CSN 1 INFN LECCE 2003 BARREL MU DT

CSN 1 INFN LECCE 2003 BARREL MU: DT status Installation plan Minicrate test f. gasparini 1

SUMMARY & CONCLUSIONS SL and Chamber assembly on schedule Installation plan compatible with DT, RPC, MB 1 supports and wheel availabilty HV cable manufacturing started Agreement for cable radial cable trays/DT cables defined Good results from MINICRATE test of May Minicrate production: small delay but still compatible with installation plan NEW LINK Alignment news 2

THE BARREL MU DRIFTCHAMBERS 4 concentric stations 60 chambers each (MB 4 70 ch. s) 680 SLs (MB 1, 2, 3: 3 SL each, MB 4 2 SL) 171. 732 wires Four assembly sites Aachen 60 MB 1 + 10 MB 1/4 CIEMAT 60 MB 2 + 10 MB 2/4 LNL 60 MB 3 + 10 MB 3/4 Torino 40 MB 4 (with two tables) 3 LHCC Comprehensive Review 10/2/01 CMS MUONS: Drift Tubes and Alignment fabrizio

The Installation Task The Barrel Muon system comprises 250 chambers in 7 flavors: 5 4 6 7 8 9 10 60 60 60 40 10 10 10 3 2 1 12 11 MB 2 MB 3 MB 4 MB 1 MB 2 MB 3 3 SL 3 SL 2 SL 2 SL 2 RPC ~2. 0 x 2. 54 m 2 960 kg 2 RPC ~2. 5 x 2. 54 m 2 1200 kg 1 RPC ~3. 0 x 2. 54 m 2 1300 kg 1 RPC ~4. 2 x 2. 54 m 2 1800 kg 1 RPC F SL SL Brown MB 4 (Torino) Have two Superlayers only Honeycomb F SL 10 Sectors will be installed in SX 5 => 210 Chambers(310 RPCs). Sectors 1 and 7 are used for the lowering fixture and will be installed in UX 5 => 40 Chambers(60 RPCs). 4

TYPICAL CHAMBER PERFORMANCE: Plateau knee Hit rate 3. 5 larger that the worst case In LHC Drift cell effic. &resolution With increasing Background CMS Note 2003/007 5

Produzione camere: Quality Control In tutti i 3 siti in produzione (A, M, Pd-LNL): • Misure di posizione dei fili • Misure di tensione dei fili • Misure di planarità (alveare e camera) • Controllo di tenuta pressione • Controllo rumore elettronica • Test con raggi cosmici • Dati immagazzinati in area web-DB 6

Madrid: Position & Tensione Fili SL at table 1 Layer 2 Layer 3 Layer 4 Mean=6. 4 m = 46. 0 m Mean=-0. 4 = 38. 2 m Mean=2. 93 N = 0. 10 N 7

PD-LNL Summario posizione fili Summario tensione fili 1 entriata=1 Layer 1 entriata=1 filo THETA <>=38 m FE <>=2. 9 N rms=0. 1 N <>=35 m HV 8 r. m. s. residui

Aachen: 9

PD-LNL Honeycomb and Chambers planarity Batch 1 Batch 2 Batch 3 Batch 4 Batch 5 10

Aachen: test cosmici 11

Pd-LNL area stoccaggio dati Top (summary) http: //www. pd. infn. it/dbcms/db/ Analysis CHAMBER: http: //www. pd. infn. it/dbcms/db/MB 3_006/ Super. Layer 13

C-beams Parts needed to assemble end test one SL: H-Comb firm I-Beams firm 1) Aluminium plates with field elecrodes 2) Aluminium beams with cathode electrodes 3) Hcomb Panels (that house services) 4) HV Boards in the gas volume 5) Amplifiers on boards in the gas volume Protvino Tools Bologna Tools Torino Al. Plates firm Dubna Aachen HVB, HVC HV Cables 1) 3400 plates (2040 double sided, 1360 single s. ) 370. 000 electrodes. Mass prod. in Dubna. Tools designed and commissioned by INFN Torino (330 plates prepared in To) CIEMAT CERN Crimping blocks Wires corner blocks LNL Torino ON SCHEDULE 2) 171. 000 I-beams = 342. 000 cathodes mass prod. in IHEP Protvino. Tools designed and commissioned by INFN Bo (13. 000 beams assembled in Bo) Gas covers profiles end plugs FE-boards ON SCHEDULE IHEP Beijing Padova CIEMAT Aachen 3) produced by Hexcel 4) boards assembled in IHEP Beijing tools and jigs from INFN Padova (delay, not critical*) 5) boards and chips by INFN Padova 6) * Delay generated by a delay in PCB availabil. LHCC Comprehensive Review 10/2/01 14 CMS MUONS: Drift Tubes and Alignment fabrizio

2897 plates cut in Torino 2139 plates produced in JINR under Torino coordination Bad channels < 1‰ Strip Production in JINR-Dubna LHCC Comprehensive Review 10/2/01 CMS MUONS: Drift Tubes and Alignment 15 fabrizio

DT Chamber Production (18 DT/site/year) 230 All DTs 180 130 120 Planned Achieved - Planned 80 30 -20 Q 401 Q 102 Q 202 Q 302 Q 402 Q 103 Q 203 Q 303 Q 403 Q 104 Q 204 Q 304 Q 404 Q 105 Q 205 Q 305 Integral of produced chambers /quarter. 16

Total Superlayers produced Some man power problem in Legnaro in Apr/June ( announced) 100 22 ch/y 18 ch/y nominal Aachen Madrid Legnaro 50 0 1/1/2001 1/1/2002 1/1/2003 1/1/2004 Date 17

Chambers OR 3 Fully Tested Super. Layers Cambers produced 50 40 32 30 20 Aachen Madrid Legnaro 10 0 1/1/2001 1/1/2002 1/1/2003 1/1/2004 Date m Madrid and Aachen begun to make the small MB 4 Chambers with 2 SL only (padova in Nov) a 18

LNL 5/9/2003 Aachen 15/8 Madrid 10/9 96 SL 102 SL 120 SL 32 CH 34 34 CH 100 ch out of 210 Chambers at ISR Total MB 1 MB 2 MB 3 20 28 25 73 by dec. 03 29 38 30 97 These figures can imply that all the dressing will be done for all chambers at ISR (to be agreed) 19

INSTALLATION CONSTRAINTS: RPC availablity ( see Iaselli) MB 1 supports Torino line Minicrates availability To cope with Minicrates: begin installation in lower parts of the Yokes To easy and speed RPC production, the priority for RPC is RB 2>RB 3>RB 1>RB 4 (see Iaselli) First MB 4 chambers will be installed in april 04 (sect. 9, 10, 11 from Aachen, CIEAMT and Legnaro). 20

MB 1 supports: To be made in IHEP Beijing, available Jan. /Feb. 2004 because of the late availability of the final drawings. Torino line: The planning was revised in detail in June with Alain. Drawings done in Torino (Mereu, Daudo) and CERN (Mimmo and Boki Goal : have the first table and related tools and references ready for end of October Bridge supports installed on the two tables by end July : done Bridges commissioning by end August beg. sept. : done Honeycomb plates for I-beam tools available end July done Alcoa plates and side profiles for I Beam tools available Finishing I-beam tool in September going Drawings of ancillary parts to make table and tool operational were planned to be available end of August going 21

Torino line versus v 33. 2: Surface assembly (Madrid is running for an year at 22 ch/year) 4 4 19 ch. /y 12 30 24 40 4 22

Torino line versus v 33. 2: underground assembly (Madrid is running for an year at 22 ch/year) 4 40 19 ch. /y 12 30 24 40 40 + 2 months 23

~ 23 ch ~ 24 ch 5 4 6 7 8 5 to 8 ch/week 3 2 YB+1, +2 2003 6 1 9 10 11 9 Torino MB 4 8 TAKES INTO ACCOUNT DELAY OF MB 1 SUPPORTS ~24 ch 2 Mb 1 in YB-1 5 4 3 2 9 1 12 10 11 9 Torino 6 YB – 1, -2 in 2005 5 4 3 2 YB 0 first 7 11 1 YB-2 NOW 12 9 10 8 132 DT/RPC packs installed by end 2004 YB 0, +1, +2 1 mar. apr 04 8 2 YB 0 jan, feb 04 1 7 YB-1 7 3 12 8 6 4 5 (may/june) 1 (then +1, +2 In fall 04) 8 12 9 10 1 11 < 64 ch. *+ 4 MB 4 sectors 8, 9 of YB-1 and – 2 that should be completed before the Mag. Test ~21 ch >42 ch* 24

v 33. 2: Surface assembly 4 4 12 30 24 40 4 YB+2, +1 YB 0 YB-1, -2 25

Complete the chambers at ISR: Accept and certificate Measure on the alignment bench Internal gas/cooling pipes HV and DCS cables R-Out cables (theta SL) Covers and protections , 26

27

28

Many types of chambers to be installed in many difficult places with different RPC configurations Many and complicate ancillary tools Are required. 29

electronics: DT electronics is sitting on the chambers in each chamber of one CMS sector : in the gas volume : F-End chips/boards & HV boards 32 F-End chips (2 Boards) feed: in the Minicrates : 1 ROB (Read Out Board) 128 wires 4 TDC (any of them can act as Master) => precise coordinates 1 TRB (Trigger Board) 128 wires 32 BTI, 4 TRACO, 1 TSServer =>2 best MU in the Board (64 cm) ONE TS Board houses the TSMaster (two parts, any can act as master) => 2 best MU in the chamber Sector collector board in the 4 th station (now on the balcony) Front End Ro. B T D C TB B T T T R S I A S C O TSB C T o S n M t r DCS SECTOR Collector LHCC Comprehensive Review 10/2/01 to the counting room C. Room 30 CMS MUONS: Drift Tubes and Alignment fabrizio

MINICRATES: Analysis of data taken in May going on : BTI modules fabrication: samples (50) and preserie (400) : First batch (~ 1000) Trigger Boards : all material available for 50 board (10 ~ 12 Minicrates) positive started CCB and SB : some adjustment after the test September Mechanical parts production of last parts: (some piece to be modified) Started or starting soon Few Mincrates will be assembled in fall 03, starting October In the plot the rate is 10 MC/month. SECTOR Collector and DDU: These are the last missing parts: prototype of DDU exist and the first Unit will be produced in beginning 2004 Sector Collector is fully designed , the input section tested in May test beam , material procurement will start soon in 2004 as planned. 31

MINICRATES versus v 33. 2: Surface assembly 250 U 200 150 100 Mag. test 50 Installed chambers without minicrates Minicrates (10/month) 32

MINICRATES versus v 33. 2: underground assembly 250 Inst. chambers Minicrates UX 200 150 100 Mag. test 50 (+ 2 months) 33

TRIGGER TEST ON THE 25 nsec BEAM OF MAY Crew from Padova, Bologna , CIEMAT, Torino and Aachen ~ 10^ 7 events were collected At different angles and in all the trigger configurations I would like to stress that this is a result of a very preliminary analysis whose aim was to check that the gross features of the trigger are as we expected in order to finalize the design of the last details of the boards. A deeper understanding of the way the “real” trigger works will require some time. 34

35 DT Trigger Performance - Sara Vanini 11

Drift. Tubes Local-Trigger 36

Each Trigger has quality bits telling the type of trigger: The BTI generates a Trigger if 3 or 4 “aligned” are found in its Superlayer : they are named L or H respectively. BTIs look each to 3 adjacent cells , with a pitch of one cell. TRACO correlates the signals from the BTIs of the two SL (same time and within the geometrical acceptance) TRACOs overlap too TSS+TSM select among all TRACOs the two best tracks, rejecting Duplicates, ghosts and fakes. There are several settings of the Trigger : default (CMS Standard) Backup, etc …. in order to recover possible failure in TSM or CCB reduce the “noise” etc…. Data shown below are from some runs in the Default mode. (standard CMS mode). 37

The Trigger needs at least 3 aligned hits in a SL identifies the time of passage of one track (Bunch crossing identification) makes and sends out a measurement of the radial (= position) and bending angle. Resolution is < 1 mm and from 5 to 15 mrad resp. depending on the Trigger type (Correlated/uncorrelated) Bending angle ( in test beam= inclination fixed depends on run) Correlated trigger (Slo & Sli) Uncorrelated (Slo or Sli) SLouter SL inner Radial angle (= position) 38

BTI outer TRACO TSS/TSM H HH * * BTI inner * H * * * L H * * HL (LH) * * * Ho * * * * N * * Hi * * H 39

* Chamber Super. Layer aa * * BTI Shift registers b * * * Good track H trigger Ghost track out of time a+b = /T L type (3 hits) a +b = T 40

TRACO can find HH HL (LH) LL four “aligned” hits in the two PHI SL 4 and 3 (3 and 4) hits in the two PHI SL 3 and 3 In case TRACOs do not find a correlation, single SL Trigger are transmitted Ho or Hi (H trigger from outer or inner SL) Lo or Li (L trigger ……. . ) TSS and TSM look to TRACO outputs etc………. 41

Trigger Output Clock triggers vs pattern unit slot clock 14 is the right one triggers in other location are "ghosts" or false triggers 42 DT Trigger Performance - Sara Vanini 13

Ronchese, Vanini, Zotto The right crossing is in the 14 th clock. Ghost triggers, found by BTIs looking for aligned hits in the shift registers are strongly suppressed by the correlation request in HH and HL But they are still present in LL, and obviously in the single SL triggers. Mechanism exist to suppress the ghosts in the LL, Lo and Li trigger. The data shown are in the “default configuration”. Analysis is going on. 43

BX Efficiency vs Incident Angle - 1 # ev. trig. at 14 bx efficiency = total # of events Event= scintillator + 2 hits in beam region tracks hitting 2 I-beams “EMULATOR” is a SW package that emulates the full chain of the Trigger Chips and Boards. It works off line and need as input the times recorded by the TDCs. “Emulator “ is not the CMS simulation, it should reproduce exactly the Trigger data. To compare with the expected performance as shown in the Trigger TDR we should compare with the full track reconstruction from TDC data. However this first result is in excellent agreement with what we expect. ~ 98 -99 % efficiency up to 350 44 DT Trigger Performance - Sara Vanini 14

BX Efficiency -2 HH HL LL drop of correlated efficiency at large angles due to geometrical cuts (single BTI 56 o traco 42 o) Ho Hi Lo Li 45 DT Trigger Performance - Sara Vanini 15

BX Efficiency - 3 HH HL Ho Hi 46 DT Trigger Performance - Sara Vanini 16

BTI test in Magnetic Field (test beam 2000, CMS note 2001/051) Prob. of one/4 layers affected by delta 16%, more than one 4% B (TESLA) longitudinal component BW Bn . 05 . 16 . 21 . 19 . 38 . 56 . 77 1. 15 . 31 H-trig = 4/4 layers L-trig = 3/4 layers BTI n+1 B (TESLA) radial component . 26. 96 47 BTIs overlap

Ho Fit vs Trigger Comparison HH… Hi 48 DT Trigger Performance - Sara Vanini 22

Fit vs Trigger comparison gap among the triggers broken lines: border between traco 9 and 10 emulation shows the same behaviour ! corr outer inner 49 DT Trigger Performance - Sara Vanini 20

Fit vs Trigger Comparison emulation with correct LUTs shows the right correlation correlated 50 DT Trigger Performance - Sara Vanini 21

Fit vs Trigger comparison - 6 extrapolation at 0 o 51 DT Trigger Performance - Sara Vanini 24

Fit vs Trigger comparison - 7 extrapolation at 20 o 52 DT Trigger Performance - Sara Vanini 25

Fit vs Trigger comparison - 8 extrapolation at -20 o 53 DT Trigger Performance - Sara Vanini 26

Detailed cross check with the TDC data are going on , and will take some time. The comparison is done selecting very good tracks and looking in the trigger data How many times there is at least one trigger signal in the right crossing (the 14 th clock) Perpendicular tracks (~ 70000 events) 4 and only 4 aligned hits in the interested cells of SL 1 and 2 HH HL Ho, Hi LL, Lo, Li No trigger 97. 0 % 1. 7 %. 9 %. 1 %. 3% Total 99. 7 +-. 02 % 0 degrees The table below give the efficiency of thr trigger for different hit configuration Single hit per cell: 4+4 4+3 or 3+4 4 + n or n+4 Global effic. 99. 83 99. 42 Including cells with more than one hit per cell Global effic, 98. 03 94. 01 M. C. Fouz 94. 38 54

CONCLUSIONS The test was very useful and validated the trigger electronics. The large statistics cumulated allows a detailed study of the trigger Performance. No bad surprise so far. An error was identified in the test pulser control, and corrected in the CCB/SB Boards. It was found that some detail of the MC mechanics had to be improved. The first five final CCB are under assembly and available at end September. Mechanics is in production. Read out part of the MC will be assembled in CIEMAT and ship to Padova for the Trigger part and full test. The final test stand for mass testing is in preparation in Padova with participation of Padova and Bologna people. 55