Solid-state radiation sensor

Have questions about a SparkFun product or board? This is the place to be.

Moderators: phalanx, TS-Tim

Postby silic0re » Tue Apr 17, 2007 5:51 pm

I agree, their response was very helpful -- they even pointed me to some potential circuit diagrams. Maybe it was my inquisitive nature? :)

I don't really have a specific intended source that I'm thinking of measuring -- I just think it'd be neat to construct a sensing device that could measure radiation. I've often thought about it as I walk through the halls at school -- there is a door around the corner from me now that has a radiation symbol on it, and I'm but 100 meters from a nuclear reactor. I often wonder exactly how much radiation leaks out, and where, as I walk by... :)
Posts: 349
Joined: Wed Apr 26, 2006 5:35 pm

Postby awright » Wed Apr 18, 2007 12:10 am

I think you can definitely build a radiation detector and that the project could be interesting and fun. But to make a useful device, you would benefit from a little study of radiation measurement technology. I say this because, while I know very little about radiation measurement myself, I do know that differentiating between various types of radiation and between hazardous and non-hazardous radiation is not a trivial matter.

Just getting a count of the number of particles detected by a G-M tube or the number of pulses from a scintillator does not necessarily tell you whether you have a radiation hazard. It is a much more complex issue than that.

One factor is the material and thickness of the window or barrier between the radiation source and the detector or scintillator. Another is the chemistry of the scintillator. And then, once you have a device that responds in some way to radiation particles, you have the problem of calibration and conversion of your output to some standardized metric of radiation risk or exposure. As I mentioned above, a Geiger Counter went wild when held next to my chest after I had gotten an internal exam involving consuming a barium milkshake. But I don't think I was in danger.

There is some remote analogy with building a sound level meter at home. You can buy a microphone capsule for less than a dollar, add an amplifier and rectifier, and send the detected DC to a meter. This would be somewhat analogous to building a device that responds to radiation.

Now we "calibrate" it by playing a steady sound on our HiFi and making it read the same as a good SLM (Sound Level Meter).

Do we now have a useable SLM? No.

1. Does your rectifier detect peak, average, quasi-RMS or true RMS?

2. How fast does your meter respond to time varying signals and how does it average them?

3. What is the frequency response of your microphone?

4. What is the frequency response of your amplifier?

5. What is the directionality of your microphone and how does it vary with frequency?

6. What is the noise floor of your microphone and amplifier?

7. How does temperature affect your system?

8. How does humidity affect your system?

9. How does your frequency response compare to internationally standardized frequency weighting networks and can you switch between the Overall, A, B, and C networks (not to mention the D and others I've lost track of)?

OK. I've made my point. Once you have taken care of all the questions above (and many others) you may have a useable SLM that gives you answers comparable to standard SLMs.

I'm not trying to discourage you here. In fact, I do think the radiation meter project could be fun. I'm just trying to encourage you to investigate what it takes to come up with a radiation detector that gives you useful information. The data is all there in the literature.

Posts: 36
Joined: Tue May 24, 2005 1:19 am
Location: Oakland, CA USA


Postby rkeyes » Fri Aug 17, 2007 7:09 pm

Hrm too bad no one has posted on this thread in over a year. Not 'active'

I am quite interested in this stuff. I'd like to hear if any progress has been made within the past year or so.

I recall, many years ago when I was a kid, reading in a magazine, perhaps Popular Electronics or whatever, a project to make a home-made radiation detector using a $50 cell, I believe the material was Cadmium Telluride. Something like that. It's difficult to remember, this was probably 25 years ago.
Posts: 5
Joined: Thu Aug 16, 2007 12:14 pm
Location: Cambridge Massachusetts

Postby saipan59 » Fri Aug 17, 2007 7:37 pm

Hrm too bad no one has posted on this thread in over a year.[/list]

Actually it's only been 4 months since the last post. :)
I did some experiments with a circuit like the one on the Hamamatsu site mentioned previously.
It 'worked', but the efficiency was really low. The circuit produced pulses in the 10-20 mV range that were clearly visible on my o-scope. But the pulses were very infrequent.
For example, using an old military glow-in-the-dark switch handle (contains Radium) about 3/8" away from the diode, there was only one pulse every second or two. A standard Geiger counter gets dozens of hits per second with the same radiation source. My scintillation probe gets *hundreds* of hits per second.

I used an ordinary IR photodiode. I tried 3 different types, but all were similar. If you look around on the Hamamatsu page, you'll find diodes that are specialized for radiation detection, but I think they are very expensive, and not carried by the standard electronics parts places.

So, the best 'bang for the buck' is still a Geiger tube.

Posts: 228
Joined: Wed Nov 22, 2006 9:04 am
Location: Colo. Springs

Postby silic0re » Mon Aug 20, 2007 12:34 pm

I also assembled a circuit like the one on the Hamamatsu site with an IR photodiode recommended by the Gammawatch folks, but unfortunately I wasn't able to sheild the IR photodiode from incident light (even after wrapping it in several layers of tinfoil and electrical tape, it would still output a voltage when light hit it). I also had quite a bit of trouble shielding the amplifier from noise, and got a wierd progressive ramping-up effect of the output, even without an input. I'm not too experienced with amplifiers, but since it was built on a breadboard, and was amplifying a great deal (probably well into the noise level of the input signal), it seems like I didn't really have a favorable signal-to-noise ratio with the setup I used. There's also the issue of being unable to sheild the photodiode from incident light too, of course :).

The circuit did end up being able to detect when pretty much anything in the room was turned on, or even plugged in. That also makes me think shielding is a huge issue with this type of circuit :).

I contacted the person who makes the Gammawatch again and reported my progress, and he actually said that my results were consistant with what they initially had, and that it actually takes quite a bit of RF sheilding and playing around to get things to work. Quite unfortunate...
Posts: 349
Joined: Wed Apr 26, 2006 5:35 pm

Postby saipan59 » Mon Aug 20, 2007 1:46 pm

I built my unit inside a box that was originally an LNA from old-style C-band satellite TV. The box is roughly a 2-inch cube, machined out of solid aluminum, with a screw-on lid. It originally had a female 'N' connector on each end - I left one in place to be the signal output, which provided a convenient and well-shielded connection to my o-scope. I removed the other 'N', and mounted the diode behind the hole. A piece of black electrical tape covers the hole to keep out light.
Inside the box is the diode, the op-amp circuit, and a 78L05 regulator which powers the op-amp.
As I recall, the noise output was in the 5 mV range. The pulses were 10-20 mV (big enough to be clearly seen on the scope), and were a couple uS in width.
The op-amp I used was a Microchip MCP6022.

I found that it helps to put as much reverse-bias on the diode as possible. For me, the limit was the rating of the 78L05 (30V).
Note that some PIN diodes have a PRV that is in that area too, so if you go too high you could fry it.

Posts: 228
Joined: Wed Nov 22, 2006 9:04 am
Location: Colo. Springs

Postby bigglez » Tue Aug 21, 2007 5:11 pm

saipan59 wrote:I found that it helps to put as much reverse-bias on the diode as possible. For me, the limit was the rating of the 78L05 (30V).
Note that some PIN diodes have a PRV that is in that area too, so if you go too high you could fry it.

Greetings Pete et al.,

Now that brings back some memories of using PIN diodes to detect a laser beam that was modulated by an RF signal (initially in the 10MHz range, but later versions at 30, 100 and 500MHz).

The PIN diode was reverse biased to 90V DC, and the recovered signal capcitively coupled to a DMOS dual gate RF FET (found to have the lowest intermod performance - the desired signal was often masked by adjacent laser noise).

To keep other signals out of the pre-amp we used machined alloy boxes, with many small perimeter screws (at about 10mm spacing) and a BNC connector for the RF out with a D-Sub nine pin for the DC power, PIN diode bias and control signals.

Good Luck, keep posting your progress. Comments Welcome!
No Longer Sharing My ID
I'm over 13 & support COPPA regulations
User avatar
Support Volunteer
Posts: 1553
Joined: Mon Sep 26, 2005 1:25 pm


Return to SparkFun Product Questions

Who is online

Users browsing this forum: Bing [Bot] and 2 guests