From: paola sala (paola.sala@cern.ch)
Date: Tue Apr 17 2007 - 17:40:47 CEST
Dear Jo
I finally arrived to the conclusion that the problem is only a plotting
artifact. ( I should have realized it earlier)
You are plotting things VS channel number. But your channel width has an
abrupt discontinuity at 20 MeV, where you suddenly change from a tin log
spacing to a fixed linear spacing. Therefore, the change of slope that
does exist in the low part of the spectrum is artificially shown as a
kink. See the attached plo, where I plot in channel mode and in energy
scale. ( I used here a larger angular acceptance, and I had to correct
some of the energy group limits in your post-processing code , see
groups.dat file)
-- by the way, lethargy is a very nice thing to plot, but it goes with
logarithmic energy scale : in this way the size of the area below your
spectrum is directly proportional to the integrated fluence between two
given energies, and one gets visually the relative importance of
different energy components.
-- in the "correct" lethargy spectrum, one sees as usual a low energy
component (few MeV), mainly due to "evaporation" processes, and a
"cascade" component at 100-few hundreds MeV. At higher energies, the
spectrum decreases, and the (now too small) energy bins are scarcely
filled.
===> it would be much better to use a log E scale, where bins become
wider at high energies
===> it would be better to use the built-in scoring, that can give you
authomatically the spectrum, without post-processing, and with tested
algorithms.
To this aim, I attach a modified inp file, where I introduce additional
"scoring regions" around the target and at the detector position. A
usryield and a usrbdx example are also in the inp.
Paola
On Thu, 2007-04-05 at 09:20 -0700, Joseph Comfort wrote:
> Hi Paola,
>
> Yes, plotting things this way helps to see the problem, so I'm glad I
> did it. Attached are the relevant files. The production target is a
> set of large wheels, with the beam hitting near one edge. I use a
> source.f to give unequal X and Y divergences to the beam. (It's too bad
> that the Fluka input does not have this feature.) I use mgdraw_targ.f
> to select particles in a narrow angular range, over 2pi phi angle. (For
> historical reasons, I convert to Geant3 particle ID's.) Then I use the
> program blowup.f to generate the data for plotting.
>
> The runs take a while. The data in the figures are for 160 million
> protons, for which I used 40 cpu processors over several days. The runs
> are collected in 40 directories (different random seeds), and 20 runs
> per directory. You will not need all of this. A single run is probably
> sufficient.
>
> There may be things in the input file and codings that are less than
> optimal, but I don't think any that might exist would provide a bias in
> the 20-400 MeV range. I don't know how much of an effect the 'extra'
> neutrons will have in going through a beamline collimator, or in the
> detector, but I would rather have something that I can parameterize
> better as a function for source.f.
>
> Thank you for looking at the issue.
>
> Best regards,
> Joe
> plain text document attachment (ni30.inp)
> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
> TITLE
> Particle production from 30-GeV protons on a Ni target
> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
> GLOBAL 1.0
> DEFAULTS PRECISION
> SOURCE
> BEAM 40.0 PROTON
> *BEAM -30.0 -0.5 -3.88 -0.306 -0.306 PROTON
> *EMF OFF
> LOW-MAT 25.0 28. -2. 293. NICKEL
> *
> GEOBEGIN COMBNAME
> 0 0 Step1 Production Target
> SPH BlakHole 0.0 0.0 0.0 3000.0
> SPH World 0.0 0.0 0.0 700.0
> SPH Shell 0.0 0.0 0.0 660.0
> SPH Vacuum 0.0 0.0 0.0 650.0
> RCC Disk1 13.0 0.0 0.00 0.0 0.0 2.17 28.0
> RCC Gap1 13.0 0.0 2.17 0.0 0.0 0.30 28.0
> RCC Disk2 13.0 0.0 2.47 0.0 0.0 1.12 28.0
> RCC Gap2 13.0 0.0 3.59 0.0 0.0 0.30 28.0
> RCC Disk3 13.0 0.0 3.89 0.0 0.0 0.83 28.0
> RCC Gap3 13.0 0.0 4.72 0.0 0.0 0.30 28.0
> RCC Disk4 13.0 0.0 5.02 0.0 0.0 0.69 28.0
> RCC Gap4 13.0 0.0 5.71 0.0 0.0 0.30 28.0
> RCC Disk5 13.0 0.0 6.01 0.0 0.0 0.59 28.0
> END
> R1bh 2 +BlakHole -World
> R2World 2 +World -Shell
> R3Detect 3 +Shell -Vacuum
> R4VacReg 6 +Vacuum -Disk1 -Disk2 -Disk3 -Disk4 -Disk5
> -Gap1 -Gap2 -Gap3 -Gap4
> R5Disk1 3 +Disk1
> R6Disk2 3 +Disk2
> R7Disk3 3 +Disk3
> R8Disk4 3 +Disk4
> R9Disk5 3 +Disk5
> R10Gap1 3 +Gap1
> R11Gap2 3 +Gap2
> R12Gap3 3 +Gap3
> R13Gap4 3 +Gap4
> END
> GEOEND
> *
> MATERIAL 0.0 0.0 0.001205 26.0 0.0 0. AIR
> COMPOUND -0.755267 7.0 -0.231781 8.0 -0.012827 20. AIR
> *
> ASSIGNMAT 1.0 1.0
> ASSIGNMAT 2.0 2.0 4.0
> ASSIGNMAT 25.0 5.0 9.0
> ASSIGNMAT 2.0 10.0 13.0
> *
> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
> * e+e- and gamma production threshold set at 10 MeV
> *EMFCUT -0.010 0.010 1.0 12.0 PROD-CUT
> * Boundary crossing current in the middle of the target (linear bin, one-way)
> USRBDX 1.0 201.0 47.0 4.0 3.0 1. CurrUD
> USRBDX +15.0 +15.0 0.0 &
> * Call MGDRAW
> USERDUMP 101.0 2. prod
> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
> RANDOMIZE 1.0 1.
> *...+....1....+....2....+....3....+....4....+....5....+....6....+....7....+....8
> START 200000.0
> STOP
data groups
& /1.9600d-02, 1.7500d-02, 1.4918d-02, 1.3499d-02, 1.2214d-02
& ,1.1052d-02, 1.0000d-02, 9.0484d-03, 8.1873d-03, 7.4082d-03
& ,6.7032d-03, 6.0653d-03, 5.4881d-03, 4.9659d-03, 4.4933d-03
& ,4.0657d-03, 3.6788d-03, 3.3287d-03, 3.0119d-03, 2.7253d-03
& ,2.4660d-03, 2.2313d-03, 2.0190d-03, 1.8268d-03, 1.6530d-03
& ,1.4957d-03, 1.3534d-03, 1.2246d-03, 1.1080d-03, 1.0026d-03
& ,9.0718d-04, 8.2085d-04, 7.4274d-04, 6.0810d-04, 4.9787d-04
& ,4.0672d-04, 3.3373d-04, 2.7324d-04, 2.2371d-04, 1.8316d-04
& ,1.4996d-04, 1.2277d-04, 8.6517d-05, 5.2475d-05, 3.1828d-05
& ,2.1852d-05, 1.5034d-05, 1.0332d-05, 7.1018d-06, 4.8809d-06
& ,3.3546d-06, 2.3054d-06, 1.5846d-06, 1.0446d-06, 6.8871d-07
& ,4.5400d-07, 2.7537d-07, 1.6702d-07, 1.1030d-07, 6.1442d-08
& ,3.7267d-08, 2.2603d-08, 1.5535d-08, 1.0677d-08, 7.3375d-09
& ,5.0435d-09, 3.4662d-09, 2.3824d-09, 1.6374d-09, 1.1254d-09
& ,6.8275d-10, 4.1400d-10/
This archive was generated by hypermail 2.1.6 : Tue Apr 17 2007 - 22:20:25 CEST