GSI Forum
GSI Helmholtzzentrum für Schwerionenforschung

Home » PANDA » PandaRoot » Tracking » very high reconstructed momenta of muons
very high reconstructed momenta of muons [message #13308] Thu, 12 April 2012 00:58 Go to next message
Malgorzata Gumberidze is currently offline  Malgorzata Gumberidze
Messages: 98
Registered: June 2005
continuous participant
From: *gsi.de
Dear tracking experts.

During test of the TMVA i was surprise to see some number of track with reconstructed momenta which are very high in comparison to the monte carlo
value (Look into the attached plot: momentum.gif, keep in mind that i have also higher values, but i put limit on x-axis).

So let's me try to explain what i was dung.

I'm using PandaRoot version nov11 and externals may11.

I have simulated muons (flat distributions using pgun) in the momentum range:
0.2 GeV/c till 5GeV/c. They in my analysis i select

1. only primary tracks
if (mctrack->GetMotherID()!=-1) continue;

2. i check if the reconstructed track is correlated to the monte carlo one:
if (pidCand->GetMcIndex()!=mc) continue;

3. if i have more then one reconstructed track which belong to the same monte carlo i select one which has momentum closest to monte carlo one.

I tried to have a look (after some advices from Stefano) if one can find some topology of this high momenta looking into: number of hits in MVD or STT or
theta or phi or chi2 from fit .......

SO in the attachment you can see several plots. All plots are done as a function of momentum, so you can see not only problematic region, but also good one. In the case when i plot theta vs phi i put cut on p > 6 GeV/c.

In general looking into theta or phi it does not look that this high momenta comes from some specific region of the PANDA. In case of STT hits it looks that they have smaller number of the hits, below 28. But generally from the topology they do not look much different then good momenta track, but maybe i have overlook something.

I hope that the names of plots are self-explaining.

So i do not know if i do something wrong, or it is like it is …. at least it looks strange.


gosia

[Updated on: Fri, 13 April 2012 08:12]

Report message to a moderator

Re: very high reconstructed momenta of muons [message #13317 is a reply to message #13308] Thu, 12 April 2012 22:12 Go to previous messageGo to next message
Malgorzata Gumberidze is currently offline  Malgorzata Gumberidze
Messages: 98
Registered: June 2005
continuous participant
From: *pool.mediaWays.net
Please find in the attachment 3 more plots, which maybe this time shows that
big statistics of this high momenta come from the regions of phi around 90 drgree,
and 2 regions of theta: 90degree and below 30 degree.


gosia
Re: very high reconstructed momenta of muons [message #13338 is a reply to message #13317] Tue, 17 April 2012 00:42 Go to previous messageGo to next message
Malgorzata Gumberidze is currently offline  Malgorzata Gumberidze
Messages: 98
Registered: June 2005
continuous participant
From: *pool.mediaWays.net
Dear All,

I have made few more test after discussions with Stefano in order to check
where the high momenta tail is coming from: tracking code or Kalman
(as it was suggested by Gianluigi).

What i have done is to switch on native tracking by using in the pid level:

PndPidCorrelator* corr = new PndPidCorrelator();
corr->SetInputBranch("SttMvdGemTrack");
corr->SetInputIDBranch("SttMvdGemTrackID");
corr->SetDebugMode(kFALSE);
fRun->AddTask(corr);


One can see that the tail at high momenta (up to 15GeV/c) is present
also with native tracking. So it means it is not coming from genfit.

It would be nice that experts can cross check this problem in the code as well.

gosia
Re: very high reconstructed momenta of muons [message #13345 is a reply to message #13338] Tue, 17 April 2012 11:13 Go to previous messageGo to next message
StefanoSpataro is currently offline  StefanoSpataro
Messages: 2736
Registered: June 2005
Location: Torino
first-grade participant

From: *to.infn.it
I suppose that chi2 and ndf should be filled also by Gianluigi code, considering that a fit is performed, while at present such data members are empty.
Gianluigi, would it be possible to add such informations to the PndTrack?

From the plots is quite evident that tracks with fahe high momentum are present. Probably there is an upper cut at 15GeV/c in the reconstruction code, but still there are tracks with momenta higher than 5 GeV/c, coming from the edges of the detector.
Would it be possible to take a look? For sure they are not coming from genfit.
Re: very high reconstructed momenta of muons [message #13351 is a reply to message #13308] Tue, 17 April 2012 13:39 Go to previous messageGo to next message
Gianluigi Boca is currently offline  Gianluigi Boca
Messages: 177
Registered: March 2004
first-grade participant
From: *gsi.de
hi Gosia and all,
concerning these high momenta muons, please consider doing these
two preliminary steps :

1) please update your code with the latest trunk version and use the jan12 externals; I believe it is always safer to use the newest versions of the code;


2) please quantify the percentage of muon tracks with 'anomalous' large momentum; from your plots it is not clear to me what is the level of the problem we are talking about

thanks

Gianluigi
Re: very high reconstructed momenta of muons [message #13356 is a reply to message #13351] Tue, 17 April 2012 15:51 Go to previous message
Lia Lavezzi
Messages: 291
Registered: May 2007
Location: Torino
first-grade participant

From: *pv.infn.it
Hi Gosia,
I just committed a fix to the class PndSttTube, so I would recommend to update the code to the last revision (15297) to be sure to include it.
Ciao,
Lia.
Previous Topic: new task and directory
Next Topic: Genfit with Forward Tracks
Goto Forum:
  


Current Time: Tue Oct 15 17:51:49 CEST 2024

Total time taken to generate the page: 0.00900 seconds