Poblems binging it to work

[beldowsk]

Hello,

I have worked through the troubleshooting points, unfortunately without finding an error.

2023_02_20 Fotos.zip

The assembled board has an internal resistance of 9.71 kOhm. A charged 9.6V battery is installed.

A "Quantum Pendant" serves as a test object. A radioactive pendant from the esoteric field. With a GammaScout it delivers the following measured values:
α + β + γ 49.6 cpm - 19 cpm (background = 30.6 cpm
β + γ 44.00 cpm - 19 cpm (background) = 25.0 cpm
γ 25.4 cpm - 17.1 (background) = 8.3 cpm

I use an onboard chip in my desktop as the AD converter. A Realtek S1220A (32 bit, 192 kHz, 30 dB input gain).

In the online oscilloscope https://ozel.github.io/DIY_particle_detector/data_recording_software/webGui/ it just fidgets around the zero line.

WaveWatch ran under processing 4th images first without and than with power supply.

The Soundcard Scope first without and than with power.

Greetings Andreas

[SteveF161]

Two things which I hope are useful:

1. A sophisticated ADC and soundcard are actually a hindrance as noise suppression interferes with the signal. I used a very cheap (< 10 GBP) , unsophisticated USB device and it worked beautifully. See my posts from 2020
2. Your cable looks like a cheap phono cable with several interposed adapters resulting in poor shielding. I found my device was very sensitive to electrical noise and a properly shielded connection is essential.

[beldowsk]

Why don't I get such images with the DIY Particle Detector or the WaveWatch? Is it because of the 32 bit resolution? The Particle Detector shows "sampling: 2048@192 kHz" and the WaveWatch is set to "sampel_freq = 192000". Then it can't be because of the sampling rate?!

[beldowsk]

Or is it the special driver


I have 3 Modes: Low Latency, Standard, Stable

[beldowsk]

90% input sensitivity UR22C, 75mV/Div, Trigger -80mV, min. -339mV, max. 117 mV, tmax-tmin=2,250ms

95% input sensitivity UR22C, 200mV/Div, Trigger -400mV, min. -751mV, max. 262 mV, tmax-tmin=2,182ms

Full input sensitivity UR22C (100%), 250mV/Div, Trigger -150mV

[ozel]

sorry, I never answered to your last point. The sample rate specified in the WaveWatch sketch is misleading. The underlying java/Processing.org code does not actually apply the sample rate as parameter to the sound device, it opens the devices always with their default rate. I also found this out only after some time trying.
Most likely WaveWatch doesn't open the correct audio device (maybe only the internal sound card), so another device needs to be specified in the code or you have to mark the Steinberg device as default sound device in Windows or something like that.

Anyway, WaveWatch does not offer anything more advanced than your fancy audio oscilloscope app, rather less! For counting of pulses per time the webpage pulse recorder can be used or even better, the python scripts in this repository.
On Windows, you can also try the Theremino MCA https://github.com/ozel/DIY_particle_detector/wiki/Software#graphical-software
Even if you have not built the alpha spectrometer (where one needs an MCA-like software), Theremino still displays raw count rates of pulses if I remember well.

[ozel]

I am surprised that the browser-based Pulse Recorder actually showed 192 kHz because until recently Firefox limited the microphone sample rates to 48 kHz. It's always a good idea to try a recent Google Chrome version as well, it has by far the best and most complete javascript audio support.
Note that you need to change the red trigger line by pressing the down arrow key until you get out of the noise floor such that it triggers only on pulses.