Small radiation detector
Posted: Tue Jul 27, 2010 2:20 pm
Lutz has suggested we group-design a small, very portable, inexpensive radiation detector, which I think is a great idea -- you can't have too many when working on nuclear stuff.
I think first we need to define some specs for the design, so we can get going on the best way to meet them.
For example, a simple geiger counter will tell you "here be radiation" but not information on what type, and differing types affect it differently, so in a lot of cases
you can't tell between a lot of low energy and a little high energy stuff, or between alpha radiation and gamma (though you can use a sheet of paper to stop the alphas and see if that makes a difference. Ditto scintillators, though you can at least get some more information from the pulse heights returned if the system is linear (not a foregone conclusion), at the cost of more complexity in the electronic back end.
Neutrons are their own game entirely, and getting energy data from them is problematic, though I've heard of it being done with at least some resolution with a 3He tube system.
Plastic scintillators see them fine if they are fast, but with random pulse heights - a recoil proton can get any fraction of the impinging energy.
If we are interested in human dosage effects, we would have to have a lot more information than a simple geiger can give us, or a simple scintillator. The trade-off there is
that if it's hard to see at all, it's probably not going to be hurting you either, so sensitivity is less important for that case. But selectivity is vital. Some kinds or energies of radiation do more damage to tissue per event than others, and the range is fairly wide for all of them.
So I believe the place to begin is with a specification -- do we want a good prospecting/survey meter? A human-calibrated dosimeter? A knock around the lab basic safety device?
We can't do them all small and cheap in the same device -- so we have to start by choosing what it is we want. Extreme sensitivity isn't ever going to be very small, as you
have to have enough detector area to intercept the radiation when the events are rare. Things that see pulse heights for human dosage are more complex electronically (and in software).
For example, this is a problem with Bill's "Henny Penny" which is a plastic scint/counter-only -- it goes wild with radiation that won't hurt you, because it sees *everything* and counts it all the same.
The unit is a TSA systems PRM-470b. It counts like 10/second on cosmic background where a 2" pancake geiger counts about 1/second -- it must be seeing the individual particles in showers. And that one counts a U glazed plate at 700 counts per second and the safety alarm goes off -- a bit much. Good for prospecting, but lousy around the lab.
So, what are we after with this idea? Let's define that first, then worry how to make one real. Easier to keep your eyes on the prize once you know what the prize is.
I would recommend against using any crystal that needs sealing. I've seen far too many "professional sealed" ones that didn't keep their seal. I have a shopping bag full of bad ones already that we got on ebay and other places. Further, if we are going for small form factor here -- you can't machine them for a custom fit. In addition, anything with iodine in it will activate around neutrons and be ruined.
This is responsible for the well-sealed but still bad NaI:Tl heads we've also bought -- they are counting their own activity, having been near a neutron source at some time.
BGO is pretty good stuff for a lot of things, and not only can you get it, I already have about 10 of them. Used end on they are decent (not as good as NaI) for spectroscopy, and the end is 1/4" by 1/2" so you can use them with a small photodetector. They don't need sealing in air. I use a chunk off one I dropped inside the fusor as a visual indicator -- it lights up just fine in there (bright) in the X rays inside the tank.
The downside of any scint vs a geiger or proportional tube is that you won't be able to see alpha radiation usually -- they are hard to make light tight enough, and so by the time you get there, they stop alphas on the way in. With a small geiger tube (really easy to make, actually -- I've done a few here) you can have a thin enough window to let them in. So that's a consideration too, since most natural radio-actives give off alphas, mainly, which even a sheet of plain paper will stop and a sheet of Al foil (what you usually wrap a scint in) will really-really stop.
I think first we need to define some specs for the design, so we can get going on the best way to meet them.
For example, a simple geiger counter will tell you "here be radiation" but not information on what type, and differing types affect it differently, so in a lot of cases
you can't tell between a lot of low energy and a little high energy stuff, or between alpha radiation and gamma (though you can use a sheet of paper to stop the alphas and see if that makes a difference. Ditto scintillators, though you can at least get some more information from the pulse heights returned if the system is linear (not a foregone conclusion), at the cost of more complexity in the electronic back end.
Neutrons are their own game entirely, and getting energy data from them is problematic, though I've heard of it being done with at least some resolution with a 3He tube system.
Plastic scintillators see them fine if they are fast, but with random pulse heights - a recoil proton can get any fraction of the impinging energy.
If we are interested in human dosage effects, we would have to have a lot more information than a simple geiger can give us, or a simple scintillator. The trade-off there is
that if it's hard to see at all, it's probably not going to be hurting you either, so sensitivity is less important for that case. But selectivity is vital. Some kinds or energies of radiation do more damage to tissue per event than others, and the range is fairly wide for all of them.
So I believe the place to begin is with a specification -- do we want a good prospecting/survey meter? A human-calibrated dosimeter? A knock around the lab basic safety device?
We can't do them all small and cheap in the same device -- so we have to start by choosing what it is we want. Extreme sensitivity isn't ever going to be very small, as you
have to have enough detector area to intercept the radiation when the events are rare. Things that see pulse heights for human dosage are more complex electronically (and in software).
For example, this is a problem with Bill's "Henny Penny" which is a plastic scint/counter-only -- it goes wild with radiation that won't hurt you, because it sees *everything* and counts it all the same.
The unit is a TSA systems PRM-470b. It counts like 10/second on cosmic background where a 2" pancake geiger counts about 1/second -- it must be seeing the individual particles in showers. And that one counts a U glazed plate at 700 counts per second and the safety alarm goes off -- a bit much. Good for prospecting, but lousy around the lab.
So, what are we after with this idea? Let's define that first, then worry how to make one real. Easier to keep your eyes on the prize once you know what the prize is.
I would recommend against using any crystal that needs sealing. I've seen far too many "professional sealed" ones that didn't keep their seal. I have a shopping bag full of bad ones already that we got on ebay and other places. Further, if we are going for small form factor here -- you can't machine them for a custom fit. In addition, anything with iodine in it will activate around neutrons and be ruined.
This is responsible for the well-sealed but still bad NaI:Tl heads we've also bought -- they are counting their own activity, having been near a neutron source at some time.
BGO is pretty good stuff for a lot of things, and not only can you get it, I already have about 10 of them. Used end on they are decent (not as good as NaI) for spectroscopy, and the end is 1/4" by 1/2" so you can use them with a small photodetector. They don't need sealing in air. I use a chunk off one I dropped inside the fusor as a visual indicator -- it lights up just fine in there (bright) in the X rays inside the tank.
The downside of any scint vs a geiger or proportional tube is that you won't be able to see alpha radiation usually -- they are hard to make light tight enough, and so by the time you get there, they stop alphas on the way in. With a small geiger tube (really easy to make, actually -- I've done a few here) you can have a thin enough window to let them in. So that's a consideration too, since most natural radio-actives give off alphas, mainly, which even a sheet of plain paper will stop and a sheet of Al foil (what you usually wrap a scint in) will really-really stop.