Code is at:

/star/u/rjreed/PPV2009/Eval500GeV02/

compiled in stardev

 

Code was run over 670080 low luminosity events.    All events were trigger 4 (BBCMB).

MuDsts are at:/star/data05/scratch/balewski/2009-muDst-F10421/10083133/mu
/star/data05/scratch/balewski/2009-muDst-F10421/10083139/mu
/star/data05/scratch/balewski/2009-muDst-F10421/10083143/mu

TPC calib & BFC used for this production of F10421
TPC sector alignment: old values (last updated for Run 7)
TPC twist alignment: old values (last updated for Run 8)
BeamLine constraint: preliminary (done with the above alignments, no SpaceCharge+GridLeak)
TPC SpaceCharge+GridLeak: preliminary for pp500

BFC:"pp2009a ITTF BEmcChkStat QAalltrigs btofDat Corr3 OSpaceZ2 OGridLeak3D beamLine -VFMinuit  VFPPVnoCTB -dstout -evout"

2)---- PPV vertex filtering. Now there are 3 groups of vertices :
a) multiple match, rank>1e6
b) one track match w/ pT>15, 0<rank<1e6
c) first 5 no match vertices, rank <0

 

There were only 1177 events with no vertices.  (0.18%)

There were 249454 events with no postive ranked vertices. (37.2%)

 

There were only 14 events where the primary vertex had a rank > 0, but rank <1e6

There were no events where a vertex with a lower order had a higher rank than the primary vertex.

 

For some reason, I couldn't get any vpd Vz.  I will look to see if there is another method.   I used:

  StMuDst* mMuDst = muDstMaker->muDst();
  StMuEvent* mEvent = mMuDst->event();
  float vpdVz = mEvent->vpdVz();
 

I also did not get any events with tracks crossing the TPC CM, which doesn't make sense.  I used the command:

mMuDst->primaryVertex()->nCrossCentralMembrane();

Jan confirms that there are no tracks crossing the TPC CM in the log files either.  Something isn't right here.

 

There were 18083 events with a bbc vertex.  (2.7%)

 

Figure 1:  Rank of the primary vertex where the following statement was used:

if rank > 1e6

fill(log(rank - 1e6) +10)

if 1e6>rank>0

fill(log(rank))

if rank<0

fill(log(rank+1e6)-10)

 

Figure 2: Rank of first 6 vertices with the same offsets as in Figure 1.

 

I will now drop all vertices with rank <0.

Figure 3: # of vertices/event with positive rank

Figure 4: Primary vertex z position distribution for rank>0 fit with a gaussian.  (Incidentally, with no rank requirement a gaussian fit on [-50,50] yields A = 7440, mu = 1.773 and sigma = 30.3

Figure 5: Primary vertex z position for vertices with rank > 0 in black.  Red is the next vertex (->primaryVertex(1)) with postive rank scaled by 30 so that both distributions could be compared. 

There are so few vertices past the primary vertex with rank > 0, it doesn't make sense to draw the other vz distributions.  I refer to the next vertex on the list as the secondary vertex since we use the word rank to mean two entirely different things.  (Maybe we need to use a new word?  Order perhaps?)

Figure 6: This the error in the z position of the primary and secondary vertices. 

 

Figure 7: Vz from the bbc.  I do not know why it is shifted in Vz.  This seems strange given how symmetric the PPV vertices are.

 

Figure 8: Vz0 - bbc Vz for rank > 0 fit with gaussian.  Adding the width of Vz0 and bbcVz in quadrature gives us (28.44^2+14.16^2)^0.5 = 31.7 which agrees with the above fit!

Figure 9: The black curve is Vz0 - bbc Vz for rank > 0.  The red curve is Vz - bbc Vz for the closest vertex to the bbc with rank > 0

 

Figure 10: # of tracks matched to the BEMC for primary and secondary vertices with rank > 0.  Black = primary.  Red = secondary scaled by 15 to put both on the same graph.

Figure 11: Same as Figure 10 but for the EndCap

 

Figure 12: # of tracks used to construct vertex for primary and secondary vertices with rank > 0.   Black = primary, red = secondary scaled by 15

 

Figure 13: Refmult associated with the primary and secondary vertex with rank > 0.  black = primary, red = secondary x 15

 

3 concerns:

- why we do not see tracks crossing central membrane & approach beam line
- why only 18K out of 670K events triggerd by BBC have offline BBC vertex
- why do we not see the VPD vertex