Page 2 of 2

Re: Writing source files

Posted: Fri Oct 12, 2018 2:45 am
by keckler
Hello,

I am currently trying to print some source distributions using the det 'df' option. For the most part it seems to be working, but sometimes when running I am getting warnings printed out that say:
Warning message from function WriteSourceFile:

P larger than 1 (1.356349E+00)
This sounds to me like it is related to the neutron weights, but I am running in analog criticality mode, so I would think all the weights are 1.0. I am running with a few different detectors to print out multiple source distributions:
det batch1 n du 0101 du 0201 du 0301 du 0401 du 0501 du 0601 du 0701 du 0801 du 0901 df sourcefile01 0.5
det batch2 n du 0102 du 0202 du 0302 du 0402 du 0502 du 0602 du 0702 du 0802 du 0902 df sourcefile02 0.5
det batch3 n du 0103 du 0203 du 0303 du 0403 du 0503 du 0603 du 0703 du 0803 du 0903 df sourcefile03 0.5
det batch4 n du 0104 du 0204 du 0304 du 0404 du 0504 du 0604 du 0704 du 0804 du 0904 df sourcefile04 0.5
det batch5 n du 0105 du 0205 du 0305 du 0405 du 0505 du 0605 du 0705 du 0805 du 0905 df sourcefile05 0.5
I checked that the warnings are still printed when the fraction parameter at the end of the detector definitions are increased from 0.5 to 1.0, and they are.

The stranger part about this is I don't always get the printed warnings. I can't seem to figure out a pattern, but sometimes I do not see the warnings printed in my output.

Any thoughts on what this means?

Thanks,

Chris

Re: Writing source files

Posted: Fri Oct 12, 2018 1:15 pm
by Ville Valtavirta
If you decrease the fraction parameter you will get less of those warnings.

The probability to store a source neutron depends on the neutron energy and the local material cross section and can sometimes be larger than unity, which requires the user to decrease the fraction parameter.

For further information on the way the storing probability is calculated, see http://serpent.vtt.fi/mediawiki/index.p ... ound_on_PN.

The PN parameter in the "set savesrc" card has the same function as the fraction parameter in the "df" option for detectors.

-Ville

Re: Writing source files

Posted: Fri Oct 12, 2018 6:17 pm
by keckler
Thanks! Looks good now.

Re: Writing source files

Posted: Thu Mar 26, 2020 12:42 pm
by novako
Hi,
I did received large number of warning messages from function WriteSourceFile: P larger than 1, so I decreased the parameter in savesrc command to
set savesrc "./source" [0.12 1 1 1 1] and adjusted the set cfe 10. It decreased the occurrence a lot, but still time to time I receive this error. In my last calculation, I got 20 in simulation of total 5e9 neutrons. Should I totally eliminate them or few is ok?
Thank you
Ondrej

Re: Writing source files

Posted: Mon Jun 08, 2020 10:37 am
by Ville Valtavirta
I think a few should be OK,

that will just mean that a couple of particles should have been included in the source multiple times, but I don't think you'll see an effect on the transient simulation.

-Ville