MWDC track reconstruction
As I mentioned last week, I made progress regarding the
track reconstruction efficiency in the BigBite MWDCs. I followed
Ole's instructions and varied the parameters that he suggested.
I made following modifications:
1.) BB.mwdc.xp.res: 5E-4 -----> 2E-3
2.) BB.mwdc.chi2_conflevel: 1E-5 -----> 5E-3
3.) BB.mwdc.maxmiss: 0 -----> 1
With these changes I get much better efficiency now. When checking
no track data (BB.tr.n==0) with EVe, now I can no longer observe
events, for which I could visually find a track. With respect to the
hits in the scintillator, I can now find tracks for 75% of event
(before it was 50%). However, If I limit myself only to those events,
which have valid hits in both chambers, then I estimated the new track
reconstruction efficiency to about 96%.
This week introduced these new parameters to my analysis files
and checked how the number of reconstructed events (at the
end of the primary analysis) changed. Beside the change to
the parameters above, I made no other change to the analysis
scripts.
Before the change (#2301):
Name Def Called Passed
-----------------------------------------------------------------------------------------------------------------------------------------------------------
BLOCK: Decode
TriggerT3 (DL.evtypebits&8)==8 5006903 4134385 (82.6%)
CoincT5 (DL.evtypebits&32)==32 5006903 1188424 (23.7%)
CoincT6 (DL.evtypebits&64)==64 5006903 570981 (11.4%)
Decode_master CoincT5||CoincT6||TriggerT3 5006903 4940387 (98.7%)
BLOCK: Tracking
u1Hits BB.mwdc.u1.nhits>0 4940387 3753149 (76%)
u1pHits BB.mwdc.u1p.nhits>0 4940387 3763381 (76.2%)
v1Hits BB.mwdc.v1.nhits>0 4940387 3719948 (75.3%)
v1pHits BB.mwdc.v1p.nhits>0 4940387 3690687 (74.7%)
x1Hits BB.mwdc.x1.nhits>0 4940387 3755166 (76%)
x1pHits BB.mwdc.x1p.nhits>0 4940387 3753469 (76%)
u2Hits BB.mwdc.u2.nhits>0 4940387 3733098 (75.6%)
u2pHits BB.mwdc.u2p.nhits>0 4940387 3758939 (76.1%)
v2Hits BB.mwdc.v2.nhits>0 4940387 3855413 (78%)
v2pHits BB.mwdc.v2p.nhits>0 4940387 3829832 (77.5%)
x2Hits BB.mwdc.x2.nhits>0 4940387 3801365 (76.9%)
x2pHits BB.mwdc.x2p.nhits>0 4940387 3821970 (77.4%)
uProj (u1Hits&&u1pHits&&u2Hits)||(u1Hits&&u1pHits&&u2pHits)||(u1pHits&&u2Hits&&u2pHits)||(u1Hits&&u2Hits&&u2pHits) 4940387 3483462 (70.5%)
vProj (v1Hits&&v1pHits&&v2Hits)||(v1Hits&&v1pHits&&v2pHits)||(v1pHits&&v2Hits&&v2pHits)||(v1Hits&&v2Hits&&v2pHits) 4940387 3513157 (71.1%)
xProj (x1Hits&&x1pHits&&x2Hits)||(x1Hits&&x1pHits&&x2pHits)||(x1pHits&&x2Hits&&x2pHits)||(x1Hits&&x2Hits&&x2pHits) 4940387 3533344 (71.5%)
ThreeProjections uProj&&vProj&&xProj 4940387 2444135 (49.5%)
uNRoads BB.mwdc.u.nroads>0 4940387 1622274 (32.8%)
uNgood BB.mwdc.u.ngood>0 4940387 1395169 (28.2%)
vNRoads BB.mwdc.v.nroads>0 4940387 1581759 (32%)
vNgood BB.mwdc.v.ngood>0 4940387 1365404 (27.6%)
xNRoads BB.mwdc.x.nroads>0 4940387 1497159 (30.3%)
xNgood BB.mwdc.x.ngood>0 4940387 1328484 (26.9%)
NoBBTrack BB.tr.n==0 4940387 3962756 (80.2%)
BBTrack BB.tr.n>0 4940387 977631 (19.8%)
Tracking_master BBTrack 4940387 977631 (19.8%)
BLOCK: Physics
HRSLGoldDp L.gold.dp>-0.045&&L.gold.dp<0.045 977631 886085 (90.6%)
HRSLGoldTh L.gold.th>-0.06&&L.gold.th<0.06 977631 863059 (88.3%)
HRSLGoldPh L.gold.ph>-0.03&&L.gold.ph<0.03 977631 840172 (85.9%)
BBTpTrackMatch BB.tp.trHitIndex>-1 977631 965224 (98.7%)
TargetSizeCut ReactPt_L.z>-0.30&&ReactPt_L.z<0.30 977631 872558 (89.3%)
Physics_master TargetSizeCut&&BBTpTrackMatch&&HRSLGoldDp&&HRSLGoldTh&&HRSLGoldPh 977631 798975 (81.7%)
After the modification (#2301):
Name Def Called Passed
-----------------------------------------------------------------------------------------------------------------------------------------------------------
BLOCK: Decode
TriggerT3 (DL.evtypebits&8)==8 5006903 4134385 (82.6%)
CoincT5 (DL.evtypebits&32)==32 5006903 1188424 (23.7%)
CoincT6 (DL.evtypebits&64)==64 5006903 570981 (11.4%)
Decode_master CoincT5||CoincT6||TriggerT3 5006903 4940387 (98.7%)
BLOCK: Tracking
u1Hits BB.mwdc.u1.nhits>0 4940387 3753149 (76%)
u1pHits BB.mwdc.u1p.nhits>0 4940387 3763381 (76.2%)
v1Hits BB.mwdc.v1.nhits>0 4940387 3719948 (75.3%)
v1pHits BB.mwdc.v1p.nhits>0 4940387 3690687 (74.7%)
x1Hits BB.mwdc.x1.nhits>0 4940387 3755166 (76%)
x1pHits BB.mwdc.x1p.nhits>0 4940387 3753469 (76%)
u2Hits BB.mwdc.u2.nhits>0 4940387 3733098 (75.6%)
u2pHits BB.mwdc.u2p.nhits>0 4940387 3758939 (76.1%)
v2Hits BB.mwdc.v2.nhits>0 4940387 3855413 (78%)
v2pHits BB.mwdc.v2p.nhits>0 4940387 3829832 (77.5%)
x2Hits BB.mwdc.x2.nhits>0 4940387 3801365 (76.9%)
x2pHits BB.mwdc.x2p.nhits>0 4940387 3821970 (77.4%)
uProj (u1Hits&&u1pHits&&u2Hits)||(u1Hits&&u1pHits&&u2pHits)||(u1pHits&&u2Hits&&u2pHits)||(u1Hits&&u2Hits&&u2pHits) 4940387 3483462 (70.5%)
vProj (v1Hits&&v1pHits&&v2Hits)||(v1Hits&&v1pHits&&v2pHits)||(v1pHits&&v2Hits&&v2pHits)||(v1Hits&&v2Hits&&v2pHits) 4940387 3513157 (71.1%)
xProj (x1Hits&&x1pHits&&x2Hits)||(x1Hits&&x1pHits&&x2pHits)||(x1pHits&&x2Hits&&x2pHits)||(x1Hits&&x2Hits&&x2pHits) 4940387 3533344 (71.5%)
ThreeProjections uProj&&vProj&&xProj 4940387 2444135 (49.5%)
uNRoads BB.mwdc.u.nroads>0 4940387 2870993 (58.1%)
uNgood BB.mwdc.u.ngood>0 4940387 2799056 (56.7%)
vNRoads BB.mwdc.v.nroads>0 4940387 2781809 (56.3%)
vNgood BB.mwdc.v.ngood>0 4940387 2717105 (55%)
xNRoads BB.mwdc.x.nroads>0 4940387 2607196 (52.8%)
xNgood BB.mwdc.x.ngood>0 4940387 2495844 (50.5%)
NoBBTrack BB.tr.n==0 4940387 3456327 (70%)
BBTrack BB.tr.n>0 4940387 1484060 (30%)
Tracking_master BBTrack 4940387 1484060 (30%)
BLOCK: Physics
HRSLGoldDp L.gold.dp>-0.045&&L.gold.dp<0.045 1484060 1331806 (89.7%)
HRSLGoldTh L.gold.th>-0.06&&L.gold.th<0.06 1484060 1295931 (87.3%)
HRSLGoldPh L.gold.ph>-0.03&&L.gold.ph<0.03 1484060 1262095 (85%)
BBTpTrackMatch BB.tp.trHitIndex>-1 1484060 1345248 (90.6%)
TargetSizeCut ReactPt_L.z>-0.30&&ReactPt_L.z<0.30 1484060 1309889 (88.3%)
Physics_master TargetSizeCut&&BBTpTrackMatch&&HRSLGoldDp&&HRSLGoldTh&&HRSLGoldPh 1484060 1102317 (74.3%)
From this comparison we see, that now we get ~40% more events. This is a significant
change in the number of events and I believe, that it would be worth it to redo
the analysis on FARM and get this additional statistics.
The low track efficiency at the end (30%) is caused by the fact, that we are not
accepting only coincidence events but also HRS-L single events. This is done because
some coincidences appear in the TS as singles. However, if only coincidence triggers
T5 and T6 would be accepted, then 92% of events would have tracks.
Effective Deuteron Pz and Pzz polarizations inside 3He
I considered Carlson/Doug's idea and tried to
determine deuteron vector and tensor polarizations (Pz and Pzz),
by taking our data at pmiss=0 and comparing them to the
deuteron elastic asymmetry. For that I made quick calculation in
Mathematica (see DeuteronElasticPzPzz.pdf ):
For the asymmetries I took AL = -0.3 +/- 0.008 and AT = -0.05 +/- 0.005.
The calculation of the elastic asymmetry was done for the center
of our elastic acceptance (theta_e = 13.2deg). This way I
obtained following values for polarizations:
Pz = 0.68 and Pzz = 1.1
I somehow expected such results, since I knew from my previous analysis,
that parameters (Pz = 2/3, Pzz = 1) gave good agreement
between the measurements and the calculated elastic asymmetry.
However, the thing that is bothering me, is that I can not find (mathematical)
reason for these values within the limits of my simple S-state-only model.
If Pz would be 1, then Pzz = 1 seems OK. However combination Pzz = 1 and
Pz = 2/3 is not consistent with the calculation of the C-G coefficients.
Last modified: 10/17/12