Cut on chi^{2} of tracks 

The primaries tracks are flagged with a cut on dcaGlobal < 3cm.

However, there may be some clean-up to do using a cut on chi^{2} (old primary track flag)

The motivation was to removed to see where the tracks (with tpc+ssd only) with a large dca_{xy} or dca_{z} came from.

This cause this shape in the dca resolution in XY for the previous test production : 2007 P08ic MinBias 

dca resolution in Z : 2007 P08ic MinBias

Following are represented the dca_{xy} and dca_{z} as a function of the chi^{2} of primaries tracks for 01,2,3 and 4 Silicon hits

 Fig 1 : dca_{xy} vs chi^{2}

Fig 2 : dca_{z} vs chi^{2}

I fixed the cut on the chi^{2} at chi^{2}<4 for all tracks

Fig.3 dca_{xy} resolution vs. 1/P

Fig.4 dca_{z} resolution vs. 1/P

 Fig 5. dca_{xy} resolution vs. 1/P for tracks with TPC+SSD for different cuts of the chi^{2} 

Fig 6. dca_{z} resolution vs. 1/P for tracks with TPC+SSD for different cuts of the chi^{2} 

observations :

• cut on chi2<30 seems to remove the large values of dca
• however, a same cut on chi2 is applied for both dca_xy and dca_z

Fig 7. dca_{xy} resolution vs. 1/P for tracks with TPC+SSD+SVT(>1) for different cuts of the chi^{2} 

Fig 8. dca_{z} resolution vs. 1/P for tracks with TPC+SSD+SVT(>1) for different cuts of the chi^{2} 

part II : studies of chi^2

The idea is to use (cut) the chi^2 instead of applying a cut on the dca, or the momentum of tracks.

However,  chi^2, momentum and dca are correlated together :

• lower momentum will gives greater dca 
• greater dca seems to appear up to a certain chi^2 (see Fig.1)
•  

Fig 2.a : chi^{2} of tracks (N=0,1,2,3,4) vs 1/P

observations :

• there are many tracks (for N==1) with low momentum and high chi2
• the cut on chi2<30 removed those tracks
• do not appear for N>1

Fig 2.b : population of tracks (with N=0,1,2,3,4) under the chi^2 cut 

Fig 2.c : chi^{2} of tracks (N=0,1,2,3,4) vs eta

For this sample of data (I did run over less files than previously), I get for the population of tracks (in percentage, have to X by 100)

•  Si = 0 : entries = 0.418126
•  Si = 1 : entries = 0.0806296
•  Si = 2 : entries = 0.305136
•  Si = 3 : entries = 0.161612
•  Si = 4 : entries = 0.0344961