-- Stefan E. Mueller Department of Information Services and Computing - Computational Science and Institute of Radiation Physics Helmholtz-Zentrum Dresden-Rossendorf Tel: +49 (0351) 260 3847 Stefan.Mueller_at_hzdr.de http://www.hzdr.de Vorstand: Prof. Dr. Sebastian M. Schmidt, Dr. Diana Stiller Vereinsregister: VR 1693 beim Amtsgericht Dresden On Sun, 22 Aug 2021, Ophoven, Niklas wrote: > > Dear Stefan, > > thank you very much for your detailed answer, this is really very helpful > for me! > > I tried to adopt your suggestions and finally I can now compile the NSpec.f > file without any errors and the NSpec.o file is successfully generated. > Then I want to build the new executable with: > > $FLUPRO/flutil/ldpmqmd -m fluka -o flukamy NSpec.o > > However, when I want to execute my input file with $FLUPRO/flutil/flukamy > -N0 -M3 Moderation_Tb I get the error: -bash: /opt/FLUKA//flukamy: No such > file or directory > > Indeed, in the flutil directory I cannot find an executable file named > flukamy, only the usual rfluka script. Any ideas what my mistake is? > > I have attached a new spectrum file (NSpec.txt), now with 3 columns as you > have suggested. The flux is not differential, it is indeed already in units > of n cm-2 s-1. I have converted the energy boundaries to GeV now already in > the spectrum file. > > "- After the sampling in energy, you have to pass the sampled value for > the kinetic energy to TKEFLK(NPFLKA): > TKEFLK(NPFLKA) = (ERG(I-1) + (ERG(I)-ERG(I-1))*FLRNDM(C))*1.D-03 > (watch out, FLUKA expects GeV, but your input has MeV energy bins)" > > I have written this information in line 180 of the source file, is that the > correct position? > > You are right, I will probably have to implement the emission coordinates of > the neutrons as well as the beam direction within the source file. Actually > I want to define the source as a cartesian volume source (10 cm distance in > z-direction to the target/sample) of same shape as the target, and all > neutrons shall be emitted into positive z-direction. > > There are 2 BEAMPOS cards in my input, because one defines the shape and > the other one the center of the source. Probably you are right and I will > have to define that in the source card. > > Kind regards, > > Niklas > > ____________________________________________________________________________ > Von: Stefan E. Mueller <stefan.mueller_at_hzdr.de> > Gesendet: Sonntag, 22. August 2021 01:21:07 > An: Ophoven, Niklas > Cc: fluka-discuss_at_fluka.org > Betreff: Re: [fluka-discuss]: SOURCE routine for neutron source > Dear Niklas, > > I had a look at your files, and have the following comments/suggestions: > > - in line 61 you put LFIRST = .FALSE., but in this way it will never enter > the initialization at line 72 where you read the file. Just move the > WRITE-line below line 72 and take out the first LFIRST = .FALSE.. > > - It is probably preferable to use the OAUXFI-Routine to read the file > instead of the OPEN-card in the input file (also your OPEN card is > incomplete, and it somehow shadows the SOURCE-card in your input file). So > I'd remove the OPEN-card and open the file like this: > > call OAUXFI('./NSpec.txt', LUNRD, 'OLD', IERR ) > > (where LUNRD is specified in the WHAT(1) of the SOURCE card: > > SOURCE 21. > ) > > Using OAUXFI has the advantage that the file is searched for in several > places: > 1) First of all: try to open the file in the current directory > 2) Second attempt: try to open the file in the original work directory > 3) Third attempt: try to open the file in the FLUPRO directory > 4) Last attempt: try to open the file in the user home directory > > (see > http://www.fluka.org/web_archive/earchive/new-fluka-discuss/0342.html) > > - If possible, I'd arrange the input file like this for each bin (i.e. > adding a column with the bin upper limit): > Emin Emax Value > > - You want to skip the first line, and then read the values with a > simple DO-loop (there may be more flexible ways to do this, but I usually > know how many values I have, so I hardcode the number which is 100 in > your case): > READ (LUNRD, '(A)') LINE > DO I=1, 100 > READ (LINE, *, ERR=10) EMIN, EMAX, H > ...calculate cumulated probability... > END DO > > (do you have the differential neutron flux in NSpec.txt? The units in the > header line give (cm-2s-1), so I am not sure if you need to multiply by > the bin-width when building the sum) > > - After the sampling in energy, you have to pass the sampled value for > the kinetic energy to TKEFLK(NPFLKA): > TKEFLK(NPFLKA) = (ERG(I-1) + (ERG(I)-ERG(I-1))*FLRNDM(C))*1.D-03 > (watch out, FLUKA expects GeV, but your input has MeV energy bins) > > - Then redefine the Particle momentum using the new value of TKEFLK > (NPFLKA): > PMOFLK (NPFLKA) = SQRT ( TKEFLK (NPFLKA) * ( TKEFLK (NPFLKA) > & + TWOTWO * AM (IONID) ) ) > > (otherwise it will be "PBEAM", which is derived from the BEAM-card) > > - I am not sure if the CART-VOL sampling specified by the BEAMPOS card is > reflected in the XBEAM,YBEAM,ZBEAM used in the source user routine, so you > might want to check this - if it is not, then just sample the particle > coordinates accordingly in NSource.f (I also noticed you have two > BEAMPOS-cards in your inpput file) > > - in general, you also might want to make sure that the continuation > character "&" is always in column 6 (otherwise the compiler will > complain about a "Non-numeric character in statement label" because it > expects a number there). > > I hope this helps, > > Stefan > > -- > Stefan E. Mueller > Department of Information Services and Computing - Computational Science > and Institute of Radiation Physics > Helmholtz-Zentrum Dresden-Rossendorf > Tel: +49 (0351) 260 3847 > Stefan.Mueller_at_hzdr.de > http://www.hzdr.de > > Vorstand: Prof. Dr. Sebastian M. Schmidt, Dr. Diana Stiller > Vereinsregister: VR 1693 beim Amtsgericht Dresden > > On Fri, 20 Aug 2021, Ophoven, Niklas wrote: > > > > > Dear FLUKA experts, > > > > > > so far there was no need for me to modify any of the available user > routines > > in FLUKA, but now I have to do so. Iam using FLUKA with the Ubuntu 18.04 > LTS > > software. > > > > In principle my simulation is quite simple: > > > > I want to simulate a rectangular shaped neutron source (homogenously > > distributed in a cartesian volume source with emission direction into > > z-axis) that hits a rectangular shaped target of same dimensions made of > > terbium(III) chloride hexahydrate. > > > > > > So far I have managed to copy the SOURCE.f file from the FLUKA directory, > > but the compilation $FLUPRO/flutil/fff NSource.f (source file is attached) > > still yields some errors. I spent some time now on solving the errors with > > material from FLUKA courses and old, similar questions in this mailing > list. > > However, as Iam not an expert in Fortran, I would appreciate any kind of > > help or guidance you could give me there. > > > > > > I want to load in the spectrum data in a histogram form, i.e. energy bins > > and relative emission probabilities. The left column in the file shows the > > energy bin boundaries, the right column gives the flux values. Note, that > > the energy boundaries on the left are lower and upper boundaries, so that > > the left column is bigger by one line compared to the right column. When I > > would simply divide the flux in each bin over the total (integral) flux, > > would this be sufficient for FLUKA to reconstruct the shape of the > > spectrum? > > > > > > I will load the spectrum data with an OPEN card in the input file (see > > attachment). The data got the logical unit number 21, is it sufficient to > > put "21" before the file ending for FLUKA to recognize that file > > adequately? > > > > > > Any kind of comments and guidance is of great help for me, thanks in > > advance. > > > > > > Kind regards, > > > > > > Niklas > > > > > > > >--------------------------------------------------------------------------- > > > --------------------- > >--------------------------------------------------------------------------- > > > --------------------- > > Forschungszentrum Juelich GmbH > > 52425 Juelich > > Sitz der Gesellschaft: Juelich > > Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498 > > Vorsitzender des Aufsichtsrats: MinDir Volker Rieke > > Geschaeftsfuehrung: Prof. Dr.-Ing. Wolfgang Marquardt (Vorsitzender), > > Karsten Beneke (stellv. Vorsitzender), Dr. Astrid Lambrecht, > > Prof. Dr. Frauke Melchior > >--------------------------------------------------------------------------- > > > --------------------- > >--------------------------------------------------------------------------- > > > --------------------- > > > > > > > >
This archive was generated by hypermail 2.3.0 : Mon Aug 23 2021 - 00:41:15 CEST