Hi, it’s me again.
Everybody seems now to build Geiger counters, and the prices for GM tubes on ebay are rising constantly.
Well, I had some of them laying around, and was also able to buy additional ones for little money.
I already have 2 Geiger counters, one old from the german army and one that I built in the 80′s, after Tschernobyl.
But up to now I didn’t have one with a PC interface, so I built one.
Not only can you use such a thing for long term measurements and environmental monitoring, it also makes an excellent source for true random numbers.
The main reason why I did this is (besides Fukushima) that I recently bought a small board with an AVR USB microcontroller (ATMEGA32U4).
I played around with the excellent LUFA USB toolbox, compiled some demos, and than had the idea to use this as a basis for the counter.
And here’s the result:
How it works:
- The ATMEGA32U4 microcontroller is programmed with a firmware that generates a virtual serial port so you can just use a terminal on the PC to connect to it.
- The micro generates a PWM of about 4kHz and 50% duty cycle. This drives the boost converter consisting of Q1, L1, D1 and C6. Q1, L1 and C6 are salvaged from an old energy saving lamp.
- The generated high voltage (up to 600V) is divided by R7, R8 and R9, and measured by the ADC11 input of the micro.
- Depending on the measured voltage, the PWM is switched on and off. The desired voltage can be changed in the code, this allows it to adapt the voltage to the tube you use.
- R6, R10, R11 and T1 adapt the pulses from the GM tube to the input level of the micro.
- The micro counts the pulses with it’s internal hardware counter and drives the LED at every measured pulse.
- The micro sends the actual counts per minute (CPM)and counts per hour (CPH) to the PC.
The software isn’t finished yet. It works, but it’s a mess and I’ll add additional features like random number generation and maybe a fancy menu using ANSI command sequences.
p.s.: Some photos
The ATMEGA32U4 board that I bought from a chinese dealer on ebay. Everything containing an USB AVR is too expensive at this time, because people use it to hack their PS3 game consoles.
I hate script kiddies.
Top side of the Geiger board. The GM tube I used is a FHZ74 from Frieseke & Hoepfner. Old german military equipment.
The tube inside the metal case is a high dosis GM tube from Philips/Valvo.
Those things are rather insensitive and give a background count of about 2-3 counts per minute.
Nevertheless I was able to measure a difeerence between work and home.
At work I get about 45 counts per hour, but at home I get 80.
Maybe because my home is more than 200 meters higher and therefore gets more cosmic radiation.
The bottom side of the geiger board with LED.
And the whole thing together.
Not fancy, but working.
Ideas I have in mind:
- Improve counting and averaging
- Implement random number generation
- Add configurable warning levels with optical/acoustical signaling
- Add dosimeter functionality
- Add a feature to save calibration tables to EEPROM to calculate Sievert/Gray rates from the CPM rates for the given GM tube.
- Add the possibility to use more than one GM tube simultaneously (i.e. low rate + high rate tube)
- Add interfaces for ionization chambers and PIN diodes
- Build a boost converter for higher voltages (> 3000V). so that homemade GM tubes with AR/CO2 filling at normal atmospheric pressure can be used.
- Build a super simple version using an ATTiny controller with RS232 or I2C interface
- Add a realtime clock with battery backup
- Build a compact standalone version, featuring a rechargeable battery, that could be left outside for longer monitoring, saving the measurement data to internal or external memory.
- Build a handheld Version with LCD, keyboard and USB.