Top Banner
Si-W ECAL Tests en faisceau du prototype Cristina Cârloganu Clermont Ferrand
27

Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

Apr 08, 2018

Download

Documents

HoàngNhi
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

Si-W ECAL Tests en faisceau du

prototype

Cristina Cârloganu

Clermont Ferrand

Page 2: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

!"#$%&'(&)*+,*&-./0&1''( Djamel BOUMEDIENE, CALICE collaboration 2312

!"#$%&'()()*+,

#-.('-,'%/%!"#$%!&#

#0)12,%,3,4,5)%/%%'(')*#

6178%.94+3157%/%+,-(./&0%

6178%7'95:39'1)*%/%121-)34)&((%

"(4+90)%/%5-6,-)3-7&8!9-:*0-

6;-<,

"8955,3.%/ =;>1-?6,,=@

#%8178%7'95:39'1)*%093('14,),'%(+)141;,<%=('%Particle Flow =('%>$"%+8*.10.

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Le prototype ECAL en faisceau test

2

HCAL Tail CatcherSi-W ECAL

CERN H6juin 2007

Faisceau

Page 3: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

!"#$%&'(&)*+,*&-./0&1''( Djamel BOUMEDIENE, CALICE collaboration 2312

!"#$%&'()()*+,

#-.('-,'%/%!"#$%!&#

#0)12,%,3,4,5)%/%%'(')*#

6178%.94+3157%/%+,-(./&0%

6178%7'95:39'1)*%/%121-)34)&((%

"(4+90)%/%5-6,-)3-7&8!9-:*0-

6;-<,

"8955,3.%/ =;>1-?6,,=@

#%8178%7'95:39'1)*%093('14,),'%(+)141;,<%=('%Particle Flow =('%>$"%+8*.10.

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Le prototype ECAL en faisceau test

2

HCAL Tail CatcherSi-W ECAL

CERN H6juin 2007

Faisceau

Page 4: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

!"#$%&'(&)*+,*&-./0&1''( Djamel BOUMEDIENE, CALICE collaboration 2312

!"#$%&'()()*+,

#-.('-,'%/%!"#$%!&#

#0)12,%,3,4,5)%/%%'(')*#

6178%.94+3157%/%+,-(./&0%

6178%7'95:39'1)*%/%121-)34)&((%

"(4+90)%/%5-6,-)3-7&8!9-:*0-

6;-<,

"8955,3.%/ =;>1-?6,,=@

#%8178%7'95:39'1)*%093('14,),'%(+)141;,<%=('%Particle Flow =('%>$"%+8*.10.

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Le prototype ECAL en faisceau test

2

HCAL Tail CatcherSi-W ECAL

CERN H6juin 2007

Faisceau

Page 5: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Le prototype ECAL SiW en faisceau test

3

mai 06

aout -sept06

DESY, 24 couches, 6 wafers/couche

CERN, 30 couches, 6 wafers/couche

CERN, 30 couches, 6 wafers/coucheoct 06

CERN, 30 couches, 9 wafers/couchejuillet -aout07

Page 6: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Le prototype ECAL SiW en faisceau test

3

mai 06

aout -sept06

DESY, 24 couches, 6 wafers/couche

CERN, 30 couches, 6 wafers/couche

CERN, 30 couches, 6 wafers/coucheoct 06

CERN, 30 couches, 9 wafers/couchejuillet -aout07

e- 1-6 GeV, 5 angles, 3 positions, 8 M triggers

ECAL+AHCAL: pi 30-80 GeV, 3 angles, 1.7 M triggersECAL: e- 10-45 GeV, 4 angles, 8.6 M triggers 30 M muons pour la calibration.

ECAL+AHCAL+TCMT: e+, e-: 3.8 M triggers, 6-45 GeV π+,π-: 22 M triggers, 6-80 GeV 40 M muons pour la calibration.

ECAL, HCAL, TCMT: 4 angles, 12 positions e+, e-,π+,π-: 200 M triggers, 6-80 GeV

Page 7: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Le prototype ECAL SiW en faisceau test

3

mai 06

aout -sept06

DESY, 24 couches, 6 wafers/couche

CERN, 30 couches, 6 wafers/couche

CERN, 30 couches, 6 wafers/coucheoct 06

CERN, 30 couches, 9 wafers/couchejuillet -aout07

e- 1-6 GeV, 5 angles, 3 positions, 8 M triggers

ECAL+AHCAL: pi 30-80 GeV, 3 angles, 1.7 M triggersECAL: e- 10-45 GeV, 4 angles, 8.6 M triggers 30 M muons pour la calibration.

ECAL+AHCAL+TCMT: e+, e-: 3.8 M triggers, 6-45 GeV π+,π-: 22 M triggers, 6-80 GeV 40 M muons pour la calibration.

ECAL, HCAL, TCMT: 4 angles, 12 positions e+, e-,π+,π-: 200 M triggers, 6-80 GeV

!"#$%

&'()*+),

-.)*&&'

/-012324

)5-67-8419

()%:;#

'

Summary of the data taken

Summary of the data taken

5239)4<)=2>?@)A

)B&)?CD978

!EF,

)979<8>)

G)*HF)IC

)J41)!K

%L)MN.

>2O>)1P<

>

!Q)+&

),)G)R)IC)J4

1)SP4<

)O-6201-

824<)1P<

>

!"#"$%&

'()*+,

-.%+/(

)&'()*+,

0('%1&/)

(23

! .24*/1&)"5$

! .24*/1

Page 8: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Management des données

4

Stockage et processing des données sur la grille (VO CALICE hébergée par DESY)

Page 9: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Management des données

4

!!!"#$%&'%()*+,-%.,/0%1&&'

!"#$%&'()*+$,'-*.&/*(&0.$1$20$3*+&3#

40/(#5$67$89:;<=*-#$>0'$'#-&/('*(&0.$&/$"((?/<@@-'&5120A/B5#/7B5#<CDDE@20A/@3*+&3#

VO Manager: R.P./ LAL, Deputy: A. Gellrich/ DESY

FE$G#A6#'/$$$*.5$30).(&.-$BB

C

8*(*$A*.*-#A#.($*.5$?'03#//&.-$67$)/&.-$("#$-'&5$

Stockage et processing des données sur la grille (VO CALICE hébergée par DESY)

Page 10: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Management des données

4

!!!"#$%&'%()*+,-%.,/0%1&&'

!"#$%&'()*+$,'-*.&/*(&0.$1$20$3*+&3#

40/(#5$67$89:;<=*-#$>0'$'#-&/('*(&0.$&/$"((?/<@@-'&5120A/B5#/7B5#<CDDE@20A/@3*+&3#

VO Manager: R.P./ LAL, Deputy: A. Gellrich/ DESY

FE$G#A6#'/$$$*.5$30).(&.-$BB

C

8*(*$A*.*-#A#.($*.5$?'03#//&.-$67$)/&.-$("#$-'&5$

Stockage et processing des données sur la grille (VO CALICE hébergée par DESY)

Simulations Monte Carlo avec Mokka (basé sur Geant4), même programme utilisé pour ILD

Production et stockage centralisés, sur la grille...

Page 11: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Le prototype ECAL SiW en faisceau test

5

Excellent fonctionnement du détecteur - très grande stabilité de l’électronique (monitorée en temps réel et calibrée avec des MIPs)

Sur l’ensemble des voies:

• 98,6% des voies fonctionnelles • bruit moyen par voie : 0.13 MIPs • dispersion du bruit voie à voie 0.012 MIPs

Page 12: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

La sélection des électrons

6

125 < Eraw / Ebeam < 375

réjection des pions avec un détecteur Cerenkov en amont

réjection du halo du faisceau

réjection des électrons qui rayonnent avant le ECAL

Page 13: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

impinging on the calorimeter at normal incidence. Here the mean value of Eraw(Eq. 5.1) is plottedas a function of the shower barycentre (x, y), defined as :

(x, y) =!i

(Eixi,Eiyi)/!i

Ei

The sums run over all hit cells in the calorimeter. Dips in response corresponding to the guardring positions are clearly visible. According to Figure 6, which shows the mean value of E raw as afunction of the shower barycentre for 20 GeV electrons, the energy loss is about 15 % when tracksimpinge in the centre of the x gaps and about 20 % in the case of the y gap.

CALICE 2006 data

!50 !40 !30 !20 !10 0 10 20 30 40

Y (

mm

)

!30

!20

!10

10

20

0

2

4

6

8

10

12

14

16

X (mm)

30

0

!40

Figure 4. Mean values of Eraw for a 15 GeV e− beam as function of the barycentre coordinates.

In order to recover this loss and to have a more uniform calorimeter response, a simple methodwas investigated. The ECAL energy response, f (x, y) = Eraw/Ebeam, is measured using a com-bined sample of 10, 15 and 20 GeV electrons, equally populated and the energy of each shower iscorrected by 1/ f .The response function f is displayed on Figure 5. It can be parameterised with Gaussian

functions, independently in x and y:

f (x, y) =

(

1!axe!

(x!xgap)2

2!2x

)(

1!aye!

(y!ygap)2

2!2y

)

The results of the Gaussian parametrisations are given in Table 1. The gap in x is shallower andwider than that in y, due to the staggering of the gaps in x [2].As illustrated in Figure 6, when the gap corrections are applied, the energy loss in the gaps is

reduced to a few percent level. The low energy tail in the energy distribution is also much reduced

– 6 –

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

L’uniformité du détecteur

7

Page 14: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

0.70.7

0.750.75

0.80.8

0.850.85

0.90.9

0.950.95

11

1.051.05

1.11.1

!40 !30 !20 !10 0 10 20 !3030 40 !25 !20 !15

(mm)

!10

x

x x

!5 0 5 10 15 20

(mm)y

y y

f( , )

f( , )

CALICE 2006 data CALICE 2006 data

Figure 5. f (x, y) function of the shower barycentre coordinates, for a combined sample of 10, 15 and 20 GeVelectrons. To characterise the x (y) response, the events were requested to be outside the inter-wafer gap iny (x), leading to an important difference in the number of events for the two distributions, since the beam iscentred on the y gap.

position (mm) ! (mm) a

x direction -30.01 4.3 0.143y direction -8.4 3.19 0.198

Table 1. Gaussian parametrisation of the inter-wafer gaps.

(mm)x

!50 !40 !30 !20 !10 0 10 20 30 40 50

4000

4200

4400

4600

4800

5000

5200

5400

5600

raw data

corrected data

(mm)y

!40 !30 !20 !10 0 10 20 30

(M

IPs

)

(M

IPs

)

raw

raw

EE

4000

4200

4400

4600

4800

5000

5200

5400

5600

raw data

corrected data

CALICE 2006 data CALICE 2006 data

Figure 6. Mean Eraw function of the shower barycentre coordinates for 20 GeV electrons, before (blacktriangles) and after the corrections (blue circles) were applied on Eraw.

(Figure 7). The correction method relies only on calorimetric information and can be applied bothfor photons and electrons.Even if it is possible to correct, statistically, for the interwafer gaps, per event their presence

– 7 –

impinging on the calorimeter at normal incidence. Here the mean value of Eraw(Eq. 5.1) is plottedas a function of the shower barycentre (x, y), defined as :

(x, y) =!i

(Eixi,Eiyi)/!i

Ei

The sums run over all hit cells in the calorimeter. Dips in response corresponding to the guardring positions are clearly visible. According to Figure 6, which shows the mean value of E raw as afunction of the shower barycentre for 20 GeV electrons, the energy loss is about 15 % when tracksimpinge in the centre of the x gaps and about 20 % in the case of the y gap.

CALICE 2006 data

!50 !40 !30 !20 !10 0 10 20 30 40

Y (

mm

)

!30

!20

!10

10

20

0

2

4

6

8

10

12

14

16

X (mm)

30

0

!40

Figure 4. Mean values of Eraw for a 15 GeV e− beam as function of the barycentre coordinates.

In order to recover this loss and to have a more uniform calorimeter response, a simple methodwas investigated. The ECAL energy response, f (x, y) = Eraw/Ebeam, is measured using a com-bined sample of 10, 15 and 20 GeV electrons, equally populated and the energy of each shower iscorrected by 1/ f .The response function f is displayed on Figure 5. It can be parameterised with Gaussian

functions, independently in x and y:

f (x, y) =

!

1−axe!

(x−xgap)2

2!2x

"!

1−aye!

(y−ygap)2

2!2y

"

The results of the Gaussian parametrisations are given in Table 1. The gap in x is shallower andwider than that in y, due to the staggering of the gaps in x [2].As illustrated in Figure 6, when the gap corrections are applied, the energy loss in the gaps is

reduced to a few percent level. The low energy tail in the energy distribution is also much reduced

– 6 –

0.70.7

0.750.75

0.80.8

0.850.85

0.90.9

0.950.95

11

1.051.05

1.11.1

!40 !30 !20 !10 0 10 20 !3030 40 !25 !20 !15

(mm)

!10

xx x

!5 0 5 10 15 20

(mm)yy y

f(

,

)

f(

, )

CALICE 2006 data CALICE 2006 data

Figure 5. f (x, y) function of the shower barycentre coordinates, for a combined sample of 10, 15 and 20 GeVelectrons. To characterise the x (y) response, the events were requested to be outside the inter-wafer gap iny (x), leading to an important difference in the number of events for the two distributions, since the beam iscentred on the y gap.

position (mm) ! (mm) a

x direction -30.01 4.3 0.143y direction -8.4 3.19 0.198

Table 1. Gaussian parametrisation of the inter-wafer gaps.

(mm)x

!50 !40 !30 !20 !10 0 10 20 30 40 50

4000

4200

4400

4600

4800

5000

5200

5400

5600

raw data

corrected data

(mm)y

!40 !30 !20 !10 0 10 20 30

(M

IPs)

(M

IPs)

raw

raw

EE

4000

4200

4400

4600

4800

5000

5200

5400

5600

raw data

corrected data

CALICE 2006 data CALICE 2006 data

Figure 6. Mean Eraw function of the shower barycentre coordinates for 20 GeV electrons, before (blacktriangles) and after the corrections (blue circles) were applied on Eraw.

(Figure 7). The correction method relies only on calorimetric information and can be applied bothfor photons and electrons.Even if it is possible to correct, statistically, for the interwafer gaps, per event their presence

– 7 –

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Correction des pertes dans les zones inactives

8

Page 15: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

Energy (MIPs)

100 200 300 400 500 600 700 800 9000

200

400

600

800

1000

1200

1400

1600

1590

Energy in even layers

Mean 273

Constant 1507 ± 12.0

Mean 256 ± 1.1

Sigma 93.41 ± 1.11

Energy in odd layers

Mean 253.7

Constant ± 12.8

Mean 233.6 ± 1.1

Sigma 86.14 ± 1.06

CALICE 2006 data

Figure 10. Energy deposits in odd end even layers by 20 GeV electrons. The statistical information corre-sponds to Gaussian parametrisations of the two distributions.

(GeV)beamE

10 15 20 25 30 35 40 45

!

0

0.02

0.04

0.06

0.08

0.1

CALICE 2006 data

Figure 11. ! as function of the beam energy.

6.2 Linearity and energy resolution

The total response of the calorimeter is calculated as

Erec(MIPs) =!i

wiEi

– 11 –

8.3

(sl

ab d

epth

)

0.2

(to

lera

nce

)

20.1

50.1

5

1.4 PCB

W

PCB

1.8

0.15 (tolerance)0.3 (H carbon)

0.4 (carbon structure)

0.1

(A

1)

2.1

(P

CB

)

0.1

(glu

e)

0.5

2 (

Si)

Si

Si

Figure 9. Details of one ECAL slab, showing one passive tungsten layer sandwiched between two activesilicon layers. The lower silicon layer is preceded by a larger passive layer compared to the upper one: thePCB, aluminium , glue, etc add to the tungsten.

The easiest method to investigate this difference is to compare in each stack the mean energydeposits in odd and even layers. For the first stack, if we neglect the shower profile, the ratio of thetwo is

R=Eodd

Eeven= 1+! ,

with ! being, approximately, the ratio of the non-tungsten radiation length to the tungsten radiationlength.When counting the layers starting from zero, the odd layers are systematically shifted com-

pared to the even layers towards the shower maximum and the measurement of R is biased by theshower development. To overcome this bias, R is measured twice, either comparing the odd layerswith the average of the surrounding even layers, or comparing the even layers with the average ofthe neighbouring odd layers:

R! =

!E1+E3+E5+E7"!

E0+E22 + E2+E4

2 + E4+E62 + E6+E8

2

"

R!! =

!

E1+E32 + E3+E5

2 + E5+E72 + E7+E9

2

"

!E2+E4+E6+E8"where En is the energy deposit in the layer number n and the brackets indicate that mean values areused. The value of ! is taken as the average of R!#1 and R!!#1, whereas the difference betweenthem gives a conservative estimate of the systematic error due to the shower shape. As an example,the distribution of the energy deposits in the odd and even layers is shown in Figure 10 for 20 GeVelectrons.The overall value of ! is 7.2±0.2±1.7% , while the individual values of ! , obtained for each

beam energy, are displayed in Figure 11. The measurement of ! using the second and third stack,as well as simulated data, leads to compatible results.In computing the total response of the calorimeter, the sampling fraction for layer i is given by

wi = K = 1, 2, 3 for even layers in stacks 1, 2, 3 and wi = K+! for the odd layers, respectively.

– 10 –

Energy (MIPs)

100 200 300 400 500 600 700 800 9000

200

400

600

800

1000

1200

1400

1600

1590

Energy in even layers

Mean 273

Constant 1507 ± 12.0

Mean 256 ± 1.1

Sigma 93.41 ± 1.11

Energy in odd layers

Mean 253.7

Constant ± 12.8

Mean 233.6 ± 1.1

Sigma 86.14 ± 1.06

CALICE 2006 data

Figure 10. Energy deposits in odd end even layers by 20 GeV electrons. The statistical information corre-sponds to Gaussian parametrisations of the two distributions.

(GeV)beamE

10 15 20 25 30 35 40 45

!

0

0.02

0.04

0.06

0.08

0.1

CALICE 2006 data

Figure 11. ! as function of the beam energy.

6.2 Linearity and energy resolution

The total response of the calorimeter is calculated as

Erec(MIPs) =!i

wiEi

– 11 –

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Facteurs d’échantillonage

9

η = différence entre les facteurs d’échantillonagedes couches paires et impaires

Page 16: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

(MIPs)recE

Entries 36892 / ndf 2! 23.45 / 19

Constant 15.1± 2026 Mean 2.6± 7924 Sigma 2.5± 259

7000 7500 8000 8500 9000 95000

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

Entries 36892 / ndf 2! 23.45 / 19

Constant 15.1± 2026 Mean 2.6± 7924 Sigma 2.5± 259

dataMonte Carlo

CALICE 2006 data

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Résolution et linéarité

10

(GeV)beamE

5 10 15 20 25 30 35 40 45

(M

IPs

)m

ea

nE

2000

4000

6000

8000

10000

12000

14000

16000 / ndf 2! 19.07 / 33

Prob 0.9747

p0 10.85± !97.48

p1 0.4739± 266.3

/ ndf 2! 19.07 / 33

Prob 0.9747

10.85± !97.48

0.4739± 266.3

"#

CALICE 2006 data

Figure 13. Energy response of the ECAL function of the beam energy.

thresholdhitE

0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9

(M

IPs )

" !

70

80

90

100

110

120

130

140

CALICE 2006 data

Figure 14. Variation of the linearity offset with the hit energy threshold.

a quadrature sum of stochastic and constant terms221

!EmeasEmeas

=16.69±0.13!

E(GeV)! (1.09±0.06)% ,

where the intrinsic momentum spread of the beam was subtracted from the ECAL data [6].222

– 13 –

Emeas = Emean + α

Emean

Page 17: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

(GeV) beam E1/0.15 0.2 0.25 0.3 0.35 0.4

( %

)m

eas

/ E

mea

s E

!2

3

4

5

6

7

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Résolution et linéarité

111

ΔE 16.7 ± 0.1 E

(%) = √ E (GeV)

⊕ (1.1 ± 0.1)

ΔE 17.2 ± 0.3 E

(%) = √ E (GeV)

⊕ (0.8 ± 0.2)

simulations MCdonnées CALICE 2006

6 GeV

45 GeV

(GeV)beamE

5 10 15 20 25 30 35 40 45

(G

eV

)m

ea

sR

es

idu

als

E

!0.5

!0.4

!0.3

!0.2

!0.1

0

0.1

0.2

0.3

0.4

0.5CALICE 2006 data

Figure 15. Residuals to linearity of Emeas function of the beam energy. All the runs around the same nominalenergy of the beam were combined in one entry, for which the uncertainty was estimated assuming that theuncertainties on the individual runs were uncorrelated.

Still to be done: The relative energy resolution expected from Monte Carlo simulations is also223

shown in Figure 16 . The agreement with data is quite good, with the measured resolution typically224

worse than the expected by a factor ! 1.02.225

Different systematic checks have been performed on the data. Variations of the linearity and226

resolution against the minimal accepted distance between the shower barycentre and the nearest227

inter-wafer gap, when the energy threshold for considering the hits is 0.5MIPs are shown in Table 3.228

In addition, this hit energy threshold has been itself varied (Table 4). In order to investigate the229

potential effects linked to the beam position, the energy response is also compared for showers230

with barycentres located in the right hand side and in the upper half of the detector (Table 5),231

respectively. The results are consistent. Since data were taken in both August and October 2006, it232

was also possible to check the response stability in time and no significant differences between the233

two data samples are observed.234

7. Conclusion235

The response to normally incident electrons of the Calice Si-W electromagnetic calorimeter was236

measured for energies between 6 and 45 GeV, using the data recorded during 2006 testbeam at237

CERN.238

The calorimeter is linear to 1% level. The energy resolution has a stochastic term of 16.69±0.13,239

whereas the constant term is 1.09±0.06.240

A simple method of correcting for non-uniformities in the calorimeter response due to non241

active regions between the silicon wafers was found. It allows to recover, statistically, most of the242

– 14 –

Page 18: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

19/03/2008 D. BOUMEDIENE | CALICE meeting, ANL 6

!"#$%&#'(')*# +,'-#. /0$%&# 12 0345'+6)*'+,,6

Y Gaps

Entries 0Mean -145.3RMS 35.21

-200 -180 -160 -140 -120 -100 -80

-0.6

-0.4

-0.2

0

0.2

0.4

Y Gaps

Entries 0Mean -145.3RMS 35.21

Y Gaps position

After subtraction of the direction : layers alignment

Systematic shift in a slab The 9th slab : singularity

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Les zones mortes utiles pour aligner le détecteur

121

• les défauts d’alignement observés en y inférieurs à 0.5 mm• troisième stack déplacé de 1 mm par rapport aux deux premiers

Page 19: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

(hit/mip < 2) (2 < hit/mip < 10) (10 < hit/mip < 50) (50 < hit/mip)

(mm) shower-XhitX-50 0 50

/1m

m (

au)

meas

E

0

0.2

0.4

0.6

0.8

1

(mm) shower-XhitX-50 0 50

(mm) shower-XhitX-50 0 50

(mm) shower-XhitX-50 0 50

(mm) shower-YhitY-50 0 50

/1m

m (

au)

meas

E

0

0.2

0.4

0.6

0.8

1

(mm) shower-YhitY-50 0 50

(mm) shower-YhitY-50 0 50

(mm) shower-YhitY-50 0 50

incident energy (GeV)0 10 20 30 40 50 60

sh

ow

er

wid

th (

mm

)

0

5

10

15

20

25

30

35

40

45

50

hit < 2 mip

2 mip < hit < 10 mip

10 mip < hit < 50 mip

50 mip < hit

incident energy (GeV)0 10 20 30 40 50 60

rad

ius (

mm

)

15

20

25

30

35

(signal=measured energy)

90% signal containment

95% signal containment (a)

incident energy (GeV)0 10 20 30 40 50 60

rad

ius (

mm

)

15

20

25

30

35

(signal=reconstructed energy)

90% signal containment

95% signal containment (b)

impact pointcenter edge cornercenter edge corner

radiu

s (

mm

)

15

20

25

30

35

90% signal containment

95% signal containment

6 GeV-e

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Dévelopement latéral des gerbes ém

1313

Page 20: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

(hit/mip < 2) (2 < hit/mip < 10) (10 < hit/mip < 50) (50 < hit/mip)

(mm) shower-XhitX-50 0 50

/1m

m (

au)

me

as

E

0

0.2

0.4

0.6

0.8

1

(mm) shower-XhitX-50 0 50

(mm) shower-XhitX-50 0 50

(mm) shower-XhitX-50 0 50

(mm) shower-YhitY-50 0 50

/1m

m (

au)

me

as

E0

0.2

0.4

0.6

0.8

1

(mm) shower-YhitY-50 0 50

(mm) shower-YhitY-50 0 50

(mm) shower-YhitY-50 0 50

incident energy (GeV)0 10 20 30 40 50 60

sh

ow

er

wid

th (

mm

)

0

5

10

15

20

25

30

35

40

45

50

hit < 2 mip

2 mip < hit < 10 mip

10 mip < hit < 50 mip

50 mip < hit

incident energy (GeV)0 10 20 30 40 50 60

rad

ius (

mm

)

15

20

25

30

35

(signal=measured energy)

90% signal containment

95% signal containment (a)

incident energy (GeV)0 10 20 30 40 50 60

rad

ius (

mm

)

15

20

25

30

35

(signal=reconstructed energy)

90% signal containment

95% signal containment (b)

impact pointcenter edge cornercenter edge corner

radiu

s (

mm

)

15

20

25

30

35

90% signal containment

95% signal containment

6 GeV-e

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Dévelopement latéral des gerbes em

1413

(hit/mip < 2) (2 < hit/mip < 10) (10 < hit/mip < 50) (50 < hit/mip)

(mm) shower-XhitX-50 0 50

/1m

m (

au)

me

as

E

0

0.2

0.4

0.6

0.8

1

(mm) shower-XhitX-50 0 50

(mm) shower-XhitX-50 0 50

(mm) shower-XhitX-50 0 50

(mm) shower-YhitY-50 0 50

/1m

m (

au)

me

as

E

0

0.2

0.4

0.6

0.8

1

(mm) shower-YhitY-50 0 50

(mm) shower-YhitY-50 0 50

(mm) shower-YhitY-50 0 50

incident energy (GeV)0 10 20 30 40 50 60

sho

wer

wid

th (

mm

)

0

5

10

15

20

25

30

35

40

45

50

hit < 2 mip

2 mip < hit < 10 mip

10 mip < hit < 50 mip

50 mip < hit

incident energy (GeV)0 10 20 30 40 50 60

rad

ius (

mm

)

15

20

25

30

35

(signal=measured energy)

90% signal containment

95% signal containment (a)

incident energy (GeV)0 10 20 30 40 50 60

rad

ius (

mm

)

15

20

25

30

35

(signal=reconstructed energy)

90% signal containment

95% signal containment (b)

impact pointcenter edge cornercenter edge corner

radiu

s (

mm

)15

20

25

30

35

90% signal containment

95% signal containment

6 GeV-e

Page 21: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Developement longitudinal des gerbes elm

1514

Page 22: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Première étude des pions avec ECAL

1614

17

!"#$%%&'&()*(+,-"./0,1#*

21#3,0"4,#/.(5#6738(9,*07,:"0,1#

;<+= ;<+=>?@A ;<+=>5BC

;<+=>?5!D ;<+=>?5!D>E= ;<+=>E=

Page 23: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

18

!"#$%%&'&()*(+,-"./0,1#*

21#3,0"4,#/.(5#6738(9,*07,:"0,1#

2;<=8* >?+; >?+;@25A9

BCB; 2D5< 2D5<@E5!C

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Première étude des pions avec ECAL

1714

Page 24: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

!"#$%&'()*+),-.)*&&' /-012324)5-67-8419()%:;# *<

The 2008 test beam at FNALThe 2008 test beam at FNAL

52=>?-6 -@A)":!"#)-619-A.)2@B8-669A)-8),CD/

E-8-)8-F2@G)B8-189A)4@)HB8 4I),-.)J

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Cette année, FNAL !

1814

!"#$%&'()*+),-.)*&&' /-012324)5-67-8419()%:;# *<

The 2008 test beam at FNALThe 2008 test beam at FNAL

52=>?-6 -@A)":!"#)-619-A.)2@B8-669A)-8),CD/

E-8-)8-F2@G)B8-189A)4@)HB8 4I),-.)J

!"#$%&'()*+),-.)*&&' /-012324)5-67-8419()%:;# *<

Plans for the 2008 test beamPlans for the 2008 test beam

&=)>&()*&()?&)@9A&()>&()*&()?&)@9A"BA69C

D*<,)979B8C)E)'&)F9G)HB-114I)-B@)014-@)09-JK

L,)979B8C)E)>*&)F9G)H014-@)09-JK!-6201-824B)I28M)

! 979B8C

L)84)>'&)F9GL()>&()>*()><()*&()?&()N&()L&)F9GOB91A.)P42B8C

"!Q9! QP"!Q9!R-182S69)8.P9C

"SM2979@)-8)*&&+)!O%T)UVR14P4C9@)P6-B)W41)>C8 P9124@

X UI4)89C8)09-J)P9124@C)E,UV/

! >Y*+),-.)Z)'Y*[)\]6.)*&&')^ 52_O!"#`":!"#

! >Y*?)59P89J091)*&&')^ 5S2_O!"#`":!"#

X R6-B)84)89C8)O!"#`a:!"#)P14848.P9C)2B)CP12BA)*&&[

!"#$%&'()*+),-.)*&&' /-012324)5-67-8419()%:;# *<

Plans for the 2008 test beamPlans for the 2008 test beam

&=)>&()*&()?&)@9A&()>&()*&()?&)@9A"BA69C

D*<,)979B8C)E)'&)F9G)HB-114I)-B@)014-@)09-JK

L,)979B8C)E)>*&)F9G)H014-@)09-JK!-6201-824B)I28M)

! 979B8C

L)84)>'&)F9GL()>&()>*()><()*&()?&()N&()L&)F9GOB91A.)P42B8C

"!Q9! QP"!Q9!R-182S69)8.P9C

"SM2979@)-8)*&&+)!O%T)UVR14P4C9@)P6-B)W41)>C8 P9124@

X UI4)89C8)09-J)P9124@C)E,UV/

! >Y*+),-.)Z)'Y*[)\]6.)*&&')^ 52_O!"#`":!"#

! >Y*?)59P89J091)*&&')^ 5S2_O!"#`":!"#

X R6-B)84)89C8)O!"#`a:!"#)P14848.P9C)2B)CP12BA)*&&[

!"#$%&'()*+),-.)*&&' /-012324)5-67-8419()%:;# *<

Plans for the 2008 test beamPlans for the 2008 test beam

&=)>&()*&()?&)@9A&()>&()*&()?&)@9A"BA69C

D*<,)979B8C)E)'&)F9G)HB-114I)-B@)014-@)09-JK

L,)979B8C)E)>*&)F9G)H014-@)09-JK!-6201-824B)I28M)

! 979B8C

L)84)>'&)F9GL()>&()>*()><()*&()?&()N&()L&)F9GOB91A.)P42B8C

"!Q9! QP"!Q9!R-182S69)8.P9C

"SM2979@)-8)*&&+)!O%T)UVR14P4C9@)P6-B)W41)>C8 P9124@

X UI4)89C8)09-J)P9124@C)E,UV/

! >Y*+),-.)Z)'Y*[)\]6.)*&&')^ 52_O!"#`":!"#

! >Y*?)59P89J091)*&&')^ 5S2_O!"#`":!"#

X R6-B)84)89C8)O!"#`a:!"#)P14848.P9C)2B)CP12BA)*&&[

... et protons de basse énergie

Page 25: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Maintenant, au FNAL ...

1914

Accent dans le futur (>2008) la reponse aux hadrons !

Page 26: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Maintenant, au FNAL ...

2014

Page 27: Si-W ECAL Tests en faisceau du prototype · C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype Le prototype ECAL SiW en faisceau test

C Cârloganu, LPNHE, 30.06.08 SiW ECAL : tests en faisceau du prototype

Conclusion

2114

Quatre laboratoires ( LAL Orsay, LLR Ecole Polytechnique, LPC Clermont, LPSC Grenoble) impliqués dans l’analyse des données.

Un article soumis fin mai sur le comissioning du détecteur (EJPh),

Un article en revue interne CALICE, à soumettre à NIM.

... et six notes internes/publiques CALICE ...

Beaucoup de choses intéressantes à faire, nouvelles contributions bienvenues!