GSI Forum
GSI Helmholtzzentrum für Schwerionenforschung

Home » PANDA » PandaRoot » Analysis » [FIXED] Problem in reconstructing neutral particle
[FIXED] Problem in reconstructing neutral particle [message #15393] Fri, 30 August 2013 23:42 Go to next message
Lu Cao is currently offline  Lu Cao
Messages: 77
Registered: February 2013
continuous participant
From: *adsl.alicedsl.de
Dear all,

I'm reconstructing this decay chain:
Ds+ -> nu_e e+ eta
                |->pi+ pi- pi0
                            |-> gamma gamma



In reconstruction, since the photons come from three different components of EMC(with different angle and energy cuttings), i.e. barrel, forward endcap, backward endcap, I need to take all possible combinations into account of the pi0 candidates. My analysis can run smoothly when the combinations are only related to barrel and forward endcap. When one/both of the two gamma(s) come from the backward endcap EMC, the macro will be crashed after 500evt with the following info:

evt 100
evt 200
evt 300
evt 400
evt 500
ERROR: attempt to call RhoCandidate::SetType("pi0 | Meson | Q=0") for a composite RhoCandidate whose daughters have total charge -3
ERROR: attempt to call RhoCandidate::SetType("pi0 | Meson | Q=0") for a composite RhoCandidate whose daughters have total charge -2
ERROR: attempt to call RhoCandidate::SetType("pi0 | Meson | Q=0") for a composite RhoCandidate whose daughters have total charge -1
*** Candidate is its own mother???  *** 


I use FillList(gam,"Neutral") to get the photon list though I know it has some problems with PdgCode setting. From the error message, it seems that some charged particles are misfilled into the neutral list.

I commented some combinations related to the bwd part, and found that in some occasions the crash happens without any error info about charge but only "** Candidate is its own mother??? **". Thus, I guess if there are some pi0 in the gamma list as well.

From the other hand, I don't understand why this crash only happens with the backward endcap EMC. If this problem is purely due to the incorrect filling of neutral list, these errors should be also equally posted by other two EMC components, but they don't actually.


Thanks in advance for all comments and suggestions!


Best,
Lu

[Updated on: Mon, 02 September 2013 14:28] by Moderator

Report message to a moderator

Re: Problem in reconstructing neutral particle [message #15394 is a reply to message #15393] Sat, 31 August 2013 00:08 Go to previous messageGo to next message
Klaus Götzen is currently offline  Klaus Götzen
Messages: 293
Registered: June 2006
Location: GSI
first-grade participant
From: *adsl.alicedsl.de
Dear Lu,


could you please attach your analysis macro? I think that would help to investigate the problem.


Best,
Klaus
Re: Problem in reconstructing neutral particle [message #15401 is a reply to message #15394] Mon, 02 September 2013 11:14 Go to previous messageGo to next message
Lu Cao is currently offline  Lu Cao
Messages: 77
Registered: February 2013
continuous participant
From: *ikp.kfa-juelich.de
Hello Klaus,

Here's my analysis macro.
Thanks in advance.

Best,
Lu
  • Attachment: ana_pi0.C
    (Size: 12.05KB, Downloaded 251 times)
Re: Problem in reconstructing neutral particle [message #15402 is a reply to message #15401] Mon, 02 September 2013 12:00 Go to previous messageGo to next message
Klaus Götzen is currently offline  Klaus Götzen
Messages: 293
Registered: June 2006
Location: GSI
first-grade participant
From: *gsi.de
Hi Lu,


I was not able to reproduce your problem up to now, but I noticed, that you don't empty the RhoCandList bwd_gam around line 202/203.

You could add an
  bwd_gam.Cleanup();

and rerun to see whether this cures the error.

The problem could be, that the bwd_gam list still holds pointers to candidates from the previous event, which after RhoFactory::Reset() point to different candidates from the current event, which might be composites, charged or whatever kind of particles.

Btw, the filtering in lines 135 and 139 (cutting on the PdgCode()) actually does not filter, since these numbers from the reco candidates do not represent the MC truth codes, but just the assigned ones from particle selectors. I.e. all the candidates in a 'PionLoosePlus' list have PDG code 211 after filling the list.


Best,
Klaus
Re: Problem in reconstructing neutral particle [message #15403 is a reply to message #15402] Mon, 02 September 2013 14:03 Go to previous message
Lu Cao is currently offline  Lu Cao
Messages: 77
Registered: February 2013
continuous participant
From: *ikp.kfa-juelich.de
Hi Klaus,

Thanks pretty much for your suggestions.
It works fine now!Smile
I'm really happy the situation is much better than I thought.

Best regards,
Lu
Previous Topic: Pnd4CFitter - how to use it
Next Topic: [FIXED] Recontruction of neutrals - macro is very slow
Goto Forum:
  


Current Time: Tue Dec 03 23:20:08 CET 2024

Total time taken to generate the page: 0.00525 seconds