LAr Frontend Board PRR Introduction 1 2 3
LAr Frontend Board PRR Introduction 1. 2. 3. 4. Overview of FEB functionality FEB performance requirements Development and evaluation of FEB Organization of FEB PRR presentations and documents 5. John Parsons 6. Nevis Labs, Columbia University 7. March 18/2004 12 sept 2001 C. de La Taille 7 th conference on LHC electronics Stockholm
Frontend Board Overview n functionality includes: n receive input signals from calorimeter n amplify and shape them n store signals in analog form while awaiting L 1 trigger n digitize signals for triggered events n transmit output data bit-serially over optical link off detector n provide analog sums to L 1 trigger sum tree J. Parsons, FEB PRR, March 18/2004 2 2
Overview of Requirements of ATLAS LAr FEB n n n read out 170 k channels of calorimeter (total of 1524 FEBs) dynamic range 16 bits measure signals at bunch crossing frequency of 40 MHz (ie. every 25 ns) store signals during L 1 trigger latency of up to 2. 5 s (100 bunch crossings) digitize and read out 5 samples/channel at a max. L 1 rate of 100 k. Hz n measure deposited energies with resolution < 0. 25% n coherent noise per channel < 5% of total noise per channel n measure times of energy depositions with resolution << 25 ns n n high density (128 channels per board) low power ( 0. 8 W/channel) high reliability over expected lifetime of > 10 years must tolerate expected radiation levels (10 yrs LHC, no safety factors) of: n TID 5 k. Rad n NIEL 1. 6 E 12 n/cm 2 (1 Me. V eq. ) n SEU 7. 7 E 11 h/cm 2 (> 20 Me. V) J. Parsons, FEB PRR, March 18/2004 3 3
Module 0 Electronics Experience n Full functionality “Module 0” FEB boards were developed Provided verification of the electronics design concepts n Used in testbeam runs with Module 0 and production calorimeter modules n n In total, almost 50 Mod 0 FEBs ( 6000 channels) were produced Have been operating reliably in testbeam for past several years (and continue to do so) n Performance meets or exceed ATLAS specifications n n Due to schedule, Mod 0 boards developed with some “short cuts”: Did not use final control signal (TTC and SPAC) distribution n FEB used Cu output cables instead of optical links n Did not pay strict attention to radiation tolerance requirements n n the main task remaining in the development of the final ATLAS FEB was to radiation harden the designs, and in particular to replace several FPGAs and other COTs with custom rad-tol ICs J. Parsons, FEB PRR, March 18/2004 4 4
Overview of main FEB components 128 input signals Analog sums to TBB 32 0 T 32 Shaper 2 LSB 14 pos. Vregs +6 neg. Vregs 32 SCAC 2 DCU 16 ADC DMILL AMS DSM COTS 8 Gain. Sel 1 Config. 1 SPAC 1 MUX 1 fiber to ROD 1 GLink 7 CLKFO 1 TTCRx TTC, SPAC signals n 10 different custom rad-tol ASICs, relatively few COTs J. Parsons, FEB PRR, March 18/2004 5 5
FEB Optical Links n one GLink output link per FEB, with rate of 1. 6 Gbps n link collaboration includes SMU, ISN, Stockholm, and Taiwan 1. 6 Gb/s J. Parsons, FEB PRR, March 18/2004 6 6
FEB Prototype Development n 128 channels/FEB n Large 10 layer PCB, with components on both sides n First FEB assembled and tested (without Vreg) during Nov. ‘ 01 n Pos. STm Vregs added Dec. ‘ 01 n First neg. STm Vregs (JQ 4) tested Dec. ’ 02 Vreg showed stability problems n Pinout change required iterated FEB design ( v. 1. 4) n n JQ 5 neg. ST Vreg tested June ’ 03 Stability problem solved, but radiation problems seen n 20 v 1. 4 FEBs made for FECT n n Two FEBs successfully tested with JQ 6 neg. STm Vreg in October ’ 03 n FEB design iterated ( v. 1. 5) in Jan. ‘ 04 to implement QPLL to reduce CLK jitter n Two v 1. 5 FEB produced and tested J. Parsons, FEB PRR, March 18/2004 7 7
Front End Crate System Test n Before proceeding to PRRs of FE crate electronics, LAr decided to perform system test (at BNL) with all boards required for one half-FEC (~ 1% of total system) n The FEC system test included 14 FEBs (1792 channels) n The main FEB results from FECT will be presented by Ioannis Katsanos n Last fall, it was observed during the FEC test that some of the FEBs showed large error rates in their output data optical links n Many detailed investigations since then have allowed us to understand that the problem was from jitter on the FEB CLK, due both to jitter in the incoming TTC signal and to jitter from the on-board TTCRx Usage of a “cleaner” TTC source allowed stable running of the FECT n We have in addition modified the FEB design to incorporate the QPLL, in order to provide more operating margin n n Stefan Simion will summarize the jitter studies, and solution implemented J. Parsons, FEB PRR, March 18/2004 8 8
Issues from FEB Final Design Review (June 18/02) 1. Concern over availability of rad-tol negative Vreg JQ 6 version from STm accepted for production n Meeting (this week) to hear results of STm’s 1000 hr lifetime test n Production quantity to be available next month n 2. Concern over availability of ADC (joint order with CMS/LHCb) n Delivery is well underway (2 lots accepted, 3 rd lot being irradiated now) 3. Concern over SMUX/GLink timing issues for optical link Delay chip added to provide additional margin n Most important timing parameters being measured for each SMUX chip n 4. Proposed use of Power-on-Reset scheme as per the CALIB board n Done 5. Diodes should be tested for radiation tolerance n Done for pre-selection, planned for production (see Stefan’s talk) 6. Develop further plan for traceability of components via serial numbers n Done J. Parsons, FEB PRR, March 18/2004 9 9
FEB PRR Presentations n FEB results from FEC system test - Ioannis Katsanos n Timing/jitter studies - Stefan Simion n Summary of FEB radiation qualification - Stefan Simion n FEB production and QA/QC overview - John Parsons n FEB HASS test - Gustaaf Brooijmans n FEB test setup at LAL - Dirk Zerwas n FEB test setup at BNL - Hucheng Chen n Summary - John Parsons J. Parsons, FEB PRR, March 18/2004 10 10
- Slides: 10