GSI Forum
GSI Helmholtzzentrum für Schwerionenforschung

Home » PANDA » PandaRoot » Analysis » Vertex fitter for two consecutive decays
Vertex fitter for two consecutive decays [message #17572] Tue, 25 November 2014 11:49 Go to next message
Karin Schönning
Messages: 65
Registered: August 2012
Location: Uppsala University
continuous participant
From: *physics.uu.se
Dear Pandaroot experts,

I am trying to use the vertex fitter to improve the resolution of the Xi- mass, reconstructing Xi- from the decay Xi- -> Lambda pi-, Lambda -> p pi- .
I have used the vertex fitter for the lambda and it works. I take the fitted lambda candidates and combine with a pion which I have checked, using MC truth match, that it is a daughter of Xi- and not Lambda. Then I try to apply the vertex fit again, see below:

------------------------------------------------------------------------ -------------------------

Xi.Combine(goodlam,pim);
Xibar.Combine(goodlamb,pip);
for (j=0;j<Xi.GetLength();++j)
{
//PndVtxPRG vtxfitterx(Xi[j]);
PndKinVtxFitter vtxfitterx(Xi[j]); // instantiate a vertex fitter

bool checkx = vtxfitterx.Fit();
double chi2_vtx = vtxfitterx.GetChi2(); // access chi2 of fit
double prob_vtx = vtxfitterx.GetProb(); // access probability of fit

if (checkx) // when good enough, fill some histos
{

RhoCandidate *Xiv = Xi[j]->GetFit(); // access the fitted cand
TVector3 lVtx=Xiv->Pos();
double Ximassv=Xiv->M();
double xivz=lVtx.Z();

cout<<"Xibar mass : "<<Xi[j]->M()<<endl;
cout<<"Xibar mass vertex: "<<Ximassv<<endl;
cout<<"Xibar vertex z: "<<xivz<<endl;

Ximass->Fill(Xiv->M());}

}


-------------------------------------------------------------------

But when running this I get into trouble, the printout is shown below. Is the vertex fitter supposed to work in this way or does it only handle single vertices?

Error in <TDecompLU::DecomposeLUCrout>: matrix is singular
Error in <TDecompLU::InvertLU>: matrix is singular, 0 diag elements < tolerance of 2.2204e-16
Xibar mass : 0.186496
Xibar mass vertex: 0.167329
Xibar vertex z: 69.2357
Xibar mass : 1.32154
Xibar mass vertex: -nan
Xibar vertex z: -nan
Xibar mass : 1.31012
Xibar mass vertex: -nan
Xibar vertex z: -nan
Xibar mass : 1.32721
Xibar mass vertex: -nan
Xibar vertex z: -nan

Does anybody have an idea of what one can do about this?

Kindest regards,
/Karin
Re: Vertex fitter for two consecutive decays [message #17574 is a reply to message #17572] Tue, 25 November 2014 12:00 Go to previous messageGo to next message
Ralf Kliemt is currently offline  Ralf Kliemt
Messages: 507
Registered: May 2007
Location: GSI, Darmstadt
first-grade participant

From: *gsi.de
Hi Karin,

The Lambda is a neutrlly charged particle. Our fitters cannot create avalid Helix & Covariance from that.
Since the Xi does fly some distance my only suggestion is you try and make an analytical helix-line point-of-closest-approach finder by hand.


Cheers
Ralf
Re: Vertex fitter for two consecutive decays [message #17577 is a reply to message #17572] Tue, 25 November 2014 12:01 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 Karin,


I don't see how you can fit a second secondary vertex (of the Xi-), since this vertex is not constraint by at least two charged tracks. A decay X -> pi+ pi- Lambda could deliver something like that, but not one with X -> pi- Lambda.


Best,
Klaus
Re: Vertex fitter for two consecutive decays [message #17578 is a reply to message #17577] Tue, 25 November 2014 12:04 Go to previous messageGo to next message
Karin Schönning
Messages: 65
Registered: August 2012
Location: Uppsala University
continuous participant
From: *physics.uu.se
Thanks for the rapid response. I wasn't aware of the inner machinery and thought that maybe one could make a fit based on four vectors instead.

Cheers,
/Karin
Re: Vertex fitter for two consecutive decays [message #17579 is a reply to message #17577] Tue, 25 November 2014 12:05 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
Why not? A neutral track has a momentum, a direction, and a covariance matrix, you can variate such parameters to find better values, moreover adding a charged track you can find the PVA point, thus finding the production point of the charged.
I think this is a feature missing in the code, but this is possible to implement.
Re: Vertex fitter for two consecutive decays [message #17580 is a reply to message #17579] Tue, 25 November 2014 12:18 Go to previous message
Karin Schönning
Messages: 65
Registered: August 2012
Location: Uppsala University
continuous participant
From: *physics.uu.se
In Erik Thomé's thesis he writes that

"Pairs of Λ¯ (Λ) and π+ (π−) were fitted to a common vertex under a ¯Ξ+
(Ξ−) hypothesis, requiring a χ2 probability > 0.001."

He used the old framework but since this is on the analysis level maybe there is only some cutting and pasting required. I will try to get a hold of him (he quitted academia and I am not sure if his email works) and check if he has some code on his own or if there was some standard part of pandora.

Cheers,
/Karin
Previous Topic: Different results for same information extracted in different ways
Next Topic: No back propagation to IP for V_0 reconstruction
Goto Forum:
  


Current Time: Wed Dec 04 03:44:58 CET 2024

Total time taken to generate the page: 0.00764 seconds