Status Report on Bonding at Catania

1
Status Report on Bonding at Catania

• Salvatore Costa
• University and INFN Catania

2
Real Module Bonding

• At previous Meeting, I reported that the first
  real module bonded in Catania had been bonded in
  early August 2002, just before our Clean Room was
  shut down for transfer to new Department
  Building.
• It was a TIB module with two 512-strip old CSEM
  sensors.

3
Test of Real Module Bonded

• Besides a global problem with a bad bias bond,
  which initially prevented the Module from being
  read at all and had to be redone, I reported on
  readout tests made on that Module, which showed
  that also some 6 to 8 channels had individual
  problems. (And I commented that being the very
  first Module, that probably wasnt so bad).
• At the time of that Meeting those tests had just
  been done and no further insights were available
  as to what we could have done wrong.

4
Test Results after Bonding

• (by CMS-like)

5
Fault Investigation

• We have visually inspected the Module and found
  the likely cause of the faults
• In short, all PA-SEN bonds look fine, but all
  SEN-SEN bonds are flawed by what in retrospective
  looks as a rather stupid mistake we used for
  SEN-SEN bonding the same loop mode (Hughes loop
  5) as for PA-SEN, which, being an asymmetric
  loop, is good to bond between pads at different
  heights but not for pads at the same level. gtgtgtgtgt
• It is conceivable that we have not created new
  pinholes.

6
The Wrong Loop Mode
Loop 5
SEN2
SEN1
AC1
AC2
AC1
DC
AC2
DC
7
Action Plan

• In principle we could ignore this problem, as all
  TIB modules we will possibly bond should have a
  single SEN.
• However, as soon as Clean Room ready in new
  Building (Exp March 2003) we will
• pull out all wrong SEN-SEN bonds
• redo them with loop 0 or 4
• re-inspect
• re-test pinholes?

8
Test Systems

• CMS-like is fully operational, as it has been for
  a while already
• ARC Progress!
• At previous Meeting I had reported that after
  replacing a faulty card we had seen it
  internally working - no Hybrid available)
• Now we have successfully read a Hybrid with
  version 4.X
• Installation of v. 5 underway

9

• break

10
DataBase Interface Update

• Salvatore Costa
• University and INFN Catania

11
Interface Dev News

• Contrary to what I said at the last Meeting, I
  decided to release yet another intermediate DEMO
  version, v. 0.3, which features
• Default values, in fields where applicable, for
  all centers!
• Updated scrollable lists for data such as tool
  types, wire types etc.
• Many bug fixes
• Various improvements requested by people
• Why? Because
• A key piece of info to read from/write to
  TrackerDB is becoming available only now
  (tomorrow?)
• Module object names, object types
• I had overlooked that we need to handle bad
  channels with a greater level of sophistication
• gtgtgt 2 questions, 1 issuegtgtgt

12
Center DB Readiness
13
XML File Contents

• lt?xml version"1.0" encoding"UTF-8" ?gt
• ltDBFile xmlnsxsi"http//www.w3.org/2001/XMLSchem
  a-instance"gt
• ltxsinoNamespaceSchemaLocation"calibration.xsd"gt
• ltobject_id"111" /gt
• ltactiongt
• ltaction_description name"PREBOND" version"1"
  object_name"MOD" object_typeTIB4"
  input_id"123" /gt
• ltresult name"centre_name" value"SANTA BARBARA"
  /gt
• ltresult name"tool_id" value"2" /gt
• ltresult name"tdate" value"2002-10-17T181637"
  /gt
• ltresult name"INSPECTOR" value"Susanne Kyre" /gt
• ltresult name"RECV_TIME" value"2002-10-17T1816
  00" /gt
• ltresult name"UNPACK" value"None" /gt
• ltresult name"STATUS" value"Good" /gt
• ltresult name"QUICK_TEST" value"No comment" /gt
• ltresult name"STORAGE" value"None" /gt
• ltresult name"SHIP_TIME" value"2002-10-17T1816
  00" /gt
• ltresult name"READY_FOR_BOND" value"yes" /gt
• ltresult name"status" valuereference" /gt

See Excel filegtgtgt
14
Question 1 Hybrid Bonding?

• At an earlier Meeting we decided I would add to
  the Interface an 8th section for Hybrid Bonding.
• Add to all Centers or to only a few? (Which
  ones?)
• Does Hybrid Bonding happen before assembly at
  Gantry? If so, Module onto which it will be
  mounted is not known yet, thus
• Record data into TrackerDB with object_nameHYB?

15
Question 2 Bad channels ?

• Currently, I read from TrackerDB list(s) (1 or 2)
  of bad SEN channels (can distinguish shorts from
  pinholes) and just throw them to the Operator.
• Need to LOAD (up to) 4 lists of bad channels
• HYB, PA (?), SEN1, (SEN2)
• and SUGGEST appropriate action

(short or pinhole)
(bad)
(short)
HYB
PA
SEN1
SEN2
Examples of possible actions next slides gtgtgtgtgt
16
Bad Channels and the DB (1)
bad
HYB
PA
SEN1
SEN2
17
Bad Channels and the DB (2)
?
short
HYB
PA
SEN1
SEN2
Do not bond only highest
18
Bad Channels and the DB (3)
pinhole
HYB
PA
SEN1
SEN2
19
Bad Channels and the DB (4)
pinhole
HYB
PA
SEN1
SEN2
20
Bad Channels and the DB (5)
?
short
HYB
PA
SEN1
SEN2
Do not bond only highest
21
Bad Channels and the DB (6)
?
short
HYB
PA
SEN1
SEN2
Do not bond only highest
22
Issue Bad Hybrid channels

• If the Interface has to suggest to skip any
  PA-SEN or SEN-SEN bonds because of a bad Hybrid
  channel, then I need to incorporate algorithms
  that translate Hybrid channel numbering (consult
  Vincent Lemaitre) into SEN strip numbering for
  all different instances of relative mounting of
  APVs and Sensors on Modules.