to 1710 cps. and see no significant drift. The slight drift we see is more likely due to temperature changes, or that the background slope is more prominent at lower count rates.

...

Steven

Thanks for the hard data.

What configuration do you use for the capacitor(s) in the voltage divider in your detectors (if it's not proprietary)? The probe I started this thread about (a very small NaI crystal and PMT) is likely intended for surveying and not spectroscopy, so I assume the capacitor(s) in the voltage divider are optimized for that purpose*. I'm curious how much difference it makes.

*As mentioned earlier in this thread, the manual for this little probe is obviously wrong and I can't get full access to the voltage divider board to figure it out, so I'm not sure how it's actually configured.

Mike S.

[Sesselmann]

Mike,

The detector I used yesterday has an ADIT 1.5" B38 series PMT with integral divider. from memory it is a 10 stage divider with 1.5 Mohm resistors, pretty sure there are three caps between D7-D8-D9-D10

Steven