About FairGeanePro [message #15561] |
Thu, 10 October 2013 18:20 |
Jifeng Hu
Messages: 31 Registered: October 2012
|
continuous participant |
From: *physik.uni-giessen.de
|
|
Hi, All,
who has experiences of FairGeanePro propagation, please have a look at my usage and results, I think I got some wrong results.
For 2 GeV momentum pion from the origin (0,0,0), isotropically generated, the flying path is expected to be the EMC shower center position. To do that, the codes are listed (see also PndPidCorrelator.cxx ):
0. Bool_t PndPidCorrelator::GetEmcInfo(FairTrackParH* helix, PndPidCandidate* pidCand) {
1. FairGeanePro *fProEmc = new FairGeanePro();
2. for (Int_t ee = 0; ee<emcEntries; ee++){
3. PndEmcCluster *emcHit = (PndEmcCluster*)fEmcCluster->At(ee);
4. TVector3 emcPos = emcHit->where();
5. fProEmc->SetPoint(emcPos);
6. fProEmc->PropagateToPCA(1, 1);
7. FairTrackParH *fRes= new FairTrackParH();
8. Bool_t rc = fProEmc->Propagate(helix, fRes, fPidHyp*pidCand->GetCharge()); // First propagation at module
9. if (!rc) continue;
10. emcGLength = fProEmc->GetLengthAtPCA();
11. }
22.}
emcGLength is extracted as the flying path. see its distribution in barrel.eps for Barrel Calorimeter, and so on.
see the Panda TDR:
http://www-panda.gsi.de/archive/public/panda_tpr.pdf
the geometry tell us the flying path is wrong.
Someone can tell me where I incorrectly make use of FairGeanePro class.
-
Attachment: backward.eps
(Size: 10.72KB, Downloaded 396 times)
-
Attachment: forward.eps
(Size: 35.28KB, Downloaded 412 times)
-
Attachment: barrel.eps
(Size: 328.60KB, Downloaded 410 times)
-
Attachment: shashylik.eps
(Size: 17.37KB, Downloaded 426 times)
|
|
|
|
Re: About FairGeanePro [message #15563 is a reply to message #15562] |
Fri, 11 October 2013 09:30 |
Jifeng Hu
Messages: 31 Registered: October 2012
|
continuous participant |
From: *physik.uni-giessen.de
|
|
Stefano Spataro wrote on Thu, 10 October 2013 18:40 | Sorry,
what is exactly wrong in these plots? They look fine to me, apart from some high lenght background due probably to some propagation/tracking problem.
|
Well, maybe I misunderstand the flying path after propagation.
That means FairGeanePro only returns the length between last point of helix to EMC hit point, if yes, those distributions look good.
Here the flying path should be longer than the straight line from vertex to EMC hit position. I thought it the total length of one track, since I only read the code instead of the document.
[Updated on: Fri, 11 October 2013 09:30] Report message to a moderator
|
|
|
|
|