ImpulseQt v3.1.6

[Sesselmann]

On Friday I published ImpulseQt v3.1.5 and this morning I woke up with a better idea and spent all Saturday coding (as you do) and 3.1.6 was born.

These are the changes...

Things Done in v3.1.6:
===================================================================
• Removed isotope checkbox and redundant code ✅
• Removed calibration window ✅
• Integrated calibration into peaks table ✅
• New Isotope library updater ✅

Things Done in v3.1.5:
===================================================================
• Add smooth count rate number to count rate histogram ✅
• Add Log function to count rate histogram ✅
• Add data table - Centroid - Counts - Resolution - Energy- gross counts - Net Counts ✅
• Highlight Regions of interest - ROI ✅
• Add Pb210 to library ✅
• Remove all histogram annotations ✅

Notes about v3.1.6
The most time consuming part was the peaks table. I got rid of the histogram annotations, they tended to overlap each other and were difficult to read. These along with the "peakfinder" were replaced by a region of interest function and a new peaks table.

Calibration Procedure

• Record a spectrum that with known photopeaks.
  Click Pop-out table for a larger view of the histogram and peaks table.
  Double-click on each known peak in the histogram to create a yellow ROI and add a new row to the table.
  In the table, double-click the Ref E (keV) cell and enter the known energy for that peak.
  Repeat for as many calibration points as you like (more points generally improve precision).
  Enable Calibration in the main window.
  The program will compute energies using a second-order (quadratic) polynomial fit.

Notes
With two points the fit is linear; with three or more it becomes quadratic.
When Calibration is on, the program checks the selected isotope library and suggests nearby gamma/X-ray lines.
The polynomial coefficients are saved with the spectrum in the JSON file; ROIs are not currently saved.
Clear all ROIs using the Clear Peaks button.

Download
from Github: https://github.com/ssesselmann/ImpulseQt/releases
from Microsoft Store (signed)*: https://apps.microsoft.com/detail/9MXB6 ... d=pdpshare

[Rob Tayloe]

I downloaded ImpulseQt v3-1-6 yesterday and spent a little while using it. I first downloaded the Windows, mac, and source versions on a Win10 machine that had just been updated with the latest Win security patches. Happily, there were no warnings from the anti-virus software.

I then used a "virgin" (i.e., had not previously had any prior versions of gamma spec software installed) Win11 machine which had all recent upgrades installed before downloading ImpulseQt v3.1.6. No anti-virus warnings and the extracted exe file ran without needing any additional drivers. I was using the GS-Max-8000 MCA with a 38x38 mm CsI detector which used 900 volts in a single wire configuration. I used a Cs-137 source beneath the detector to do the initial configuration.

I set the HV to 902 volts (-U060) and set the pulse view to obtain the full pulse (-fall030). A screen capture of the pulse is shown below.

The initial spectrum acquired with the Cs-137 source had the 662 keV peak way too far to the left (with 2048 bins). Since varying the max setting was said to serve as a de facto gain control, I varied this setting so that the 662 keV was located just a bit less than the mid-range of the bin space. This ended up being a -max2500 setting. A screen capture of this spectrum is shown. I also made several other short term measurements of other sources to determine if the spectrum looked "reasonable". Those screen captures are also shown.

I still find it necessary to STOP and then re-START the data acquisition following a source change and filename revision after each measurement to fully clear the bin buffer.

[Sesselmann]

Rob,

Thanks for feedback...

Great to hear that the warnings have gone, I assume it is because Impulse Qt is now registered on the MS Store (even though 3.1.6 is not officially released on the store as I am writing this).

I still think you are running your detector at way too high voltage, very few PMT's require 900V !!

My recommendation is to drop the voltage right down 580 - 650 typical, no real need to mess around with -max unless absolutely necessary -max 13,800 default works for most people.

The correct position for your Cs137 peak is at (channels/3000)*662 i.e. if you are running (2048 channels/3000 channels) * 662 = peak in channel ~450

• Did you figure out how the new calibration works ?

• Did you try the pop out table ?

Clearing Histogram
There is no need to Stop and Start twice.

The last histogram recorded will persists on the screen (even after restert), until a new filename is entered and START is Clicked.

Upon clicking START the function checks:

Does the filename already exist?

IF YES, then pop up a warning ; "FILENAME.json already exists... Overwite ? YES/NO ; ELSE proceed to clear all variables and record new filename

If YES: Clear all variables and START - screen will clear and a new histogram will overwrite the old filename

If NO: Go back to change filename



Steven

[Rob Tayloe]

Steven,

I have continued experimenting with ImpulseQt v3-1-6 and offer the following comments and observations.

With respect to the HV setting. For the CsI detector, my determination of the HV setting was done initially using the Spectrum Techniques UCS 30 MCA and the Spectrum Techniques USX software configuration methodology. This method requires use of a Cs-137 source with the scintillator. The HV is gradually increased until pulses are detected, then the coarse and finally fine gain settings are increased so that the 662 keV peak is about centered in the channel space. The chosen voltage was 900 volts and that has been the basis for the use for that detector.

Never-the-less, based on your recommendation, I decided to use a lower HV setting. I started with a GS-USB-PRO device (s/n 1451) because I could use the signal port with an oscilloscope and observe the pulses as the voltage was lowered. With the same CsI detetector as used above, I settled on a HV of 669 volts. I opted to use the PRA software with the GS-USB-PRO and CsI detector to determine if reasonable spectra could be obtained. I altered settings on the PRA data acquisition and analysis page as shown in the screen capture below. Specifically, I set the min height/arb unit to 0.5, the bin size/ arb unit was set to 0.02, and the boost gain was set to 10. These settings were chosen so that the Cs-137 662 keV peak was shifted a bit to the right. Shown below are screen captures from spectral measurements for Cs-137 and Co-60.

Next, using the same detector configuration, I hooked up the GS-MAX-8000 and used ImpulseQt v3-1-6. Configuration was done using a Cs-137 source. The -max setting was at 18300. The HV was set to 668 volts (-U032). The -fall setting was 030 and the settings and pulse view is shown in the screen capture below. I attempted to acquire spectra with the Cs-137 and Co-60 sources as was done with PRA and the GS-USB-PRO. The count rate was very low and the spectra acquired after a few minutes were not useful. Screen captures are below.

I noticed that the notes are not indicated as being written to the desired filename. Perhaps this is a bug.

[Sesselmann]

Rob,

I am scratching my head... 🤔

Why would you set the boost gain in PRA to 10 ?
Marek added the "Boost gain" function as an exeptional gain control for experiments, sometimes required for proportional detectors that don't have the same level of gain as a PMT. The signal from a PMT does not require much amplification to be read by a sound card, it only needs a bit of stretching and shaping, which is done with an opamp and a low pass filter. Boost gain is normally set to 1 in PRA when you are using PMT type detectors.

I don't know what's going on with your GS-MAX, maybe it is better to set it up with a NaI probe first -- for NaI you may need to change the -ris and -fall.

Once you see the whole spectrum, it will be easier to fine tune.

Steven

Typical NaI settings...

Probe Voltage ~ 550 to 700V

GS-MAX-8000 Settings for NaI Probe

Histogram