Impulse Executable for Windows

[Sesselmann]

Victory at last after a long battle with Windows....

Anyone game to test this app and provide some feedback?

I have been developing this app on a Mac, so it has been failrly easy to release regular updates in .app format, but Windows threw a bunch off errors every time and I had to investigate each individual error and search for solutions, but I now have a version that has been tested on my PC.

This version works with all models of the Gammaspectacular as well as the Atom-Nano and GS-MAX products.

https://www.gammaspectacular.com/blue/s ... ds/impulse

Steven


Intuitive layout - No hidden settings

tab2

Collaborate with other users, share your spectra, this feature is unique to Impulse.

tab5

[Sesselmann]

As more people are trying Impulse and giving me feedback, I have been fixing bugs and making small improvements.

The latest versions for both Mac and PC are up to date as of yesterday.

https://www.gammaspectacular.com/blue/s ... ds/impulse

Steven

[Sesselmann]

In preparation for the release of the new GS-MAX I have been doing further upgrades to Impulse.

Not too many changes for those using the Gammaspectacular, but if you are already on the GS-MAX-14 or Atom-Nano devices then the latest version will offer more control.

Dowload
https://www.gammaspectacular.com/blue/s ... ds/impulse

Steven

Spectrum page

Serial device command input

Th228

newplot-20.png (89.47 KiB) Viewed 452 times

[jbeale]

I was able to run the program and get what may be a plausible spectrum, with my first test of a surplus 3" R6233-02 PMT and 40mm x 40mm NaI scintillator, using a Theremino "PMTAudioAdaptor - New Version" circuit. This PMT is spec'd for 1kV bias voltage which is was I used initially, but then I realized looking at the raw signal in "Audacity" that almost all my pulses were clipping the max range, and I had to turn down the DC bias voltage a lot, to get lower net gain. I'm not sure if my situation is common, but you might include an illustration of what the distortion curve is supposed to look like, when most of the input peaks are not clipping.

Cs-137 spec

Cs137-spec.png (13.9 KiB) Viewed 414 times

PMT + scintillator and reflector ring

[Sesselmann]

John,

Your approach to using Audacity to set the gain is good 👍, the sound card has a dynamic range of around +- 1.2 Volts so if your pulses are clipping at the top, then your gain is too high.

Some huge pulses from cosmic rays and other noise will always hit the ceiling but 99% should be within the range.

If you are using Impulse with a soundcard spectrometer like the GS-PRO I recommend using 1000 channels and setting the bin size to 30, you then adjust your gain so that the main peak from 137-Cs (662 kev) lands around channel 230 (gain is still too high in your plot above).

Pulse shape acquisition
First you run the mean shape acquisition, this function collects a given number of pulses and saves a csv file containing the mean pulse shape for your detector setup.

Distortion curve
After running the pulse shape you can run the distortion curve function, this function compares a given number of pulses to the mean and sorts it by deviation.
Typically you will see a hockey stick curve, where the distortion is relatively low for most of the range but then rises rapidly at the end. This tells us that the pulse shape is relatively conssistent over most of the the energy range except for the very large pulses. Now you can read the value on the "y" axis and determine where to set your "Shape tolerance". Setting the tolerance too low could result in a spectrum where to top end (right side) is missing.

Hope that helps, and please let us see some nice looking spectra 👍

Steven

[Jim Kovalchick]

Steven,
I am intrigued by Impulse being able to read a serial data stream from the ATOM. Would it be possible to emulate the ATOM's protocol with a serial data stream from a microcontroller board like Arduino, Teensy, or Pi Pico? I have been tinkering for a while now with pulling data from Canberra ADC's using microcontrollers acting as the MCA. I'm relatively happy with the spectra I get, but there would be some additional advantages if I could just port the data straight over to Impulse from the microcontroller. If the ATOM's communication protocol is available, I would like to give it a try.
Thanks,
Jim Kovalchick

[Sesselmann]

Jim,

Sure, Impulse is open source Python, so you can create a fork and try here: https://github.com/ssesselmann/impulse

Inside the folder called shproto you will find the files relating to the serial protocol, I did not write this myself, I cloned it from Amber here and modified it: https://github.com/Am6er/nanopro

It would be great if we can add more device options to Impulse.

Steven

[Jim Kovalchick]

Steven,
Thanks for the answer. I'll have to admit that it was all gibberish to me. My question was really much more basic than all that. I'll try again. Does the serial communication from the Atom contain digitized pulses or does it contain the output of multichannel analysis? In other words, is it sending a digital representation of analog pulses or does it send energy/channel information? If it is the latter then maybe I have a chance. Also, does it send data one pulse/gamma at a time or does it send a spectrum's worth of data all at once? If it is the latter, I won't bother because I already have the ability to dump an accumulated spectrum all at once for plotting.
Please forgive my toddler level questions, but I am a sixty+ year old who once coded in machine code, basic, and fortran forty years ago, but I have done almost nothing with contemporary coding. I have started learning C++ to tinker with microcontrollers and will probably play a little with micropython. I am willing to go after python, but until I do that I am not set up well to go into the files you pointed me to and discern how the Atom communicates. thanks.

[Sesselmann]

Jim,

Sorry I missed your question. The ATOM serial devices and GS-MAX devises are MCA's and typically send a complete 8192 channel spectrum to the PC once per second.

It is possible to change the mode so it sends a pulse shape, but not in the same way as the Gammaspectacular.

Steven