Harshaw 8S8/2A-X

[Svilen]

I was going to open the Bicron detector from my previous post and rebuild the voltage divider the way I need it, but I found this detector - Quartz & Silice holland b. v. (Harshaw crystal) 8S8/2A-X and it seems it works! I made a 12M voltage divider as Steven explains in one of his articles (thanks for that!). Here is the place to ask - are the 5% tolerance resistors really a bad idea? I will get some precise ones, but this is what I had at home so I used them for the test.
The detector is quite old guessing from the company name, but looks in very good condition, it could be that it was not used too much or even not at all. I hope the crystal inside survived the years.
Now I finally got my GS functional (so excited) and wanted to show a background and an thorium mantle spectrum here, hoping to get some advice on how to improve them or just somebody with experience to help me evaluate the performance of what I have, since this is my first time using the GS and this detector. But to let it work some hours, I wanted to ask you guys, do you think I abuse it with 850V or this is OK for this tube? I know here read some experts, that build them and maybe you can guess what for a PMT is inside. With my volume to 100% and anode voltage of 850V I get background pulses with amplitudes around 40-50 AU in intune. I know the detectors can get also higher voltage, but don't want to shorten it's life, so I will be very thankful for advise. Than I'll let it run and show you what I got (unshielded, since I still have no lead castle).

[Sesselmann]

Divider looks good, and I don't think the resistor tolerance is critical, more important that you use metal foil resistors and not carbon, as the carbon resistors can add noise.

The voltage between dynodes will be roughly 1/12 so if you run at 850V, you have a potential between dynodes of around 70 volts. Increasing or lowering the voltage by 100V is only going to make a difference of 8 volts.

Find the optimum voltage by recording the cpm in 50V steps between 600 and 1000 volts and identify the plateau.

If you are running the PMT at optimum voltage and the sound card at 100% and still find that your spectrum only covers half the arbitrary unit scale, you can tweak the preamplifier inside the GS, but get the rest working first.

[Svilen]

Thank you for your response Steven!
The count rate stops to increase significantly after 850V, so I guess 850-900V should be the optimal value for this tube.
This is what I get from InTune with the thorium mantle and volume set to 100%: Noisy?
[broken link removed - Svilen]

Then I used these settings, and got this here as a background spectrum:
[broken link removed - Svilen]

And from the thorium mantle (well at least there is some difference):
[broken link removed - Svilen]
The spectrum in a bigger window (I didn't figure out how to change the range of the horizontal axis):
[broken link removed - Svilen]

Can you please tell me what is right and what is wrong here? What should I do further? I see it is also not filtering any pulses, something should be wrong I guess. Thank you very much in advance!

[Sesselmann]

Yes looking at your Intune screen shot I can see you have zoomed in by the scroll bars in the image, which means your signal has low volume. The Intune window at full height is 100 arbitrary units, so that represents maximum line level (about +- 1.7 volts)

There are two ways to go, software or hardware. You can change the boost gain, top right Data Acquisition and Analysis. This essentially boosts the current signal, alternatively you can open the GS and adjust the blue 25 turn trimmer potentiometer which controls the pre-amplifier volume (its the one without the red seal).

Steven

PS: Can you please change your login handle or add a signature with your real name, we like to keep the forum personal and friendly.

[Svilen]

Excuse me please, now the login name is my real one.

At the moment the pulses I get are between 40-50 au, which amplitude would be the best to achieve after turning the trimmer?
Well, we have a problem - there are actually no trimmers in my GS !?

[Sesselmann]

Svilen,

Thanks for updating your name. So I guess you must have the GS-1100A model, in which case the gain is fixed to 10% with the resistor combination 1MR and 100KR, my other model GS-1100-PRO has a trimmer potentiometer across this divider. In this case I recommend you use the boost gain in the software.

The standard objective is to set the volume so your pulses extend over the entire +- 1.7 volt range, but even if your detector falls a but short of this you should still get a good spectrum. Unlike other commercial MCA's which only offer 4096 channels, PRA is 16 bit, so the resolution of channels is over 32,000, all you need to do is to set the bin size to 0.01 or 0.005 and you will see the finer details.

Chances are you just have a PMT with very low gain, this might even be a good thing, as low gain PMT's often have better resolution than the ones with high gain. The gain can vary significantly even in the new PMT's I buy for my own detectors, so I have to specify a preferred range when I order.

Steven

[Svilen]

Thanks again, Steven!
Yes, the GS is the 1100A model. I still can change the resistors to increase the gain if you think it is a good thing to try? Otherwise, I boosted the software gain to 2 and below is what I got:

Thorium mantle, ~2000 sec acquisition, no shielding, gain boost 2, bin size 0.01:
[broken link removed - Svilen]

Same, but with bin size 0.05, cause it maybe looks better:
[broken link removed - Svilen]
Here compared to what Tom from anti-proton.com shows, maybe some lead shielding can really improve it?
[broken link removed - Svilen]

How do you find it?

[Sesselmann]

Svilen,

Those Thorium spectra look pretty good, I personally don't think you need to loose any sleep over it. One day you might find another detector with more gain and everything will be fine.

If you want to play with the hardware, I would suggest changing both gain resistors with a small 1M ohm potentiometer, then you will have that extra control over the volume, but just remember when you increase the amplifier gain you also increase the noise.

Steven

[Svilen]

Thanks Steven! The thorium mantle is quite hot, I wonder if it will detect something weaker, but hope so. Do you think this spectrum is good enough to try calibration with it?

I very likely wouldn't touch the hardware for now, but just in case I try it in the near future - would you please tell me which resistors they are (maybe I can guess it looking at the circuit, but to make my life easier and avoid mistakes) and maybe two example values you think I can set the trimmers to, as a starting point?

[Sesselmann]

Svilen,

The preamplifier gain is determined by two resistors 1M and 100k values which then give around 10% gain. To increase the gain tou would need to increase the value of the 100k resistor or replace both resistors with a 1M trim pot.

This is a very basic single transistor amplifier.

If you have a very hot sample you should make sure that your count rate does not exceed a reasonable limit, if the count rate is too high you will loose a lot of counts, so just pull back the source until you have a reasonable count rate retention.

GS-Preamp.JPG