TEC X5 Setup (1) 1 Front + 1 Back petal, fully equipped (28 modules F.P. + 23 modules B.P. + AOH + DOHM) Petal assembly IPN-Lyon (F.P.), RWTH-Aachen I.

TEC X5 Setup (1)

• 1 Front + 1 Back petal, fully equipped (28 modules F.P. + 23 modules B.P. + AOH + DOHM)

• Petal assembly IPN-Lyon (F.P.), RWTH-Aachen I (B.P.)

• Constitutes one full TEC control ring• Full optical link with optical PMC-FEC (VME-

FEC will be added in coming days)• 2 x FED9U in same crate (68 chans F.P., 56

chans B.P.)

TEC X5 Setup (2)

• Petals placed back-to-back in insulated cold / dark box

• Series cooling loop: B.P. -> F.P. -> copper cooling plate

• Cooling plant 500 W @ -20C

• Temp. / Humidity monitoring + DCU information being read

Installation experience

• Mechanical installation, cabling, etc.: 2 days• FEC communication established immediately• First Landau: 4 days after arrival• Approximate strip statistics after installation at

X5 - 1 dead readout channel (256 strips) not included:– Dead: 99 / 31488 -> 0.31 %– Noisy: 8 / 31488 -> 0.03 %– (No loss of strips after 800 km car transport Aachen

to CERN)• One module with HV short

Early Results

• FED9U works– Internal pedestal + common mode noise subtraction– Zero suppression being studied – first results

encouraging

• ENC noise close to expected:– Estimated independently using (a) calibration pulses

(b) signal peak (assume 22,000 e- in 320 micron Si)– Fit with known detector capacitance gives:

ENC = 511 + 25C (C ranges 11 pF R1 to 26 pF R7)

ENC noise from cal runs

y = 24,76x + 511,09

R2 = 0,8272

400

500

600

700

800

900

1000

1100

1200

0 5 10 15 20 25 30

Cstrip (pF)

EN

C n

ois

eNote: online data, very rough analysis!

ENC noise from signal peak

Voltage scan – FP R1/R2

Voltage scan – BP R1/R2

Selected Landaus, Peak Mode

V = 300 V

S/N vs Ring (Peak Mode)

0

5

10

15

20

25

30

35

40

45

0 1 2 3 4 5 6 7 8

Ring

S/N

(p

ea

k m

od

e)

0

0,1

0,2

0,3

0,4

0,5

0,6

Th

ickn

ess

/ EN

C n

ois

e (m

u /

e)

S/N Thickness / ENC

V=300 V

FED Zero Suppression

Note: Problem with internal FED common mode subtraction for these runs

First results at low temp.Temperature of sensor leg = -5 °C approx.

(Compare S/N room temp = 18)

Problems

• One dead R5 readout channel (256 strips)

• HV short on FP stereo module 2.4

• BP 2.1 does not receive HV

• One R5 module with very low noise (broken bonds?)

• High CMN on R5 module associated with noisy strip

F.P. 2.1 does not get HV

Dead readout channel

???

ST sensor CMN problem

To Do this Week

• No beam• Improve thermal contact of T and H sensors on modules• Improve cooling circuit and thermal isolation of box• Logging of DCU values• Commission VME FEC• Try higher fast control speeds• Pulse shape studies (cal runs)• Repeat bad strip study• Finalize installation of slow ctrl / interlocks