Radioactive Mineral Sample Prospecting Equipment - Advice Sought / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Radioactive Mineral Sample Prospecting Equipment - Advice Sought

first previous 2 3 4 5 next last

Move Thread

LAN_403

Proud Mary

Tue Mar 30 2010, 09:01PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

A Russian design of the 1960s, which I'll call a 'GM wand,' deserves a mention here. In this topology, the detector, batteries, and electronics are organized co-axially in an aluminium tube, which must be very useful for 'sniffing' into places otherwise difficult to access with the 'brick with a handle' devices of the Cold War.

Given that all but a very few ore exposures in Britain are to be found in and around sea cliffs, I could see the GM wand as being very useful. Once could even construct it so that it could be folded in two or three, for convenient transportation.

IntraWinding

Wed Mar 31 2010, 10:22AM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

I've bought the CI-21G / SI-21G

Denys says he has no data on either of these tubes, except size:
CI-21G / SI-21G...........260mm x 18mm
STS-6 / CTC-6..............195 mm x 17mm

I've asked why he charges a higher price on the smaller CTC-6.

'GM wand': I could add the increased access of the GM Wand to my 'metal detector' styled device by making the angle of the sensor head adjustable all the way to 180 degrees. Could be handy for sweeping high up on a cliff face.

Alan

Proud Mary

Wed Mar 31 2010, 01:47PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

IntraWinding wrote ...

I've bought the CI-21G / SI-21G

Did you think scientific instruments were sold by the yard?

I would guess that the SI-21G has an active length of ~220mm

Tubes of this length often need an anode voltage of 950-1150V, but without any data you'll have to determine the centre plateau voltage yourself by the usual method of drawing a graph of volts (350-1250V) against counts.

IntraWinding

Wed Mar 31 2010, 03:27PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

Proud Mary wrote ...

Did you think scientific instruments were sold by the yard?

Very witty! Well it kind of looked like that was how he was selling them...

---

Proud Mary wrote ...

Tubes of this length often need an anode voltage of 950-1150V...

Quoted from Wikepedia:
The halogen tubes were invented by Sidney H. Liebson in 1947, and are now the most common form, since the discharge mechanism takes advantage of the metastable state of the inert gas atom to ionize the halogen molecule and produces a more efficient discharge which permits it to operate at much lower voltages, typically 400â€“600 volts instead of 900â€“1200 volts. It also has a longer life because the halogen ions can recombine whilst the organic vapor cannot and is gradually destroyed by the discharge process (giving the latter a life of around 10^8 events).

So the longer tubes tend to use organic vapour and have a shorter life?

Alan

Proud Mary

Wed Mar 31 2010, 04:17PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

I'd say that was more often true than not, but where the tube wall is unlined stainless steel, attack by bromine or iodine quench gas must be taken into account. Lining the steel skin with, for example, tungsten, adds very considerably to the cost, and the complexity of the factory plant needed to do it, problems both in the context of Soviet five year plans.

It is true that organically quenched tubes do not last as long as the halogen quenched species, but the number of counts till failure is still very large indeed - the kind of figure unlikely to be reached outside the world of fission and high level waste.

IntraWinding

Thu Apr 01 2010, 12:50PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

Proud Mary wrote ...

...the kind of figure unlikely to be reached outside the world of fission and high level waste.

Probably do me then

Alan

IntraWinding

Mon Apr 19 2010, 02:17PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

I'm not getting the kind of counts vs. voltage graph I expected.
(measurements at 25V intervals):

1271686013 2261 FT86405 Geiger Plot 2 Small

I'm driving the tubes central wire from a positive voltage source through a 4M7 resistor (guessed value) with the tubes outer case connected to the base of a Darlington pair (2 x 2N2222) with the emitter grounded and the collector driving an 8 ohm speaker to a fixed +12V supply. It picks up too much mains hum if my table lamp's on, but otherwise seems to work. To measure the pulse rate I replace the 8R speaker with a 10R resistor and use the pulses to drive a charge pump (small capacitor and two diodes) into an RC (4uf & 10MR) whose voltage I monitor. Since V=C/Q this should give a voltage proportional to pulses/time I think

The well defined clicking I hear starts to turn into a 'squark' at around 600 volts (I'm working without an oscilloscope for the next few weeks). I think the flat part of the graph from about 600V up is the tube moving to the continuous conduction region, but rather than conducting continuously the tube capacitance and 4M7 ballast are causing oscillation by the same mechanism seen with the classic Neon lamp oscillator.

I'm a bit worried that this is a low voltage tube (400-600) and I've damaged it by running it at over 1000V in early tests.

Alan

MinorityCarrier

Mon Apr 19 2010, 03:26PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

I'm surprised Proud Mary hasn't got here first.

Your counts vs bias graph shows you're just starting to hit the operating voltage plateau of your GM tube. You keep gradually increasing the bias voltage and the counts should stay the same for some span of voltage. When the counts start to climb again stop!

You select an operating voltage in the middle of the plateau where the count-rate was unchanged.

Proud Mary

Mon Apr 19 2010, 03:33PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Nothing finishes off GM tubes more quickly than too small an anode resistor. I'd suggest changing this to 10M before doing anything further - it can always be reduced later if needs be. It is important that the anode resistor should be connected directly to the anode terminal by the shortest possible wiring.

As a rule of thumb, the cathode resistor/anode resistor ratio should be ~1:45, so your cathode resistor should be (preferred value) 220K if your anode resistor is 10M. The cathode resistor should also be connected directly to the cathode terminal with the shortest possible wire. The other end of the resistor should be soldered to Earth with minimal wiring, or a long tube like yours may well start to think of itself as an antenna, a hand-capacity detector, or passer-on of feedback from the amplifier.

Take off the signal from the cathode via a cap, say silver mica 50-100pF, plugged into a high input impedance first amp stage (a bootstrapped FET for example) so as to put the minimum load on the tube.

If your tube gets bogged down with stray capacitance from meandering wires, it will not achieve its minimum dead time, and is likely to become unstable. Excess capacitance in the anode circuit will also shorten the life of the tube, because all of the charge stored in that capacitance must be discharged through the gas in much longer pulses than would otherwise have been the case.

The squarking you describe sounds like the continuous conduction at the upper end of the plateau. Back off from there until the signal stops clicking altogether, and then park it midway between the two.

My crystal ball tells me that you should try Va 390V, a Russian preferred value for GM tubes.

It is possible that your tube may have given up the ghost in response to over-voltage, but not very likely, if the over-voltage was not too long sustained.
Even then, it is more likely to have changed its parameters, making it unsuitable for use as a replacement part, but as you don't know its original values it is not likely to be too much of a problem.

I ordered an SI-22 from the Ukraine a few weeks ago, and had hoped to get it by now, but for the volcano.

IntraWinding

Mon Apr 19 2010, 05:09PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

Thanks folks.

MinorityCarrier: I don't think that plateau at 600-700V is the Geiger plateau. I think that kink between 300 & 400V is more to do with it, but I could be wrong. Perhaps my test circuit isn't giving the tube a chance to operate properly yet. I need to redesign my test circuit and plot another graph before I'll have any confidence over this.

Proud Mary: I was using 10M with very short lead length to the Anode when I tried the tube at 1000V, and then only for minutes. I don't have any criteria for selecting the value of this resistor so just thought I'd give it a try at the other value I see used in many circuits.

The tube certainly works in the region of 390V but it would be nice to see the plateau on a graph before proceeding.

I originally figured that ideally the Geiger tube outer jacket (Cathode) should be earthed, so the signal would have to be picked up from the Anode side, (with a series: V+ -- 220K -- 10M --tube and a KV rated capacitor tapping off a signal between the resistors), but the thought of suffering unintentional ECT through headphones led me to try the Darlington approach. I figured holding the Cathode at the double Vbe of 1.2V would be as good as earthing it. I realise now I slipped up there, but I'm inclined to go back to taking the signal from the Anode side so I can properly earth the tube side. What do you think?

Alan

first previous 2 3 4 5 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob