November

November Work

Changes for the globals hits positions (x,y,z) according to the alignment :

shift in X. dX=0.418 cm
shift in dZ=-0.083 cm
shift in phi : dPhi=0.0181 rad

I do first the 2 shifts in X and Z (2 translations) then I get another globals positions x1,z1
Then I do the rotation :

x' = x1cos(delta phi) + ysin(delta phi)
y' = -x1sin(delta phi) + ycos(delta phi)

Finally I have a new set of globals positions (x',y',z1) The same transformations are applied to the wafer geometry and there is another shift , which is the shift observed between wafers :

between wafer 7 and 8 : +/- 0.25mm
then +/-0.5mm
wafer 0 --> z neg . wafer 15--> z pos.
positions here are in cm

Verification of the sign of the rotation
As we don't know if dPhi = 0.0181 rad must be apply in which way ,I look the 2 cases : dPhi = 0.0181 and dPhi = - 0.0181
for event 167 of /star/data10/reco/cu62productionMinBias/FullField/P05if_test/2005/073/st_physics_6073006_raw_1020001.event.root

primaries	with ssd ,tpc ,svt points	with ssd ,tpc points	with svt ,tpc points	with tpc points
dPhi = 0.0181	10	5	14	8
dPhi = - 0.0181	1	5	23	8

So the rotation is dPhi = +0.0181 rad ( trigonometric )
Changes the tracking parameters
coeff[0]= intrinsicZ
coeff[1]= driftZ
coeff[2]= crossZ
coeff[3]= intrinsicX
coeff[4]= driftX
coeff[5]= crossX

ecross=coeff[0]+coeff[1]*dz/(cosCA*cosCA) +coeff[2]*tanCA*tanCA;
edip=coeff[3]+coeff[4]*dz*cosDipInv2+coeff[5]*tanDip*tanDip;

The defaults parameters are:
coeff[0]= 0.002
coeff[1]= 0.
coeff[2]= 0
coeff[3]= 0.002
coeff[4]= 0.
coeff[5]= 0.
so I just change
coeff[0]= 0.07

Some results :
for event 167 of /star/data10/reco/cu62productionMinBias/FullField/P05if_test/2005/073/st_physics_6073006_raw_1020001.event.root
with ssd parameters from new parameters :
in StMySsdMaker::fill , counter of StiTracks = 85
in StMySsdMaker::fill , counter of primaries StiTracks = 37
in StMySsdMaker::fill , counter of StiTracks with ssd,svt,tpc points in primaries =10
in StMySsdMaker::fill , counter of StiTracks with ssd,tpc points in primaries =5
in StMySsdMaker::fill , counter of StiTracks with svt,tpc points in primaries = 14
in StMySsdMaker::fill , counter of StiTracks with tpc points in primaries = 8

with ssd parameters from defaults parameters :
in StMySsdMaker::fill , counter of StiTracks = 84
in StMySsdMaker::fill , counter of primaries StiTracks = 36
in StMySsdMaker::fill , counter of StiTracks with ssd,svt,tpc points in primaries = 8
in StMySsdMaker::fill , counter of StiTracks with ssd,tpc points in primaries =1
in StMySsdMaker::fill , counter of StiTracks with svt,tpc points in primaries = 19
in StMySsdMaker::fill , counter of StiTracks with tpc points in primaries = 8

QA :INFO - StMySsdMaker::summarizeEvent : , Reading Event: 1 Type: NONE Run: 6073006
QA :INFO - # track nodes: 199
QA :INFO - # primary tracks: 99
QA :INFO - # V0 vertices: 0
QA :INFO - # Xi vertices: 0
QA :INFO - # Kink vertices: 0
QA :INFO - # TPC hits: 4180
QA :INFO - # SVT hits: 467
QA :INFO - # SSD hits: 62
QA :INFO - # FTPC hits: 491
QA :INFO - primary vertex: -0.654004 0.295796 -0.403706

Positions

Positions

(Click on the picture to enlarge. A new window will open.)


Picture 1: from the last node including a ssd hit : 15 nodes	Picture 2: ssd hits used in Sti : 62 hits

7 November

Same as previous but in (phi,z)

(Click on the picture to enlarge. A new window will open.)


Picture 1: from the last node including a ssd hit : 15 nodes	Picture 2: ssd hits used in Sti : 62 hits

A better way to look at this event :
red circles : positions from the ssdHitCollection
blue crosses : positions of the nodes in the ssd for the primaries

For a closer look ( 2.5 < phi < 5 and z>0 ) , we have :

(Click on the picture to enlarge. A new window will open.)


Picture 1: 3 nodes ( from the primaries )	Picture 2: ssd hits used in Sti : 7 hits

COMMENTS : the tracking is well using the possible ssd hits but as we have 38 primaries it must be using more than only 16 hits of the ssd

Without and with ssd
for event 167 of /star/data10/reco/cu62productionMinBias/FullField/P05if_test/2005/073/st_physics_6073006_raw_1020001.event.root

Sti without ssd	primaries = 35	tpc alone = 9	tpc+svt = 26
Sti with ssd	primaries = 38	tpc alone = 8	tpc+svt = 14	tpc+ssd = 5	tpc+ssd+svt = 11

9 November

We take a look to another event , with less ssd hits and tracks so the assciation will be more simple to understand
For this event ,we have :
StiStEventFiller::fillEventPrimaries() -I- Primaries (1):21 (2):21 no pipe node:0 with IFC:0
StiStEventFiller::fillEventPrimaries() -I- GOOD:21
in StMySsdMaker::fill , counter of StiTracks = 41
in StMySsdMaker::fill , counter of primaries StiTracks = 21
in StMySsdMaker::fill , counter of ssd points for all StiTracks = 4
in StMySsdMaker::fill , counter of StiTracks with ssd points in primaries = 4
in StMySsdMaker::fill , counter of StiTracks with ssd,svt,tpc points in primaries = 3
in StMySsdMaker::fill , counter of StiTracks with ssd,tpc points in primaries = 1
in StMySsdMaker::fill , counter of StiTracks with svt,tpc points in primaries = 11
in StMySsdMaker::fill , counter of StiTracks with tpc points in primaries = 6
QA :INFO - # track nodes: 97
QA :INFO - # primary tracks: 57
QA :INFO - # V0 vertices: 1
QA :INFO - # Xi vertices: 0
QA :INFO - # Kink vertices: 0
QA :INFO - # TPC hits: 2065
QA :INFO - # SVT hits: 485
QA :INFO - # SSD hits: 23
QA :INFO - # FTPC hits: 199
QA :INFO - primary vertex: -0.603848 0.256902 -6.52965
QA :INFO - QAInfo: Done with Event [no. 52/run 6073006/evt. 164/Date.Time 20050314.82513/sta 0] Real Time = 210.57 seconds Cpu Time = 70.92 seconds


Picture 1: my result : blue circles = hits from the ssdHitCollection red crosses : positions of the nodes	Picture 2: from Lilian : black crosses = ssd hits red crosses : projections of tracks

0 < phi < 3 (rad)

red crosses = tpc hits from the tracks
blue triangles = ssd hits from the tracks (4)(The blue point at ~(0,0) is not a ssd hit in a track , it is the position of the primary vertex !!
purple triangles = ssd hits from the ssdHitCollection
-->There are some good associations ( the only 4 tracks with ssd hits in there ) but we see that there are another ssd hits which must be ( and they're not)associated to the tpc tracks.

Same as before but for 0 < phi < 360 (degre)

This next plot shows the tracks , for this event , only with tpc and svt points.(red are tpc points and green are svt point(s))

-->It is clear that there is tracks which must included a ssd hits and they don't do it.

To illustrate more clearly this , show the next plot :

red : tpc points used for the tracks
blue : svt point(s) used for the tracks
purple : ssd point used for the tracks
green : ssd point from the hitCollection

Sti without ssd	primaries = 21	tpc alone = 6	tpc+svt = 15
Sti with ssd	primaries = 21	tpc alone = 6	tpc+svt = 11	tpc+ssd = 1	tpc+ssd+svt = 3

10 November

By changing the chi2 cut , I get some changes so I check the number of associations by changing it.
The defaults parameters ( seen in StiSsdDetectorBuilder.cxx ) are :
use = 1
active = 1
max chi2 = 5.
min window = 1.5
max window = 4.
search window scale = 10
intrinsicZ = 0.07
driftZ = 0
crossZ = 0
intrinsicX = 0.002
driftX = 0
crossX = 0

chi2	primaries	tpc alone	tpc+svt	tpc+ssd	tpc+svt+ssd
5	21	6	12	0	3
10	21	6	11	1	3
15	21	6	11	1	3
20	21	6	10	1	4
25	21	6	9	2	4
30	21	6	9	2	4
40	21	6	9	2	4
45	21	6	8	2	5
50	21	5	8	3	5
100	21	5	8	3	5

I look at the chi2 of the tracks including a ssd hits for the 2 cases : max chi2 =5 (default) and max chi 2 = 50 :


Picture 1: max chi2 = 5	Picture 2: max chi2 = 50

A max chi2 = 50 for the ssd doesn't have a so big impact for the chi2 of the track.

For chi2 = 50 , I get :


Picture 1: chi2 = 50	Picture 2: from Lilian : black crosses = ssd hits

Things to do
Use of the parameters
Sti/StiKalmanTrackNode.cxx
double StiKalmanTrackNode::getWindowY();
double window = searchWindowScale*::sqrt(mFE._cYY+myEyy);
double StiKalmanTrackNode::getWindowZ();
double window = searchWindowScale*::sqrt(mFE._cZZ+myEzz);

11 November

I check the positions of the hits from the hitCollection ( saved in StEvent ) and the positions of the nodes used in the tracks .
We can see ,several times , in the last plots that there is some little shifts between these 2 positions.
I look , without any transformations , what are these positions : we expect no shifts between hits positions in global frame and the globals positions of the nodes used in the tracks( unless the nodes are NOT the real hits positions but are the projection of the tracks to the ssd layer )
From the hitCollection :
ssd pos X = 12.8051 ssd pos Y = 18.7523 ssd pos Z = -20.4447 ladder =2 , wafer = 3
ssd pos X = 12.6566 ssd pos Y = 18.8367 ssd pos Z = -9.1216 ladder =2 , wafer = 5
ssd pos X = 21.8206 ssd pos Y = 6.72739 ssd pos Z = 6.59799 ladder =4 , wafer = 9
ssd pos X = 23.1171 ssd pos Y = -3.92153 ssd pos Z = 13.9644 ladder =6 , wafer = 11
ssd pos X = 23.1331 ssd pos Y = -3.88704 ssd pos Z = 16.1356 ladder =6 , wafer = 11
ssd pos X = 22.095 ssd pos Y = -6.13356 ssd pos Z = 26.593 ladder =6 , wafer = 14
ssd pos X = 22.1 ssd pos Y = -6.12286 ssd pos Z = 27.2661 ladder =6 , wafer = 14
ssd pos X = 22.091 ssd pos Y = -6.14218 ssd pos Z = 30.3502 ladder =6 , wafer = 15
ssd pos X = 22.1189 ssd pos Y = -6.08181 ssd pos Z = 34.1498 ladder =6 , wafer = 15
ssd pos X = 19.1959 ssd pos Y = -12.5577 ssd pos Z = -9.5172 ladder =7 , wafer = 5
ssd pos X = 10.6094 ssd pos Y = -19.9997 ssd pos Z = -14.936 ladder =8 , wafer = 4
ssd pos X = 5.51539 ssd pos Y = -22.0053 ssd pos Z = 17.2095 ladder =9 , wafer = 12
ssd pos X = -7.66417 ssd pos Y = -21.5368 ssd pos Z = 27.8943 ladder =11 , wafer = 14
ssd pos X = -16.5239 ssd pos Y = -16.6397 ssd pos Z = -14.6103 ladder =12 , wafer = 4
ssd pos X = -19.2966 ssd pos Y = -12.4482 ssd pos Z = -28.9249 ladder =13 , wafer = 1
ssd pos X = -17.6322 ssd pos Y = -14.2582 ssd pos Z = -19.6214 ladder =13 , wafer = 3
ssd pos X = -19.2351 ssd pos Y = -12.5151 ssd pos Z = 18.9332 ladder =13 , wafer = 12
ssd pos X = -22.5 ssd pos Y = 0.522744 ssd pos Z = 4.5502 ladder =15 , wafer = 9
ssd pos X = -18.6988 ssd pos Y = 13.0983 ssd pos Z = -19.1209 ladder =17 , wafer = 3
ssd pos X = -19.391 ssd pos Y = 12.3455 ssd pos Z = -18.2249 ladder =17 , wafer = 3
ssd pos X = -18.5536 ssd pos Y = 13.2561 ssd pos Z = 18.8884 ladder =17 , wafer = 12
ssd pos X = -18.5601 ssd pos Y = 13.2492 ssd pos Z = 19.4312 ladder =17 , wafer = 12
ssd pos X = -17.0097 ssd pos Y = 14.9352 ssd pos Z = 23.7898 ladder =17 , wafer = 13

From the Nodes :
his position from nodes in global frame are X =-18.733250 , Y = 13.050172 , Z = -19.139466
Hit det?: Ssd/Layer_0/Ladder_18/Wafers

his position from nodes in global frame are X =-19.505014 , Y = 12.186285 , Z = -18.240529
Detector: Ssd/Layer_0/Ladder_18/Wafers

his position from nodes in global frame are X =-17.935170 , Y = -13.835025 , Z = -19.647880
Detector: Ssd/Layer_0/Ladder_14/Wafers

his position from nodes in global frame are X =-19.478441 , Y = -12.175165 , Z = 18.920508
Detector: Ssd/Layer_0/Ladder_14/Wafers

his position from nodes in global frame are X =10.091873 , Y = -20.905036 , Z = -15.702535
Detector: Ssd/Layer_0/Ladder_9/Wafers

his position from nodes in global frame are X =-19.500963 , Y = -12.162728 , Z = -28.916233
Detector: Ssd/Layer_0/Ladder_14/Wafers

his position from nodes in global frame are X =-17.024572 , Y = 14.914395 , Z = 23.783445
Detector: Ssd/Layer_0/Ladder_18/Wafers

his position from nodes in global frame are X =12.698761 , Y = 18.805343 , Z = -9.205307
Detector: Ssd/Layer_0/Ladder_3/Wafers

Without transformation ( chi2 =50 )
in StMySsdMaker::fill , counter of StiTracks = 41
in StMySsdMaker::fill , counter of primaries StiTracks = 21
in StMySsdMaker::fill , counter of ssd points for all StiTracks = 9
in StMySsdMaker::fill , counter of StiTracks with ssd points in primaries = 8
in StMySsdMaker::fill , counter of StiTracks with ssd,svt,tpc points in primaries = 5
in StMySsdMaker::fill , counter of StiTracks with ssd,tpc points in primaries = 3
in StMySsdMaker::fill , counter of StiTracks with svt,tpc points in primaries = 8
in StMySsdMaker::fill , counter of StiTracks with tpc points in primaries = 5

15 November

To illustrate the event above , see this plot : green : used in the tracks
blue : from the ssdHitCollection

It was done without change the geometry of the ssd , eg the transformations : Global transformations

On 10 events :

tpchits >	primaries	tracks with ssd hits	chi2	rms of chi2
0	1058	411	0.884	0.25
10	959	361	0.9	0.236
15	821	319	0.907	0.227
20	753	289	0.912	0.226
25	696	269	0.915	0.224
30	617	240	0.918	0.218
35	455	178	0.922	0.205


Picture 1: tpc hits > 10	Picture 2: tpc hits > 35

Return on event 167
config 1 = defaults parameters + no geometry transformations + no hits transformations
config 2 = defaults parameters + geometry transformations + no hits transformations
config 3 = defaults parameters + geometry transformations + hits transformations
config 4 = new parameters( intrinsicZ = 0.07 ) + geometry transformations + hits transformations
config 5 = new parameters( chi2 = 50 ) + geometry transformations + hits transformations

config	primaries	tpc alone	tpc+svt	tpc+ssd	tpc+svt+ssd
1	37	10	23	0	4
2	37	10	22	1	4
3	36	8	19	1	8
4	37	8	16	2	11
5	38	6	13	7	12

10 events again


Picture 1: config 1	Picture 2: tuned parameters + no geometry transformation	Picture 3: config 5

beginning on self alignment for svt+ssd : - self alignment for svt+ssd

Jonathan Bouchet

Last modified: 3 november

chi2	primaries	tpc alone	tpc+svt	tpc+ssd	tpc+svt+ssd
5	21	6	12	0	3
10	21	6	11	1	3
15	21	6	11	1	3
20	21	6	10	1	4
25	21	6	9	2	4
30	21	6	9	2	4
40	21	6	9	2	4
45	21	6	8	2	5
50	21	5	8	3	5
100	21	5	8	3	5

chi2	primaries	tpc alone	tpc+svt	tpc+ssd	tpc+svt+ssd
5	21	6	12	0	3
10	21	6	11	1	3
15	21	6	11	1	3
20	21	6	10	1	4
25	21	6	9	2	4
30	21	6	9	2	4
40	21	6	9	2	4
45	21	6	8	2	5
50	21	5	8	3	5
100	21	5	8	3	5

chi2	primaries	tpc alone	tpc+svt	tpc+ssd	tpc+svt+ssd
5	21	6	12	0	3
10	21	6	11	1	3
15	21	6	11	1	3
20	21	6	10	1	4
25	21	6	9	2	4
30	21	6	9	2	4
40	21	6	9	2	4
45	21	6	8	2	5
50	21	5	8	3	5
100	21	5	8	3	5