Preamp for scintillator probe use on GM Geiger Counter

Linear and non linear

Preamp for scintillator probe use on GM Geiger Counter

Postby George Dowell » Sun Sep 11, 2011 8:51 pm

Geiger Counters designed specifically for GM tubes have an input sensitivity in the hundreds of milliVolts.
This is fine for GM tubes as they can provide a signal well into a few Volt range, and there is no need
to add system gain which increases system noise.

When using scintillator probes on such meters, the gain needs to be increased to the 30 to 50 milliVolt range, otherwise
the smaller signals from weak energy sources can be ignored.

The design I developed works fine to convert CDV-700 and all other GM only meters for use with scintillators, giving them an effective input sensitivity ( a.k.a. LLD, Discrijminator,Threshold, Gain) of 25 mV.


Geo
Attachments
Preamp%20Inside.jpg
Inside the preamp
Preamplifier%20Schematic.jpg
Schematic
Preamp%20Outside.jpg
Preamp Finished
Preamp%20Outside.jpg (28.07 KiB) Viewed 3472 times
User avatar
George Dowell
 
Posts: 41
Joined: Fri Sep 02, 2011 12:27 pm

Re: Preamp for scintillator probe use on GM Geiger Counter

Postby Doug Coulter » Mon Sep 12, 2011 10:47 am

I like this trick, muxing the HV but sticking a preamp into the signal chain all on one wire. That's my kinda way to roll. I'm kind of surprised those old geigers are that sensitive. We are getting >100v pulses out of our Russian tubes here (not cheating, we're running them at the nominal volts and impedances from the datasheet), and have to divide 20::1 and use a high impedance series leg to keep from hitting the protection diodes in the cmos too hard, even with that. Your scint heads must have unusually high output?
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 2964
Joined: Wed Jul 14, 2010 8:05 pm
Location: Floyd county, VA, USA

Re: Preamp for scintillator probe use on GM Geiger Counter

Postby George Dowell » Mon Sep 12, 2011 11:00 am

Doug, none of my scintillators are special. I do try to use 9 or 10 stage PMTs for good gain but everything is standard from there. I use either 60 meg or 100 meg bases and a single wire feed for most applications. Only on the main MCA do I use a commercial base with preamp inside and lower dynode strings resistance.

By standardizing across the board we get pretty good probe to probe performance and more important my shop to the next guy's shop performance.

Geo
User avatar
George Dowell
 
Posts: 41
Joined: Fri Sep 02, 2011 12:27 pm

Re: Preamp for scintillator probe use on GM Geiger Counter

Postby Doug Coulter » Mon Sep 12, 2011 11:38 am

I guess I'll have to look into your work on those old CD counters to see why they are so voltage sensitive (low impedance input that cuts down the geiger tube output?) since I'd never design for millivolt input levels from something that intrinsically makes 10's of volts (at least). We had to add serious attenuation to ours to get down to 5v swings. While I could make an exception for really well built old stuff, most of the "cargo cult" in the old stuff turns me off, personally. I prefer to know how it works inside, and to be able to get parts to fix it. I can't say how much non working NIM stuff Bill has managed to collect, no schematics, and what's wrong with it is always the house-numbered hybrid chip you can't get anymore. After your 4th or 5th non working NIM MCA/preamp/threshold box/power supply, how much more does one tolerate? We're just going to make our own, at least we'll know how to fix them and be able to get the parts.

I don't see how anyone can really trust that which they don't understand, but that's my personal take. If you don't know how it works when it's working right, how can you tell when it's not? Despite what I look like, I'm real conservative on a lot of things, this being one of them. The "clown suit" is a ruse to instantly detect those who aren't paying attention to what's under the cover of the book.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 2964
Joined: Wed Jul 14, 2010 8:05 pm
Location: Floyd county, VA, USA

Re: Preamp for scintillator probe use on GM Geiger Counter

Postby chrismb » Mon Sep 12, 2011 1:49 pm

We've been talking about this on the other board, George. For the Russian 3He tubes, these are around the 10mV and very flakey to anything but mega-input impedance. I threholded mine in the preamp, generating a ttl output there, either + or - pulse, which I fed into a crate that was generating the volts and picked up the detection.

Seemed a sensible way to get the signal away from the pre-amp without introducing further EMC breaches, so I decided on that topolology for my fully-portable version too - a separate box with the power source (CCFL) so as to keep the magnetic interference away, but also sending back down that power line the ttl which gets 'dealt with' in the power box end, feeding out to a counter and also to an audio line so that it can be counted and recorded.

When I tried it I thought I might get feedback into the preamp, but it didn't work out like that. Seems to work fine, though I did have to put a big resistor in series with the coupling cap to avoid lots of false counts with my crate, as it is already set up to be sensitive enough for scintillators.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 7:32 pm

System PARAMETERS. Setting, understanding their meanings.

Postby George Dowell » Mon Sep 12, 2011 5:16 pm

This is just a stub/ I'll drop the whole article here somewhere in a fresh thread when it is better formatted for this forum.

MOVING TARGETS.

We have a number of parameters to play with in the Home Rad Lab equipment.

Ludlum makes it easy for end users because they know their market well. All their Model 3's are
nothing more or less a set of parameters. a particular set of tools. By this I mean, the Model 3 is ideal for use with a GM Pancake and GM thin walled metal beta gamma probe.

There are no input sensitivity adjustments, because they are not needed. To have them COULD compromise the proper operation of the instrument in the field ( Double Pulsing etc).

Model 3's DO have adjustable HV, but no way to meter it. The meter, probes, scales, HV, even the cable are set up as a unit by a bench tech (me- as a hobby) and that's that. Nothing to adjust so nothing to screw up.

Models 3's have been made for decades. Early Model 3's are very different in HOW they get the operating requirements to work, but ALL Model 3's operate to the same set of requirements ( Battery Check, number of ranges- 4, adjustable HV 400-1500, no meter readout of HV, Fast/Slow response switch, audio on-off switch.fixed input sensitivity).

Add an alarm feature, it becomes a Model 3A.
Change one thing, make it 3 ranges instead of 4 ranges, it is now a Model 2!!


There will always be a Model 3 because there will always be a requirement for a meter that does what it does, nothing more, nothing less.
Depending on the vintage, the input sensitivity (that is the lowest signal in milliVolts to which the meter responds) is 30 or 40 mV, negative going.
Ideal for GM and even SOME scintillator applications > by no means ALL scintillator applications, far from it).
Much more on this key parameter shortly.

Jump next to a Ludlum Model 12. It is not an improvement or advancement on a Model 3, it is simply a different bag of tools. Like the Model 3, the 12 has been made for decades with all the technical electronic advancements available at the particular vintage, but they all do the same things.

With the added features on the '12 you can monitor the HV on the meter, and most importantly set up the meter's input sensitivity (LLD, Lower Level Discriminator, Threshold)> 1 mV to about 100 mV range here and the HV range is expanded to 400-2500V.

Next up are the "analyzers" which have all the features of a Model 12 but add an additional control called WINDOW ( ULD, Upper Level Discriminator). Adjustable from 0-60 mV above LLD.

With the WINDOW control you can set another parameter, where the meter will respond ONLY to a signal that is above LLD and below ULD. This makes it by definition a Single Channel Analyzer. Very handy if needed, totally devastating if ser wrong.



So one up the line it goes until the Model 2221 which has every control, but each and every control MUST BE SET to make it do what you want it to do with a particular probe.

THERE IS A REAL NEED TO UNDERSTAND THE FUNCTION OF LLD WHY IT IS SET A CERTAIN WAY FOR CERTAIN PROBES>PERIOD> THIS IS DONE WITH A PULSER> MORE ON THIS ANOTHER TIME.

The probes.

Different class probes work vastly differently and a meter used with that class probe must be set up a certain way.

GM probes can give huge pulses/. The HV must be set somewhere within the operating range of the probe, they are not too critical.

Scintillator probes can give smaller pulses, but the height of those pulses means something. The height of the pulse is related to the HV setting, and for some scintillator materials, the energy that creates the pulse in the first place. By diddling with the HV setting, LLD and ULD, a window can be made just about at any pulse height to respond to a given energy. Usually we do this in a precise manner so that a certain gamma energy is equal to a certain pulse height ( and sometimes read all at once as in MCA).


Proportional probes give VERY small pulses and the height of those pulses can mean something.

One last thing while I pause for discussion. Sometimes we use a scintillator with MORE THAN ONE type of
detector in it, reading say Gammas and Betas, or Alphas and Betas at the same time but independently. This is accomplished by pulse height detection.


The picture shows a typical Geologists Gamma Ray Spectrometer with remote probe, servo loops for temperature feedback and "downhole" HV generation and pulse shaping. With these we look at 5 things at the same time. 1- TC or Total Counts of all energies, #2 U window - we monitor the Bi-214 line at 1764 keV for Uranium , #3 Th channel We monitor the Tl-208 @ 2620 keV for TH 232 , and the K-40 line (1460 for K-40). The last channel is called a stability channel and is an artificial channel introduced by either an isotope or a small LED lamp. The precise position of this channel is adjusted by HV electronics to keep the whole system calibrated. There are a number of schemes, the newest uses onboard K-40 for continuous calibration, the fraction added being removed by software.

Geo
Attachments
Scintrex%20#2.jpg
Geologists Gamma Spec
User avatar
George Dowell
 
Posts: 41
Joined: Fri Sep 02, 2011 12:27 pm

Re: Preamp for scintillator probe use on GM Geiger Counter

Postby chrismb » Sun Sep 18, 2011 12:25 pm

Today I finally got time and energy enough to start board assembly for my pre-amp. The one thing I have come to ponder over is the connections to implement the method we are both using to feed the signal back down the HV line.

The thing is, if we look at your circuit, as the HV voltage comes onto the line, it will lift the capacitor, C4, until such time as the resistors R3 R4 R5 and R9 charge up the capacitor, or otherwise through R9 and the resistance/short down to ground in the 3V supply. The cap charges up through these resistors. RC is still very low, micro-s in duration, but do you think there is any need to put in some diodes so reverse condution can charge up the cap if the HV comes onto the line too quick?

My output is from a CMOS chip output so will be even higher impedance on that side of the signal coupling cap. I'm just wondering what I should do about it. That being said, I did run the circuit before and nothing 'blew up', but it seems we are leaving it to the God's of incidental resistances.

(I also have a 220k resistor between the coupling cap and the HV line, which slows the charging down on the 'HV' side of the cap.)

I'm thinking that maybe I need to stick an inverting pnp in there, so that the coupling cap can be connected to the low side held to ground for power-up/most of the time.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 7:32 pm

Re: Preamp for scintillator probe use on GM Geiger Counter

Postby George Dowell » Sun Sep 18, 2011 3:33 pm

The circuit is stable and there have been hundreds of examples built up but only after thorough bench testing of the design. .
Input diodes have been added with good results, but by and large we find they are not needed. Variations of the HV + Signal Splitter Combiner circuit is also well represented in the field in the form of "adaptor boxes" for using two wire probes on one wire electronics and the reverse.

We tend to test at the front end, then stick closely to the design in production.

Geo
User avatar
George Dowell
 
Posts: 41
Joined: Fri Sep 02, 2011 12:27 pm

Re: Preamp for scintillator probe use on GM Geiger Counter

Postby chrismb » Sun Sep 18, 2011 3:48 pm

Can you clarify what you mean by 'input' and 'front end'?

I'm just pointing out that if the HV input from the scaler flies high(/low) very quickly, you will pull Q3 high(/low) with it, as there is nothing to directly charge up C4 (viz. that holds the 'ground-side' low). It looks like you're relying on the transient immunity in Q3 to deal with any such possibilities, maybe it's good enough for the job.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 7:32 pm

Re: Preamp for scintillator probe use on GM Geiger Counter

Postby Doug Coulter » Sun Sep 18, 2011 4:23 pm

I'd say that unless your crate simply switches HV hot, from an already fully charged internal filter cap, that there would be no problems with this, particularly if the 3v batteries were switched in at the time -- that 240 ohm will take care of charging the cap fine without much overvoltage. You have to pick up a sense of perspective about "fast" vs the other time constants involved here. There's a big difference between "too fast to see on a voltmeter needle" and "fast enough to make any real current through a .001 uF cap". Most transistors will turn on and draw some current in a zener mode if nothing else, and not hammered too quick. Anything like this the provides the HV through a sense impedance so it can also see a signal on the same wire has, you know, that sense impedance in series with the HV drive. It's usually from 10k and on up - 100k more typical, I've seen 1 and 10 megs there in some things. You should probably make tests, then comment, since Geo HAS made tests.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 2964
Joined: Wed Jul 14, 2010 8:05 pm
Location: Floyd county, VA, USA

Next

Return to Analog

Who is online

Users browsing this forum: No registered users and 1 guest