SpecMaker - create your own Test Spectra

Home Page | Test Spectra Page | Peak Measurement Challenge

SpecMaker is a small program, based on an Excel spreadsheet, designed to create spectra with Gaussian peaks in it at known positions and with known areas so that the performance of one's spectrum analysis program can be gauged.

The program starts by creating a continuum by specifying the continuum level at the start and the end of the spectrum as specified in a text file. The program then makes either a log-linear or simple linear interpolation across the spectrum. As an alternative, an actual background spectrum can be loaded instead of the mathematically generated continuum.

The program then consults a list of peak energies and peak areas. For each item on the list, a series of counts is calculated, based on the required peak area and representing a Gaussian function distributed across the required peak width taking into account the peak width calibration. Those are then added to the appropriate channels of the continuum.

The program then imposes statistically distributed scatter across the whole spectrum, the width of the scatter depending upon the channel count. Finally, the spectrum data is output to the required filename in the ORTEC CHN spectrum format. Numerical tests have established that, at least for all normal purposes, the scatter imposed by SpecMaker is statistically reasonable.

Recently the SpecMaker program has been improved, and slight inaccuracies in the coding remedied, and validated, and is now labeled as SpecMaker 2025.

SpecMaker 2025
SpecMaker 2025 Manual
Spreadsheet for assessing SpecMaker results
Interesting Spectra

SpecMaker 2025 Plus

While updating Practical Gamma-Ray Spectrometry to its 3rd Edition, in which the correlation peak search procedure was recommended, the author became curious as to whether it could be imlpemented in Visual Basic. So, having a ready source of known peaks of any area and at any position to search for, those created by SpecMaker, it seemed convenient insert the peak search function in SpecMaker. So, SpecMaker 2025 Plus is a modification of SpecMaker 2025 in which peaks can be created in a spectrum, even very small ones, which can be used to test the correlation search function, created within the program.

The correlation search turns out to be very good and, while it can be used to check the reality of peaks created by SpecMaker, it will also find the peaks in the external background if one has been instructed to sit below those peaks. Of course, any spectrum could be loaded and searched without the need to impose other peaks. Maybe, this will be developed to become another option for routine gamma spectrum analysis. Development of SpecMaker Plus helped improve and validate the creation process of SpecMaker 2025. At the moment it is an interesting oddity waiting for development.

SpecMaker 2025 Plus
SpecMaker 2025 Plus Manual
Interesting Spectra

=============================================

The Test Spectra Measurement Challenge page carries a number of spectra that are extremely useful in gaining an understanding of how your spectrum analysis program copes with peaks that might not be easily detected. The spectrum sets are:

  • Limit of Detection - Three spectra each containing 50 peaks containing a number of counts that would be at the Limit of Detection, taking into account the level of the background beneath the peaks. The three spectra have background, respectively, 100, 1000 and 10,000 counts per channel. The statistical expectation is that your spectrum analysis program would detect, and measure, at least 47 peaks.
  • Critical Limit - Three spectra, similar to the 'Limit of Detection' spectra, in which the peak count corresponds to the Critical Limit. Statistically, one would expect your program to detect 25 of the peaks. How good is your program at detecting peaks near to the limits of detection?
  • Continuums - Four spectra each containing 50 peaks. The four spectra have different background levels and 50 peaks of the same area, giving peak to background ratios of: 10:1,1:1, 1:10 and 1:100. Of course, although the peaks are the same area, their width will be different depending upon their energy. How good is your program at measuring 50 peaks across the spectrum, identical apart from statistical scatter?
Dr. Gordon Gilmore, the creator of SpecMaker, has now retired and is not in a position to support the program. The link below will allow you to download files containing all relevant information. The creator will accept no responsibility for untoward consequences of decisions made based on the use of SpecMaker.

SpecMaker 2025
SpecMaker 2025 Manual
Spreadsheet for assessing SpecMaker results
Interesting Spectra

SpecMaker 2025 Plus
SpecMaker 2025 Plus Manual
Interesting Spectra

Appendices Data Example Spectra Links Spreadsheets Test Spectra

This website was created by Dr Gordon Gilmore for the benefit of readers of Practical Gamma-Ray Spectrometry (3rd Edition).
Problems with this page? Report it on the Feedback page.