GSI Forum: Bugs, Fixes, Releases » Problem with PndKinVtxFitter and Ks decays

Home » PANDA » PandaRoot » Bugs, Fixes, Releases » Problem with PndKinVtxFitter and Ks decays

Show: Today's Messages :: Polls :: Message Navigator

Problem with PndKinVtxFitter and Ks decays [message #14243]

Wed, 28 November 2012 13:01

2dd180b0
Messages: 2
Registered: September 2011

occasional visitor

From: *ikp.kfa-juelich.de

Dear all,

I've come across a problem while trying to reconstruct the following decay:
D0 -> Ks pi+ pi-, Ks -> pi+ pi-

The procedure I use is to first use PndKinVtxFitter to fit the pi+ pi- from the Ks decay to a common vertex, and then use PndKinVtxFitter to combine the fitted Ks with another pi+ pi- pair, to reconstruct the D0. The problem is that the fit to the D0 -> Ks pi+ pi- always returns nan for the chi^2.

The fitting code I use is exactly analogous to that used by Marius to reconstruct psi(3770) -> D+ D-, D+- -> K+- pi+ pi-, and we think that the problem is due to PndKinVtxFitter not being able to properly handle the neutral Ks candidate.

I've attached the analysis code and EvtGen decay file that I used for this analysis.

Best,
Sean

Attachment: psi3770.dec
(Size: 0.20KB, Downloaded 274 times)
Attachment: run_ana_dobbs.C
(Size: 4.51KB, Downloaded 281 times)

Report message to a moderator

Re: Problem with PndKinVtxFitter and Ks decays [message #14244 is a reply to message #14243]

Wed, 28 November 2012 13:14

Ralf Kliemt
Messages: 507
Registered: May 2007
Location: GSI, Darmstadt

first-grade participant

From: *pool.mediaWays.net

Hallo Dean,

It is a "known issue" that the vertex fitters don't handle neutral particles correctly. In the new year I'll invest some time to repair that behaviour.

Bear in mind that the vertex fitters primarily are made for charged tracks under the common vertex hypothesis. You can, however, fit on the vertex of your pions from the D0 and select for the chi2. The cleaned pion sample is used to combine with the K0 to the D0. You can then apply the 4C fit (knowing the initial system) on both D0 and aD0 for further improvement.

Cheers
Ralf Kliemt

Report message to a moderator

Re: Problem with PndKinVtxFitter and Ks decays [message #14245 is a reply to message #14244]

Wed, 28 November 2012 14:09

Albrecht Gillitzer
Messages: 49
Registered: March 2004

continuous participant

From: *ikp.kfa-juelich.de

Hello Ralf,

I have another problem with PndKinVtxFitter: After fitting the mu+ and mu- tracks from J/psi decay to a common vertex (and in case of more than one found combination taking the one with the best chi2) and applying a chi2 cut, the mu+mu- invariant mass distribution has a clearly larger width than obtained with the raw, unfitted tracks. I would expect the opposite since after fitting and chi2 cut the result should be closer to reality.
Is this also a "known issue"? And could you please also look into this problem?

cheers, Albrecht

Report message to a moderator

Re: Problem with PndKinVtxFitter and Ks decays [message #14246 is a reply to message #14244]

Wed, 28 November 2012 14:13

2dd180b0
Messages: 2
Registered: September 2011

occasional visitor

From: *ikp.kfa-juelich.de

Dear Ralf,

Thanks for the information! I'll try doing what you've suggested.

Cheers,
Sean

Report message to a moderator

Re: Problem with PndKinVtxFitter and Ks decays [message #14247 is a reply to message #14245]

Wed, 28 November 2012 15:18

Ralf Kliemt
Messages: 507
Registered: May 2007
Location: GSI, Darmstadt

first-grade participant

From: *pool.mediaWays.net

Hello Albrecht,

From the fitter algorithm point of view I think it is not necessary that the composite mass resolution has to improve with a vertex fit, especiually with only two particles at hand. In principle these fitters adjust the momentum directions of the daughter particles.
Could you plot the dependence between the chi2 and the mass resolution?

Cheers
Ralf

Report message to a moderator

Previous Topic:	[FIXED] Crash in PndDrcHitProducerIdeal with Mai2011 packages
Next Topic:	[WARN] Removing candidates while looping over TCandList

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Nov 27 07:53:58 CET 2024

Total time taken to generate the page: 0.00954 seconds