Portable radiation detector based on a ionization chamber.
The KiCad PCB sources and BOM can be found in the repository's subfolders.
Check out https://hackaday.io/project/27508-open-radiation-detector for more information.
-
October 2017: Designed, build and document a prototype to participate in the Hackaday Prize.
-
5/May/2017: The radiation sensor concept has been incorporated into a connected sensor board. It looks like the effect of nearby electrostatic fields still affects the voltage output, regardless of the back plane being connected to either guard voltage or ground (maybe this is due to the large size of the electrode?). The board uses the PNP configuration polarized with 10 MOhm 1% and an op-amp driver (LM2904DR) to minimize the effect of ADC current consumption. It includes a duplicate circuit unconnected from the electrode, to be used as a compensation for temperature effects.
- 15/Feb/2017: It works! At first the result with the new transistors had lots of noise as well. But adding a copper plane behind the main electrode seems to have done the trick. So maybe the transistors were not the problem after all! The pictures show a new prototype being tested with and without a radiation source. Its response ranges from 70mV to 160mV and it is way more stable than before. Schematic is Charles Wenzel's Experimenters Ionization Chamber with an FMMT634TA NPN darlington transistor.
- 8/Feb/2017: The Darlington transistors used until now (MMBT6427, MMBTA14, MMBTA28, MMBTA63, MMBTA64) may not be suitable for detecting the low currents involved in this detection problem (nA or even pA). Yet again I have resorted to Charles Wenzel's awesome website. A proper surface-mount alternative to the MPSAW45A power darlington transistor may be the FMMT634 and FMMT734. So I've ordered a bunch of these transistors in order to test the concept again.
- Week of 7/Feb/2017: The new approach is to use an actual ion chamber that uses a copper electrode insulated from ground by an outer guard ring. The chamber itself is made of a standard size, commercially available PCB shielding can (in the pictures you can see my attempt at creating one with tin sheet from a metal can). The first prototype actually seemed to have some reaction during the tests. However the second prototype did not. G-code to mill the electrode.
-
On a second thought, the "finger electrodes" may not be a good idea. Since there is a large contact area between each finger, there surely is a lot of leakage current that will mask-out any ion-induced current (one of the biggest concerns explained by Charles Wenzel's ion chamber website)
-
19/Jan/2017 Finally got a .9uCi AM241 source to test, it will be the benchmark for the detector. Unfortunately, the first attempts with the amplifier circuit yield very noisy measurements, that mainly depend on wind and humidity. Surprisingly (to me), it is also very dependent on static electricity. I've also tested again the spark particle detector, but there seems to be absolutely no relation between the sparks and the proximity of the radiation source. Maybe the finger electrodes are too "pointy" because of the manufacture method (milling)? pic1 pic2
-
Week of 9/Jan/2017: Simulation of circuit amplifiers "click type", "LED scale type"
-
Self-ask the question: How to make a simple low-cost radiation detector? Begin to learn about ionization chambers thanks to Charles Wenzel's website.
-
Week of 1/Jan/2017: First learn of spark particle detectors. Design, build and test a PCB electrode (but without any radiation source to be tested). video
This work is licensed under a Creative Commons Attribution 4.0 International License. https://creativecommons.org/licenses/by/4.0/
Garcia-Saura, C. (2017). Open Radiation Detector. GitHub. https://doi.org/10.5281/ZENODO.10446305
Latex / Bibtex citation:
@misc{garciasaura2017rad,
author = {Garcia-Saura, Carlos},
title = {Open Radiation Detector},
publisher = {GitHub},
year = {2017},
doi = {10.5281/ZENODO.10446305},
url = {https://doi.org/10.5281/ZENODO.10446305}
}