| Problem with TGeant4? [message #19347] |
Mon, 27 June 2016 17:05  |
C. A. Douma
Messages: 88
Registered: September 2015
Location: Groningen
|
continuous participant |
From: *kvi-cart.rug.nl
|
|
I run a simulation where I have only NeuLAND in the setup. I shoot 1 GeV neutrons
onto NeuLAND (14 m distance, 4.5 degrees angular spread, box generator). When I
run this simulation under TGeant4, I only find a few different tracks is the MCTrack-branch
of the output-file. When I however run the same simulation under TGeant3, The MCTrack
branch is much larger. In further analysis, this results in that Geant4 gives only
40% detection efficiency for NeuLAND, while Geant3 given 99%. Does anyone know
why my Geant4 results are wrong?
Christiaan.
PS: The histogram included is the size of the MCTrack branch under Geant4.
|
|
|
|
|
|
| Re: Problem with TGeant4? [message #19351 is a reply to message #19347] |
Tue, 28 June 2016 12:43  |
Dmytro Kresan
Messages: 166
Registered: June 2004
|
first-grade participant |
From: *gsi.de
|
|
Geant3 is well known to reproduce neutron interaction accurately, also for lower energies. We urgently need a Geant3 / Geant4 / Experiment comparison and validation, in order to select the best suitable physics list. Till this is done, I would not trust any Geant4 simulations for NeuLAND.
But still, you can try to improve it. For that you need to modify the file R3BRoot/gconfig/g4r3bconfig.in
1. Comment out the old ion physics with #
#/R3B/phys/addPhysics binary_ion
2. And add two following lines:
/R3B/phys/addPhysics ion_inclxx
/R3B/phys/addPhysics qgsp_bert
3. In addition, in this file you can also set your electron cut (rangeCutForElectron):
/mcPhysics/rangeCutForElectron 1000000. mm
Attached is the distribution of MC tracks showing effect of this new physics list. I suppose the range cuts still have to be adjusted to achieve event better agreement.
Please note. In case of Geant3, the secondary MC tracks are saved to output only if they produced at least 1 MC point in a detector. To have the same for Geant4, you need to open file R3BRoot/gconfig/g4R3bConfig.C and change line 62 to:
stack->SetMinPoints(1);
Best regards,
Dima
-
Attachment: c1.png
(Size: 35.19KB, Downloaded 287 times)
|
|
|
|
|
|
| Re: Problem with TGeant4? [message #19354 is a reply to message #19351] |
Wed, 29 June 2016 11:27 |
C. A. Douma
Messages: 88
Registered: September 2015
Location: Groningen
|
continuous participant |
From: *kvi-cart.rug.nl
|
|
Dear Mr. Kresan,
Your advice reminded me of an earlier discussion we had about how to adapt the Geant4 Physics list to
cut of the delta electrons for heavy ions. When making those and your new modifications, all problems
were solved. The Geant4 detection effciency for NeuLAND is now about 99% (not resolving efficiency, that one
is about 95%).
I included the physics list that I used for this. It is indeed quite different from the current one!
Christiaan.
|
|
|
|
GSI Forum
Search
Help
Members
Login
Home
