11/13/2017 at 19:21 •
It was very rewarding to see our project reach the top 10 from the >1000 entries to the Hackaday prize :-)
Technology needs to be free for all, and this is only possible with a well motivated and supported community of makers. We are very grateful to Hackaday for organizing such a big contest to promote open projects. We are also thankful to the Open Source Hardware and Software Demonstration (OSHWDem) that took place La Coruña (Spain) this weekend, it also contributed to a great diffusion of the Open Radiation Detector and many other projects.
Some people has contacted us to test the detector with Radon. We will be manufacturing a batch of improved sensors in the near future that shouldn't require calibration and have a digital interface. Get in touch if you are interested!
10/30/2017 at 23:38 •
We are working on a new design that improves upon versions 0.9 and 1.0:
- It will have rear-facing LEDs to allow easy readout from both sides of the detector.
- Instead of using a Darlington arrangement as the frontend, we are investigating whether it is possible to directly use a really good operational amplifier. This should have better linearity, temperature stability, and maybe allow calibration.
- Also, it will use an SMD buzzer to further facilitate pick-and-place assembly.
For the frontend amplifier, we are using the µCurrent by EEVblog as a reference.
Do you have any ideas on how to improve the design? Please share them in the comments!
10/19/2017 at 11:44 •
Edited 21/Oct/2017: We are very excited to be on the Hackaday Prize finals! Thank you Hackaday for this great opportunity to share our project with the world :-)
10/16/2017 at 15:36 •
After testing the prototype and making some minor corrections, we have released the KiCad sources and BOM for version 0.9. Bear in mind it is still a very early version of the detector, suitable for tinkerers only! Here is the link to our GitHub repository:
We are already working in the next version of the detector, one that supports the buzzer and a row of LEDs that represent radiation intensity with a scale. (a strip will be much easier to read than a single led's brightness gradient).
I want to take this opportunity to thank Hackaday for promoting the creation of open source projects. The Hackaday Prize has been the biggest motivation for me to advance this project, document it and finally release it. I really hope this project can help to lower the cost and facilitate the detection of radiation sources worldwide.
I also want to thank the Grupo de Neurocomputación Biológica at Universidad Autónoma de Madrid for providing the material resources to create the prototypes, which have accelerated the project a lot.
Finally, the project has reached a point where the power of the community will help take this project to the next level. Did you find the project interesting? Do you have any suggestions to improve detection effectiveness? Please share your thoughts in the comments!
10/15/2017 at 01:37 •
The prototype has been successful! Our latest prototype can measure radiation from the source in a smoke detector.
At first the signal was too dim and we needed to adjust the circuit's gain, after that the sensor could work by itself and notify with the on-board LED (link to video).
10/14/2017 at 10:01 •
Here are some detailed pictures on the construction of the most important part of the detector: the inner amplifying electrode.
The central electrode is a surface-mount stud that captures the tiny ionization currents within the chamber. It is tightly coupled with the Darlington transistor that amplifies the signal (see the previous log entry for the schematic), this is done in order to reduce electrical noise.
10/11/2017 at 22:57 •
Signal amplification and conditioning has been implemented with the following circuit:
10/11/2017 at 00:15 •
This project provides two different things: a new design of manufacture-friendly ion chamber, and also a proof-of-concept device that demonstrates its sensitivity to radiation.
From the start, the idea has been to create a hand-held detector that can be used to check if an object is radioactive or not.
10/09/2017 at 22:45 •
A big problem with our initial ion chamber prototypes was that any nearby electrostatic fields had a large effect over the radiation measurements.
Even the temperature-corrected version of the radiation detector was unusable: it produced false detections when bringing a charged pen near the detector (a harmless plastic pen that had been electrically charged by rubbing it with a cloth).
After many test we established that the problem was caused by the electrode topology. To explain this, here is a diagram of how the early prototypes looked:
10/08/2017 at 23:20 •
The circuit we are using to measure radiation is derived from a DIY detector by Charles Wenzel: http://www.techlib.com/science/ion.html. It uses a high-gain Darlington transistor to amplify the tiny currents generated between the inner electrode and the outer shield.
For our design we have chosen the FMMT734 transistor, an up-to-date part with standard SOT-23 surface mount footprint. We use the PNP configuration rather than the commonly used NPN; this way it is possible to have the outer shield at ground voltage.
However in our first tests this circuit produced measurements that were very dependent on temperature. When temperature raises around the detector, the gain of the Darlington transistor is reduced and the output signal is substantially diminished.
This is why the fourth iteration incorporates two transistors: one of them is connected to the electrode, and the other one is left open. The radiation measurement is then corrected by subtracting the reference output voltage from the signal coming from the main amplifier.
Here are two pictures of the fourth prototype. The radiation sensor is the large metallic box, the rest of components are part of a plain old, boring IoT box (Internet of Things is soo 2016..).
The fourth prototype was successful - there is a clear signal whenever the radiation source is put close to the detector. Also, the offset and the effect of temperature are removed by subtracting the reference signal from the unconnected reference transistor:
The long-term test in a WiFi connected platform has also shown that the sensor is not capable of measuring the background radiation levels. Temperature changes over the course of hours and days, and this still has a notable effect that obfuscates any traces of low radiation levels (even after correcting with the reference transistor).
However the sensor does produce a signal when radiation sources are close to the detector. We believe this sensor will be useful in a portable hand-held device that can be brought near objects to check if they are radioactive or not.
Stay tuned for more updates, and tell us what you think! :-)