Please describe the issue. In case of reporting a bug, please provide step-by-step instructions for reproducing the error.
Dear Fluka experts,
I am working on the efficiency calibration of an HPGe detector using a cylindrical source and monoenergetic beam, for now. I am using the DETECT and DETGEB cards to obtain the pulse-height spectrum.
However I am facing two issues:
Low Efficiency: My calculated efficiency from the .tab.list file is very low (around 10^-6 %) for low-energy photons. There might be a physics or geometry issue that I’m missing.
Normalization and Units: I would like to confirm the unit of the values of the 3rd column of the .tab.list output. Is it a density expressed in GeV^-1.primary^-1? If so, to calculate the efficiency, I would have to multiply the area under the peak by the bin width expressed in GeV?
I have attached my files for reference. Any insights on the simulation or the correct processing steps would be greatly appreciated.
Thank you in advance for your help.
Input files
Please upload all relevant files. (FLUKA input file, Flair project file, user routines, and data files)
Depending where you look in a .lis file, you will see 4 columns. The first two are the bin limits, the third the distribution value (indeed normalized in GeV$^{-1}$ per primary (or primary weight)) and the percent error. You will not find the efficiency in this file even using the coincidence feature of the DETECT card.
Where in your lis file were you looking? Could you perhaps attach it here?
To calculate the efficiency, I multiplied the values of the 3rd column of the .lis file by the bin width (the difference between the 2nd and 1st coloumns). With this method, my results are very low: 10$^{-6}$ %. And even with a flat beam directed to the detector, the efficiency stays very low.
If you set a lower energy limit of zero in your DETECT card, you will notice that the integral of the third column is equal to 1. The high value of the first bin will be due to the contained zeros, corresponding to photons (most of them) that did not deposit any energy in the crystal. If you subtract this value from 1, then you already get the total efficiency.
Now, since the integral is 1, the values of a bin should be understood as “counts in the channel per emitted photon”, i.e. exactly what is in your experimental spectrum, except there it is “counts in the channel per all photons emitted during acquisition”, and indeed you divide that by the number of photons your calibrated source emitted during acquisition to obtain the full energy peak efficiency. So there is in fact no need to multiply the third column by the bin width, which is the reason you produce very low values.
In fact, if you do not apply any GEB, the value of the bin corresponding to your source photon energy (minus any substantial background in adjacent bins to be subtracted), will directly give you the peak efficiency for that photon energy. (Unless the energy falls exactly on the bin limit, in which case two adjacent bins may need to be added.)
The application of GEB is then only really relevant if you want to compare to an experimental spectrum, especially where adjacent lines are not distinguishable due to detector resolution. For the tests you are running with monoenergetic photons, you make your post-processing much easier without it.
You can of course put two copies of the same DETECT card in your input (differently named), and apply GEB to only one of them and compare the two results.