Problem with gamma source intensity distribution

I am trying to produce gamma lines of equal intensity, I tried:
momentum_energy = sample_discrete_momentum_energy( [filename], [unit] ), and used a dat file for energy distribution of equal intensities.
When I plot the flux using USRTRACK in the linear mode, it gives equal intensities which is fine, while in the log mode, it gives decay lines as in the pic. Is that normal? although the lines’ intensities are the same.

GAMMA2.flair (2.3 KB)

gamma_source_lines.dat (131 Bytes)
source_newgen_gamma.f (20.8 KB)

Dear Soulaiman,

The unit you get from a fluence (USRTRACK or USRBDX) scoring is [1/GeV/pr]. Since the width of the bins of a logarithmic scoring is not uniform, the values are different.

However, if you calculate the integral value of the bins, you will get the same result, showing that the sampling is indeed uniform

Cheers,
David

1 Like