Well, you make new revisions if the old ones are not good enough.
In this case, it was noise due to the modular nature of Rev2.0. Now everything is on one board and it works great. I can now always use linear scale within the RedPitaya gamma spectrometer app. Otherwise nothing much changed. The CSA is the same, so is the low and high voltage power supply. I still use a NMH0505SC and a Hamamatsu C10940-53.
What changed is, that I added a AtMega8 for displaying the high voltage on a 0.96″ Oled I2C display.
Also I added a 1206 footprint for the CSA output capacitor and a divider followed by a buffer/amplifier, in case the signal of the CSA is too large or too small. I also can calibrate the divided high voltage with a ten turn trimpot now.
To use your RedPitaya as a Gamma spectrometer you connect your CSA to it and start the MCA software of your choosing. It is really as simple as that. The difficulty here is to make a CSA and get some HV PSU for the detector.
The LTC6228 is especially usefull for the purpose of a CSA. It is very low noise. Only 0.88nV/Sqrt(Hz). Also very fast. The GBWP is 890Mhz. Perfect for fast rising signals. In addition to all that the pinout offers easy layout, due to the output being made available on a special feedback pin on pin one, with pin two being the negative input. USB is only for power. Rev4.0 might digitize the signal on board.
Here’s the schematic and the board:
In order to see how much of a improvement Rev3.0 brings I have here these two spectra of my Ra226 watch. Spectrum was made with REV 3.0 and spectrum two with Rev2.0. It’s not the same detector, but that doesn’t matter. The noise level was higher with any of my detectors.
