Dear colleagues,
we are carrying out a very simple FLUKA simulation to determine the (pion decay) photon spectrum we can expect from energetic ions impacting a thick target source. A beam of protons is directed at a cuboid composed of a substance which is mostly hydrogen, with small quantities of helium, carbon, oxygen and nitrogen (first approximation to the chemical composition of the atmosphere of the Sun). The simulations run fine and produce understandable results for moderate proton energies, up to a few GeV, but if we go to 10 GeV we start to find a rare event that causes a crash with the error message
Abort called from STCPPR reason INVALID WHYDR Run stopped!
STOP INVALID WHYDR
If we change the composition of the target so that it is only hydrogen, or helium, or carbon, or even a mixture of one or two of these, there is no error. So I guess this is something to do with transport in our medium. We have experimented with explicitly setting thresholds for different species but we have not managed to eliminate the error. I wondered if this was something to do with the transport of spallation fragments so we compiled and ran with DPMJET but this doesn't solve the problem.
Because this is a rare event we can work around it by doing several runs with modest numbers of primaries. Not all of them will crash and we can use results from the successful ones in the usual way. But this is slow and fiddly and we would like to understand what caues the error in case it highlights something we should be aware of in our results.
Example .inp, .out and .err files are attached. Thanks in advance for help.
Alec MacKinnon
Dr Alexander L MacKinnon
Senior Lecturer
Centre for Open Studies/School of Physics and Astronomy
University of Glasgow
GLASGOW G12 8QQ
UK
(+44)(0)141 330 1857
http://www.gla.ac.uk/study/short/staff/alexandermackinnon/
Please note my normal hours of working: Monday, Tuesday and Wednesday a.m. in Open Studies; Wed p.m., Thursday and Friday in Physics and Astronomy
__________________________________________________________________________
You can manage unsubscription from this mailing list at
https://www.fluka.org/fluka.php?id=acc_info
Received on Wed Jul 13 2016 - 21:11:20 CEST