b904 integration

3
John Coughlan , Tracker Electronics Integration , 12 April 2005 presented by John Coughlan RAL B904 Integration Pre-Series Manufacture Final Production Plans

Upload: kellsie

Post on 04-Jan-2016

32 views

Category:

Documents


0 download

DESCRIPTION

B904 Integration. Pre-Series Manufacture Final Production Plans. presented by John Coughlan RAL. B904 Integration. Firstly, apologies if this has already been covered in 2 previous meetings and/or is stating the obvious… - PowerPoint PPT Presentation

TRANSCRIPT

John Coughlan , Tracker Electronics Integration , 12 April 2005

presented by John Coughlan RAL

B904 Integration

Pre-Series Manufacture

Final Production Plans

John Coughlan , Tracker Electronics Integration , 12 April 2005

B904 Integration Firstly, apologies if this has already been covered in 2 previous meetings and/or is stating the obvious…

…and approaching this problem from point of view of validating FED boards.

Scope : only considering OFF Detector Electronics i.e. crates in USC55.

Strategy : to have final Tracker DAQ system i.e. final racks destined for USC55 operational as test bed in B904

Schedule and Resources : need to identify one overall responsible for B904 hardware (CERN based). Subdivide tasks for hardware/firmware/software. Milestones.

Arrive at System Test with all final components for 2 partitions asap;

So many ‘new’ components still to be integrated… APVE, FMM, OptoFEC, OptoFED, LTC, TTCci, TTC splitter, CAEN CC, FRL…

Distinguish System Tests vs Acceptance Tests

FED System Test : Test 2? full crates together with final Tracker DAQ elements see above (and assoc S/W). Synchronisation.

FED Acceptance Test : Follows on from UK acceptance tests see next slide

Individual FEDs will be tested as far as possible with Opto inputs/SLINK at RAL. But using FED Testers not final system.

Require system at B904 for CMS to validate FEDs (and other components)

Envisage Full Crate of FEDs (Opto inputs or simu test frames) running continuously “soaking” with central Tracker DAQ

Q. CMS Acceptance / Validation Policy?

Either refill this crate with random sample of FEDs from batches or re test every FED.

Both are doable, question of logistics. Target of ~ 50 boards / month commencing October 2005.

FEDs go into storage at B904? awaiting installation in USC55.

John Coughlan , Tracker Electronics Integration , 12 April 2005

Acceptance/Validation of FEDs

1. Custom Tests atAssembly Plant

BScan, VME crate

3. Tests at CERNPrevessin 904

DAQ Integration

2. Tests at RAL &IC

OptoRx, Full crate

4. Installation at CMSUSC55

0. Quality Controlsduring Assembly

processAOI, X-ray

0. Assembly ProcessQuality Checks

E.g. AOI

2. Fit FrontPanelDeflector Bar

Jumpers

1. VisualInspection

Multimeter tests

3. Serial NrFit FP label & 2D

Bar code

4. Insert in CrateCheck formechanics

5. Power On CrateVerify LEDs

6. Boundary ScanSave results

7.Program EPROM(change Jumper)

8.Power Reset

buttonVerify LEDs

10.Test Serial EPROM

Load with Ser Nr, Date etcJumper for Write Protect

9.Test VME Access

11.Test FPGA loadingInsert CFlash Card

Power ResetVerify LEDs for Load Done

Flashing Clocks

12.Run Test Bench Programs:Exercise FPGA Registers

Read TemperaturesRead Voltages

Scan DACs and OptoRx settings capture data

13.Final Tests

Box up

Results of Operator Checks andProgrammable Tests should go to

Dbase

For details of tests and checksand acceptance criteria see

detailed diagrams

VME Crate Testing for Analogue

Test Flow from Assembly Plant to USC55500 boards to test over 10 months. Essential to catch any manufacturing faults early.

Boundary Scan Testing for Digital

Testing by Assembly plant operatives