A Very Tiny Gamma Ray Detector


When you think of a radiation detector, you’re probably thinking of a Geiger tube and its high voltage circuitry. That isn’t the only way to measure gamma radiation, though, and [Alan] has a great circuit to measure even relatively weak radiation sources. It uses a very small photodiode, and draws so little power it’s perfect for projects with the smallest power budgets.

The detector circuit uses a miniature solar cell and a JFET wired up in a small brass tube to block most of the light and to offer some EM shielding. This, in turn, is attached to a small amplifier circuit with a LED, Piezo clicker, and in [Alan]’s case a small counter module. The photodiode is actually sensitive enough to detect the small amounts of gamma radiation produced from a smoke alarm americium source, and also registers [Alan]’s other more powerful radioactive sources.

The circuit only draws about 1mA, but [Alan] says he can probably get that down to a few micoAmps. A perfect radiation sensor for lightweight and low power applications, and gives us the inspiration to put a high altitude balloon project together.

48 thoughts on “A Very Tiny Gamma Ray Detector

      1. If you want a scintillation detector, then get a scintilation detector, and it will certainly work much better. It will also allow you to resolve energy. (i.e. gamma spectroscopy, radionuclide identification etc.)

        But it will be more expensive, bigger, fragile, considerably more complex, have higher power requirements, require a HV supply etc.

  1. We need one of these on every mobile phone made from now on… imagine a world populated with mobile, connected, radiation sensors… just try to smuggle some plutonium, or a pony nuke.

    1. DHS had that “let’s put radiation detectors everywhere!” idea too. All they’ve done so far is annoy a lot of cancer patients.

      Where radiation detectors have come in handy: ports, mostly for turning back a lot of goods with radioactive metal.

      1. Not just that-but at domestic steel mills in the scrapyard (most “new” steel contains a lot of “old” steel). A single load of radioactive scrap can ruin your whole day.

        As far as other forms of radiation detection-Argonne Natl. Labs invented a polymer a decade back that can detect neutron flux. I’m certain the TSA folks and the FBI have detectors based on that nowadays.

      2. Since the the detonation of the first nukes most materials are contaminated anyway, not much (they are essentially entirely safe) but enough to make them useless for certain applications.

        That gets really annoying if you have to do science where you need a radiation free hardware (like say detectors in particle-accelerators). Thats why WW1 ships that sank prior to the Manhatten project are so precious. The Metals in those are very clean and where sheltered against the fallout(s) on the bottom of the ocean.

        Another good thing of those radiation detectors? You can always find you bananas ;)

        1. They even used lead from roman shipwreck for neutrino detectors.

          Ship from WW1 can also contain radioactive source. Radium was routinely used for luminescent paints. Some radioactive metals where used in metallurgy pre-curie era.

        2. One of the first things I had to do was have a total body count done before I could start working at Savannah River Site and the table I had to lay down on had to be made from pre WW2 metal due to all metal after has some radioactive material of some sort.
          And yes the tech said that I did have a banana that am since there was a spike for potassium 40 a radioactive isotope that is in bananas.

          One thing would be hard to do is to calibrate the detector, since Gamma sources are not an everyday item.

          I even with a mica window I don’t know if this sensor will pick up alpha particles but might be a good beta. And if you have a old civil defense Geiger detector it has a beta source on the side of the case.

          1. No, it wont detect alpha. Alpha won’t even make it through the plastic over the photodiode. And it would probably suck at beta as well. These sensors are just not very sensitive.

  2. I wonder if it would get a bit more sensitive to (Gamma) radiation if you add some Diamond, Saphire or Glass in front of the Phototransistor to get the (Gamma-)radiation to slow down emitting Cherenkov radiation (light). Since the phototransitor is actually build for that it could work better.

    1. Note that Gamma radiation IS light. It traveling into sapphire/etc. won’t generate Cherenkov radiation, but it might result in the Gamma being diffracted slightly. Diamonds do fluoresce under X-ray/Gamma… so there might some potential advantage.

      1. At the minimum you would want an avalanche photo diode. THere are some miniature Scionex pmt based scintillates on ebay from time to time which would be even more sensitive.

  3. @Poose I heard about Co-60 loaded belt buckles, just what you don’t need as a consumer..

    I was horrified to see just how many were contaminated, over a lifetime ONE belt could add up to 400 chest X-rays worth, which is not a small amount of radiation at all.

    Re. detectors, I made one using a mono CMOS video camera from B&Q and a sheet of silver paint coated mica, which later evolved into a sheet of thin pyrolytic graphite
    as this is better for low energy particles such as alphas.
    Something like 75% make it through a 0.1mm sanded sheet :-)
    Decrease clock to increase sensitivity, the “sweet spot” is 4MHz instead of 13.5.
    Something about increasing charge accumulation maybe?

    Tried to publish it in electronics magazines but they said “too specialised”… :-(

  4. Cool idea. I think a lot of people are unaware of how many devices are capable of detecting radiation even though that is not their primary function. In fact 99% of the people reading this have a decent radiation detection device in their possesion right now. Your camera phone. As I posted in a comment above, there is a serious Android app to use the phone’s camera as a radiation detector and they have even done calibration on several devices using the camera and comparing it’s sensitivity to high end detection equipment. See the app here: http://goo.gl/r005D
    Here is a section of the app’s readme:
    The App is using the camera sensor to detect radiation, like a geiger mueller counter, of course with a smaller area. We tested several mobile phones at the Helmholtz research facility in Munich, using a professional radiation device in the range of 2-10 µGy/h till 1-10 Gy/h (CS137 and CO60). The CMOS sensors can detect primary gamma radiation and some higher beta radiation (depends on the shieldings in the mobile phone). Typically not going into saturation as most GM tubes. See FAQ on our homepage for what you can measure and how.

  5. http://4hv.org/e107_plugins/forum/forum_viewtopic.php?144983.0#post_144992



    Unfortunately no pictures that I am aware of, the problem is that it wasn’t as reliable as I’d hoped possibly due to the alphas “sand blasting” the row and column drivers for the sensor. If I made this again it would include a lead shield and just expose the centre of the chip…

  6. Oh, and http://4hv.org/e107_plugins/forum/forum_viewtopic.php?75355.post

    If anyone wants to try this, please PM me on the forum or mandoline at cwgsy dit dah net yadayada.
    remove at and morse to mail.

    I did notice that really cheap and nasty ™ old webcams from laptops are quite good for this, even the plastic used passes a few alphas.
    What would be really nice is if manufacturers used bare chips, but they probably don’t because they corrode in time like I found.

    The problem would be fixed if the sensor was encased in say an Aixiz module with PG end window held in place with Epoxy and then silver painted and electroplated with a mask to protect the centre of the PG, and then backfilled with say argon and some silica gel..

  7. If I were to update this, format it as a scientific quality report with data, and submit it to a major journal such as “Nature” or “Scientific American” do you think they’d publish it?

    @Longfist yes I just noticed (facepalm!)

    This also has implications for detecting of other “interesting” nuclides such as 40K and 137Cs at a substantial distance, the details of which are yet to be worked out.
    Essentially this is a modification that might detect alpha particle entanglement which could carry a lot further than a simple alpha and also show information about the source not obtainable by any other method… !


  8. There are quite of few project using PIN photodiodes already, mostly using the BPW-34, but also other types such as the BPW21, S1223, etc. There are a few that go even further, using large area, ultra sensitive photodiodes such as the PS100-7-CER (also expensive).
    The main disadvantage is these constructions are quite insensitive when it comes to gamma radiation, and totally worthless for beta and alpha which are stopped in the plastic case of the tiny photodiodes.
    Some builders went even further to dismantle the epoxy/glass casing to expose the semiconductor junction of the photodiode, and so making them able to detect alpha and beta as well.
    The pulse is proportional to the energy of the incoming radiation allowing experimenters to create simple spectrography devices but the drawback is high energy radiation will not transfer the entire energy to the tiny photodiode layer, giving misleading results.

    An interesting toy, but still a toy.

  9. Also worth mentioning, in case people haven’t found this out some of the Apple desktops use an exposed chip on their cameras.
    ie 19″ and 22″ Mac flat panels with the iSight camera, v2 and v3 use these.
    I also located a similar sensor used in older Nokias and despite the color filters it still works fine.
    A valuable method of “nuking” filters is a UV laser and this may ruin the camera for normal use but the alpha sensitivity remains.

  10. Whoa.. you can get ZnS:Cu powder on ebay although ZnS:Ag sheet from unitednuclear works much better.
    look for the guy selling “electroluminescent powder” , although it isn’t quite as good for many uses it is fine.
    Also a little note, for it to work at all as an alpha sensor it has to be in direct contact with clean graphite.
    I found that even Superglue ™ fumes ruin the effect.
    I’m trying this exact project with some thinned down silicon from a dead 8GB memory card as its a lot thinner than
    the graphite previously used.

  11. Update:
    I found that a method originally invented by Andre Geim works well.
    Take the powder and pick up a “blob” (actually just enough to be barely visible) and then stick the tape directly onto the graphite having first carefully cleaned and dried it.
    This method does have the advantage that the sheet is essentially a monolayer already so the efficiency should be quite high.
    Also worth noting, if you are like me and have 2006 vintage Osram GlacierGlo EL material its worth drying it out in a low oven before use for 4-6 hours at 130C.
    This also works for damp strontium glow powder to get it to mix properly with Epoxy and is very effective indeed.

  12. Scintillation!
    Also found that bismuth mixed with phosphor (eg ZnS) is the “magic sauce” needed to sense gammas and X-rays.
    The recipe I am experimenting with at the moment is Bi2O3 (from jarmondbrinkley on ebay) and mix it with activated carbon, heat up to around 550-660C for a few hours to get finely divided Bi then make the ZnS as normal but substituting Bi for Cu.
    Appears that some samples of SrAl2EuO5 are contaminated with trace amounts of Bi and Cl to change the emitted wavelength of light.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.