US Tracker Group Status Sep. 1, 2005 J. Incandela For the US CMS Tracker Group
Dec 21, 2015
US Tracker Group StatusSep. 1, 2005
J. Incandela
For the US CMS Tracker Group
Sep. 1, 2005 - J. Incandela 2
US Hybrids Inventory (approx)
Hybrids as of 5/7/05
FNAL IT UCSB IT
L12pu 31 - 170 -
L12pd 116 - 167 -
L12su 81 - 94 -
L12sd 55 - 79 -
L34pu 31 - 170 -
L56pu 121 - 203 -
R5p 33 -
R5s 55 41
R6 19 7
R7 20 -
•Have maintained TEC production
•We have not built new TOB modules for several weeks
• Developed backplane wirebonding (see below).
• As a result of this stoppage, we now have a decent stock of TOB components
• Will allow us to sustain uninterrupted production for some time…
Sep. 1, 2005 - J. Incandela 3
US Module Assembly (as of 8/29)
Week Cumulative
A B F A B F % A or BL12pu 0 0 0 11 1 0 100.0%L12pd 27 1 2 27 1 2 93.3%L12su 0 0 0 0 0 0 0.00%L12sd 0 0 0 0 0 0 0.00%L34p 30 0 0 799 23 5 99.4%L56p 0 0 0 1305 34 13 99.0%R5N 2 0 1 141 6 1 99.3%R5S 17 0 1 106 2 3 97.3%R6 39 0 0 186 12 0 100.0%R7 0 0 0 128 14 2 98.6%Total 115 1 4 2703 93 26 99.1%
Sep. 1, 2005 - J. Incandela 4
Module Assembly Issues
• Past Week• Both sites had 2 F grade modules due to sensor angles slightly
out of specifications.
• UCSB has only one operational wirebonder for modules. Repair scheduled for Sep.12. Need to improve speed of repair response from K&S (can do so with priority contract ~20k$/year). Machine is however usable for backplane bonds and we are using it for TEC modules now.
• Preparations for TEC at FNAL• UCSB R7 plate designs have been adapted to FNAL gantry
requirements.
• Manufacture of parts this month.
Sep. 1, 2005 - J. Incandela 5
US Hybrid Testing Totals (8/29/05)
Weekly CumulativePass Fail Pass Fail % Pass
L12su 0 0 0 0 0.00%L12sd 40 0 40 0 0.00%L12pd 41 0 130 0 0.00%L34p+L12pu 0 0 1042 15 98.58%L56p 3 0 1634 22 98.67%R2N 1 0 93 2 97.89%R2S 1 0 78 0 100.00%R5N 0 0 285 3 98.96%R5S 1 -1 214 3 98.62%R6 28 0 286 6 97.95%R7 0 0 158 1 0.00%Total 115 -1 3960 52 98.70%
Continue to recover previously failed hybrids by repairing them. Yield now 98.7%, was 97.1% in June.
Sep. 1, 2005 - J. Incandela 6
US ARCS Testing Results
•2675 Total Tested
•2640 Grade A
•10 Grade B
•25 Grade F
•This week:
•34.5 of 47616 channels bad (0.072%)
•Total Production:
•1057 of 1740280 channels bad (0.061%)
Total US Modules Tested
0
500
1000
1500
2000
2500
30001
/3/2
00
5
2/3
/20
05
3/3
/20
05
4/3
/20
05
5/3
/20
05
6/3
/20
05
7/3
/20
05
8/3
/20
05
Grade A
Grade B
Grade F
Total
US Modules
1
10
100
1000
10000
1/3
/20
05
1/2
4/2
00
5
2/1
4/2
00
5
3/7
/20
05
3/2
8/2
00
5
4/1
8/2
00
5
5/9
/20
05
5/3
0/2
00
5
6/2
0/2
00
5
7/1
1/2
00
5
8/1
/20
05
8/2
2/2
00
5
Grade A
Grade B
Grade F
Total
WIEN BOX:3 modules have
failed in LT
Sep. 1, 2005 - J. Incandela 7
Module Quality
•Total Production:
•1057 out of 1740280 channels are bad (0.061%)
Modules
1
10
100
1000
10000
100000
1000000
10000000
3-J
an
7-F
eb
14
-Ma
r
18
-Ap
r
23
-Ma
y
27
-Ju
n
1-A
ug
Ch
an
ne
ls B
on
de
d
0.000%
0.010%
0.020%
0.030%
0.040%
0.050%
0.060%
0.070%
0.080%
Pe
rce
nta
ge
Ba
d C
ha
nn
els
Total Bad Channels
Total Channel Tested
% Bad Total
Sep. 1, 2005 - J. Incandela 8
Completed fractions
0
200
400
600
800
1000
1200
1400
1600
1800
2000
L12p
u
L12p
d
L12su
L12sd
L34p
L56p
R5N
R5S
R6
R7
US Module Producton
Total Needed
Total Good
Sep. 1, 2005 - J. Incandela 9
Ag Epoxy Issues: HV Resistance
Delta Sensor Resistance (316 sensors) After LT Testing - After Assembly
1
10
100
1000
-10
-8 -6 -4 -2 0 2 4 6 8 10
ohms
Fre
quen
cy
Sep. 1, 2005 - J. Incandela 10
Ag Epoxy biasing data: 2 months on the shelf
Delta Sensor Resistance (144 sensors) July 2005 - May 2005
1
10
100
-250
-150
-50
-30
-10
-8 -6 -4 -2 0 2 4 6 8 10 30 50 150
250
ohms
Fre
quen
cy
Sep. 1, 2005 - J. Incandela 11
Backside Bonding – TEC Modules
•We have backside bonding fixtures built for R5N, R5S, R6 and R7 modules.
•Modules are placed on the bonding fixture without carrier plate.
•Much care was taken to make sure teflon blocks are free of metal chips since the front side of the sensors are very delicate.
•5 – R7, 18 – R5S and 21 – R5N modules (46 total) have been backside bonded and arcs tested without problem.
Sep. 1, 2005 - J. Incandela 12
Backside Bonding – TEC Modules
•Bonds height average .30 mm.
•Last box of 20 modules were bonded in less than 1 hour.
•We are putting on 15 bonds. This is what the TIB people have done.
•Bond pull strengths have all been over 7 grams.
Sep. 1, 2005 - J. Incandela 13
Backside Bonding – TEC Modules
• We encapsulate the backside bonds for protection from damage.
• Single wire bonds have been encapsulated on test pieces with sylgard 186 and DC 3140.
• They are being thermal cycled from -30C to +50C.
• After 30 cycles both still conduct fine.
• We will continue thermal cycling them for at least 50 cycles.
• They are held at the extremes of -30C and +50C for 30 minutes during each cycle.
• The encapsulant height can be close to the frame thickness. To make sure we do not go outside this envelope, we will squeegee off any excess to constrain to .80 mm height.
Sep. 1, 2005 - J. Incandela 14
New TOB Modules
• Solder a gold plated 100 m thick Kapton tab to the frame circuit.• Initial bonding tests of bonding show no problems.
• Built fixtures for SS and DS-r frames that hold the Kapton tabs in place while they are soldered.
• Assures minimal gap.
• One small recess and one block of Teflon must be added to our assembly plates to accommodate these tabs and allow them to be glued to the back of the sensors during assembly.
• The blocks were made and glued on some assembly plates yesterday.
• They will get machined today.
• Will restart TOB production next week
Sep. 1, 2005 - J. Incandela 15
Backside Bonding – New TOB Modules
Sep. 1, 2005 - J. Incandela 16
Backside Bonding UCSB – Old TOB Modules
•For most modules we can silver epoxy a wirebond tab to the portion of the frame Kapton bias ring (photo).
•For modules that already have epoxy all around the ring, we will run a wire to the other side of the frame as proposed by FNAL.
Sep. 1, 2005 - J. Incandela 17
FNAL: Wire Connection to Backside
FNAL to use magnet wire with acceptable insulation rating soldered to HV on topside, then run to back side where it connects to a Au-plated Kapton piece that bridges the 2 sensors. No interference problem with the carrier plates or with double-sided rod construction.
Sep. 1, 2005 - J. Incandela 18
ROD Workshop
• We had a US rod production workshop Aug 21-22 at UCSB• Focused on electrical testing, since mechanical issues under control
• Goal was to firm up details of procedures, tests, etc., so that we will be really ready for production when the new ICC cards show up
• Generated set of 22 action items• Minutes at http://hep.ucsb.edu/cms/rod/workshopAug2005.html
• Several items need actions taken a by CERN experts:• Awaiting optical FECs from CERN
• Should make system more robust!
• We will likely need help from Europe for installation
• Digital Tests
• Still waiting for CERN experts to help us define these!
• ICC cards swap
• Have many rod frames in US that need to have ICC cards swapped
• We prefer this is done at CERN. We can send manpower.
Sep. 1, 2005 - J. Incandela 19
Integration team evolution
• US integration team starting to build up• J. Spalding will be my deputy at CERN (resident as of Oct.)
• S. Tkaczyk will also be resident and will help oversee testing group
• J. Lamb who is my graduate student at UCSB and who has substantial rod testing experience will arrive in Sep. and stay resident indefinitely.
• We expect to have a post-doc and another student coming by January and several excellent and very experienced people have been contacted and are considering residence at CERN.