Re: Low energy - High energy neutron spectrum mismatch with USRTRACK

From: Roy Keyes <>
Date: Thu, 26 Feb 2009 11:13:38 -0700

Thank you Anna and Alberto,

It looks like my problem was that I was multiplying the values that
Fluka gave me by the bin width (as described in the USRTRACK section of
the manual...). I was in fact reading the low energy values in the
correct order already.

At the high/low boundary, the values, when multiplied by their bin
energy widths are:

Low energy: 1.1050E-04 * 3.596700E-04 = 3.9743535e-08

High energy: 9.3207E-05 * 4.0000E-03 = 3.728280e-07

which gave me the order of magnitude mismatch.

I've attached a semilog plot, which has the data plotted as Alberto
suggested, and it looks much better. Now I can address my other issues :)

Thanks again,

Anna Ferrari wrote:
> Hello Roy,
>> I suspect I am making an obvious mistake somewhere.
> you are right. There is not a discontinuity in your high energy-low
> energy neutron spectra, simply you did not correctly read
> the output file: you have read the low-energy part in the inverse mode.
> The results of the high energy part are written starting with the lower
> energy bin (20 MeV, 24 MeV); the binning of the low energy part of
> the spectrum starts with the GROUP 1, that corresponds to the
> higher energy interval (19.6 MeV, 20 MeV).
> Your contents are: 1.105 10^-4 in the bin (19.6,20)
> 9.32 10^-5 in the bin (20,24)
> and you will find a continous behaviour.
> Just two suggestions more:
> 1) in your .inp file you have set the USRTRACK card with: linear
> binning in (0, 0.2) GeV, 50 bins. For neutrons FLUKA will force
> the binning to fit with the 260 neutron groups (logaritmically
> binned) between the thermal region and 20 MeV; above 20 MeV
> you get 45 linear energy bins. If you want to better control
> your linear binning in the high energy part, you can use
> the USRTRACK in two steps, for the two energy ranges.
> 2) you can also use a non-formatted output (using a negative number in
> what(3)) and then run the post-processing program ustsuw (see the manual
> for the details): in this way you will get - between other outputs -
> a file [output]_tab.lis, which shows in the first two columns the energy
> boundaries, in the third the corresponding bin content and in the last
> one the statistical error.
> cheers,
> Anna
>> Hello Fluka users,
>> I am trying to measure the neutron fluence energy spectrum of
>> annihilation neutrons produced by antiprotons incident on a copper
>> target. I am using USRTRACK to score the neutrons, but when I calculate
>> and plot the "low" and "high" energy parts of the spectrum together,
>> there is a large mismatch in the spectrum at 20 MeV (see attached plot).
>> I am using the 260 group low energy neutron transport library and am
>> scoring up to 0.2 GeV. To obtain the fluence energy spectrum, I am
>> multiplying each bin by its energy width in GeV.
>> I am not sure if my 0.0564 cm thick target qualifies as "very thin" and
>> would lead to the artifacts mentioned in section 10.1.1 of the Fluka
>> manual.
>> If anyone has any idea why I am seeing such a disjointed spectrum, I
>> would greatly appreciate hearing back. I suspect I am making an obvious
>> mistake somewhere.
>> Thanks,
>> Roy

Received on Fri Feb 27 2009 - 12:59:18 CET

This archive was generated by hypermail 2.2.0 : Fri Feb 27 2009 - 12:59:21 CET