From: Alberto Fasso' (fasso@SLAC.Stanford.EDU)
Date: Tue Dec 12 2006 - 00:42:04 CET
The reason is probably in your test
ETRACK.EQ.PBEAM
both ETRACK and PBEAM are real (actually double precision) quantities, and
it is well known that comparisons of such quantities should be done only
by .LE. .LT. .GE. or .GT. Equality can be spoiled by rounding errors.
Suppose for instance that PBEAM has been input as 21.5, but ETRACK
for some reason is 24.99999999999999D0. For computing purposes it is the same
thing, but the test of equality would fail. I suggest that you change your test
into:
IF (ICODE.EQ.101 .AND. ETRACK.GE.PBEAM*ONEMNS) THEN
(ONEMNS is predefined in FLUKA as 0.999999999999999D+00)
Alberto
On Mon, 11 Dec 2006, me@marychin.org wrote:
> Dear FLUKA friends,
>
> I started monoenergetic photons in homogeneous lead. Simulations were
> repeated for different incident energy, ranging from 7.5 to 29.5 MeV. My
> USDRAW in mgdraw.f has:
> IF (ICODE.EQ.101 .AND. ETRACK.EQ.PBEAM) THEN
> J = 0
> DO I = 1, Np
> IF (Kpart(I)==8) THEN
> J = J + 1
> END IF
> END DO
> WRITE (IODRAW) J
> END IF
> so that only photons at incident energy undergoing inelastic
> interactions scores. Slowing down photons were not allowed to score. I
> basically mean to count the number of (g,n), (g,2n) and (g,3n) events.
> When the count is plotted against the incident energy, I expect
> the plot to resemble the photoneutron cross section. Generally results
> appear as expected except that at incident energies 21.5, 25.5 and 29.5
> MeV, I miraculously get zero counts. Zero counts were obtained at
> exactly the same three energies when the simulation was repeated with
> carbon and calcium. I can't understand the abrupt zero and the
> discontinuity. Could someone please help. Have I missed something?
>
> My results for lead is as follows. Please note the region of interest I
> have zoomed into between 21 and 22 MeV. 21.4, 21.49 and 21.6 all
> produced non-zero counts while 21.499 and 21.5 produced zero.
> MeV (g,0n) (g,n) (g,2n) (g,3n)
> 7.5 709 1747 0 0
> 8.5 0 11923 0 0
> 9.5 0 37225 0 0
> 10.5 0 85953 0 0
> 11.5 0 181338 0 0
> 12.5 0 266842 0 0
> 13.5 1 328983 0 0
> 14.5 0 242696 17314 0
> 15.5 1 90913 76994 0
> 16.5 0 20916 80017 0
> 17.5 2 5257 65996 0
> 18.5 4 1634 52304 0
> 19.5 8 572 35742 0
> 20.5 9 235 21941 0
> 21 13 188 18131 0
> 21.1 10 190 17868 0
> 21.2 16 148 17409 0
> 21.3 10 169 17280 0
> 21.4 10 155 17156 0
> 21.49 12 157 17694 0
> 21.499 0 0 0 0
> 21.5 0 0 0 0
> 21.6 12 175 17529 0
> 21.7 11 158 17700 0
> 21.8 18 162 18257 1
> 21.9 13 154 18703 9
> 22 18 152 19278 18
> 22.5 20 161 22008 210
> 23.5 40 168 22886 2608
> 24.5 32 118 12426 5929
> 25.5 0 0 0 0
> 26.5 20 75 2865 9170
> 27.5 19 94 1715 10760
> 28.5 32 85 951 9397
> 29.5 0 0 0 0
>
> Attached inp file, as well as photoneutron cross sections for lead. The
> cross section doesn't seem to explain the trend. Calcium, carbon and
> lead are unlikely to have absolute-zero valleys at exactly the same 3
> energies anyway?
>
> Thanks very much.
>
> mary
>
>
>
-- Alberto Fassò SLAC-RP, MS 48, 2575 Sand Hill Road, Menlo Park CA 94025 Phone: (1 650) 926 4762 Fax: (1 650) 926 3569 fasso@slac.stanford.edu
This archive was generated by hypermail 2.1.6 : Tue Dec 12 2006 - 09:48:00 CET