Hello,
I'm interested in building a geiger counter, there are a lot of guide for arduino, but no one explain the difference between sensors.
on ebay I've found a lot of them, but I don't know if the difference is only the precision or something else.
I need it just to play with common object, maybe some ambient scan
I have a primer here.
http://albert-cordova.com/Cdv700/geiger-counters.htm
Yes there are huge differences. What you will need to even get started is a high voltage power supply that is appropriate for the tube you select. somewhere between 300 and 1000 VDC is most common.
Some tubes are just for gamma/X-rays, Some are gamma and beta, still others are good at Gamma, Beta and Alpha tubes.
Beware of over voltage in your tube it will shorten its life or kill it! Especially the cheap Russian tubes that operate at 300 volts, 600 volts for one second is fatal.
Halogen quenched tubes have a longer life than organically quenched tubes as a general rule.
For a newbie I would recommend a gamma only tube as they tend to be more robust that the other tubes. You will need to know the dead time of the tube or how long it takes to reset between events.
At first you should just worry about counts per minute or (CPM) once you get that part down you can attempt to convert the CPM readings to m/R or m/SV but you will need to know the calibration constant in order to do that.
I hope this helps.
If you have more questions feel free to ask.
Tracy
Sadly, there has been a huge up-tick of interest in Geiger counters.
I have project plans for an Arduino Geiger counter on my blog . . .
They include the circuit for a 555 based 500V power supply and info and links on Geiger tubes.
I agree with Radioactive that a metal jacket tube is the best bet for a start. They are indeed more robust. (I've heard that sometimes the surplus tubes that use mica windows and such, don't survive the pressure difference experienced in shipping.)
I've had good luck with SBM-20 metal tube, and would recommended it. I found it worked pretty well in detecting beta if the sample is held near the tube. (I've gotten up to 6000 CPM with old lantern mantles that used thorium-321.)
I found the glass walled tube, such as the CL-3BG to be much less sensitive.
Again, there is more info in the blog.
I'd also like to say here that thoughts, prayers, and hopes go out to the Japanese people who are dealing with this current tragedy. I wish you all the best.
John
Nice work bHogan. 
A nice place to cross reference tubes and find there operating characteristics is LND INC.
If the tube has a number and is commonly used LND INC manufactures a replacement and has it in there DBase.
I have used CCFL inverters with a bridge rectifier and a hv cap to supply the high voltage to GM tubes as well as photomultiplier tubes for scintillators.
Thanks for some very useful information.
I have a tube on its way from the Ukraine. I was contemplating using the 300V inverter from a single use camera flash gun for the HT generation.
I hear the comments about the dangers of over-voltage. If the voltage is less than specified, does it just reduce the sensitivity, or simply stop it working at all?
Running the tube in an under voltage will reduce sensitivity and increase dead time. Dead time is the time the tube needs to reset itself from an event. At some point there will not be enough voltage for the tube to fire as you decrease the voltage.
This tube will operate at 450 volts even though its a 500 volt tube, At about 400 volts it ceases to fire even when exposed to a strong radioactive source.
http://www.lndinc.com/products/149/
Just to report back - using the pcb from the flash unit of a disposable camera worked really well.
I replaced the big 330V electrolytic with a 1nF 1kv cap and tweeked the voltage on my bench PSU to get 400V (right for my tube).
I used a 10M resistor from 400V to the anode of the tube and a 470K resistor from cathode to GND. This gave me a 4V pulse for each event across the 470k resistor - just right for an Arduino!
As a test source I took apart a smoke detector which ups it from about 20 counts per min to several hundred - my tube is beta and gamma only.
Cool you are detecting the very weak gammas from the AM241. It produces no beta and around 20,000 alphas a second.
Am-241
Atomic Number : 95
Half-Life : 432.2 y
Jp : 5/2-
Sn : 6641.00 keV
Sp : 4480.10 keV
Possible Parents
Parent Fraction (%) Decay Mode
Pu-241 99.998 % b-
Decay Products
Daughter Fraction (%) Decay Mode
Np-237 100 % a
???? 4.30E-10 % SF
Emission Products
Number of Alphas : 32
Number of Betas : 0
Number of Gammas : 172
Number of X-Rays : 22
Si:
Just to report back - using the pcb from the flash unit of a disposable camera worked really well.I replaced the big 330V electrolytic with a 1nF 1kv cap and tweeked the voltage on my bench PSU to get 400V (right for my tube).
I used a 10M resistor from 400V to the anode of the tube and a 470K resistor from cathode to GND. This gave me a 4V pulse for each event across the 470k resistor - just right for an Arduino!
As a test source I took apart a smoke detector which ups it from about 20 counts per min to several hundred - my tube is beta and gamma only.
OK you get my vote for inventiveness today. 8) Well done!
I have been looking at the kit from SparkFun. What you did is way too inventive for my modest abilities.
We use these things professionally -- and the ones we use are real expensive but don't even do what we need done. So as soon as the rush on radioactivity detectors cools down....
Thanks everyone.
the ones we use are real expensive
At the risk of sounding smug -
Tube - £13 ($20)
Single use camera - free
Smoke detector - £5 ($8)
Hi all!
Cooking Hacks has launched Radiation Sensor Board. There are two versions. One with Geiger Tube and the another one without it to use with your own Geiger Tubes.
Way back in 1986 (Chernobyl) I resurrected my 1967 giger counter project and revamped for a computer input.
The text is here:-
http://www.doc.mmu.ac.uk/STAFF/A.Wiseman/Acorn/BodyBuild/BB86.html
And here are the diagrams:-
