Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Options for Optical Links at SLHC Francois Vasey, CERN/PH-MIC with input from ATLAS and CMS colleagues Margins of existing ATLAS and CMS TK OL Options for OL upgrades Some proposals The joint optoelectronics working group Conclusion 28-Sep-06 [email protected] 0 Optical Links for the ATLAS and CMS Trackers ATLAS Digital links (40/80Mb/s) CMS Analogue readout links (40MS/s, 8bits) Digital control links (80 Mb/s) 17.5k Channels 40k+ 3k Channels VCSEL 850nm MM EELaser 1310nm SM 28-Sep-06 [email protected] 1 The Optical Link Margins (1: Tx) Tx Radiation Hardness Operating point shifts Thermal management becomes crucial Thermal resistance Annealing Compliance voltage BEFORE annealing Optical Output Power [W] 1000 100 10 1 0.1 0.01 0 20 40 60 80 100 120 140 80 1.0 60 0.8 Ith Eff of LD1 40 0.4 20 0.2 LHC 1.0 2.0 3.0 4.0 Fluence [10 n/cm ] 20MeV 15 28-Sep-06 [email protected] 0.6 2 2 Relative Efficiency [W/mA] Preliminary Threshold Ith [mA] Input Current [mA] The Optical Link Margins (2: Rx total) Rx Radiation Hardness Leakage current damage definitely appears to be ‘saturating’ Responsivity going to zero No significant annealing Ultimate limit reached 10 10 10 10 10 10 10 -3 leakage current (A) leakage current (A) -4 -5 -6 -7 10 10 10 10 -9 10 1 154 2 1.0 photocurrent at 100uW (uA) 0.6 0.4 0.2 LHC 1 -5 -6 -7 -8 -9 0.8 0.6 0.4 0.2 0.0 0.1 10 154 2 1 10 annealing time (hrs) fluence (10 n/cm ) [email protected] -4 1.0 0.8 0.0 0.1 -3 0.1 1 10 annealing time (hrs) 10 fluence (10 n/cm ) photocurrent at 100uW (uA) 10 -8 0.1 28-Sep-06 10 3 Optical Link Margins (3: Rx SEE) Rx SEE cross-section Bit rate dependent Incidence-angle dependent 28-Sep-06 [email protected] 4 Optical Link Margins (4: fiber) Fiber Radiation Hardness (0.05-0.1dB/m) To be reconfirmed 28-Sep-06 [email protected] 5 Optical Link Margins (5: link) Fiber Capacity CMS SISM: >>10GHz∙km Electronics ATLAS SIMM: 50-80 MHz∙km ATLAS GRIN: 1000 MHz∙km Binary 40/80MBps Analogue 100MHz Digital 80MBps 20cm from beam axis Geometry 18cm from beam axis 28-Sep-06 [email protected] 6 Margins for operation at SLHC (6: summary) Tx and Rx need careful study Fiber Generalized Bottleneck Geometry Radiation tolerance probably OK Capacity increase OK in SM, NOT OK in SIMM, OK in GRIN Electronics Are we reaching the limit? Rx SEE sensitivity to be mitigated Possibility to move away from beam axis to gain margin against radiation damage Good overall agreement between ATLAS and CMS observations 28-Sep-06 [email protected] 7 The Options for an Upgrade (1) Reuse the existing system Check margins carefully Boost analog link capacity by developing CODEC and MODEM Develop TRx, SERDES and CODEC See paper by P. Moreira Check compatibility and margins of legacy fiber and connectors ? Start from scratch See paper by S. Dris Reuse the fiber plant only Cod Ser Tx Cod/Mod Possibly using components identical to existing ones A combination of the above Rx Des Dec Dec/Demod Geometry dependent system implementation 28-Sep-06 [email protected] 8 Novelty The Options for an Upgrade (2) Start from scratch Reuse the fiber plant only Reuse the existing system components development Resources 28-Sep-06 [email protected] 9 Proposal 1: The ATLAS SCT Straw-Link (T. Weidberg, Oxford, Oct 05) 2500 links 2.5Gbps Start from scratch TTC FanOut M1 M2 M30 Clock Data MUX 28-Sep-06 PDA PDA LD [email protected] 10 Proposal 2: ATLAS Pixels K.K. Gan, Ohio State University Reuse existing cable plant 1m micro twisted pair, 8m SIMM fiber, 70m GRIN fiber, 1Gbps 1m40 TP 2Gbps 9m SIMM + 20m GRIN 0.65Gbps 1.3 Gbps 28-Sep-06 [email protected] 11 Proposal 3: The versatile bi-directional digital link GBT Chipset Opto 1 S. Marchioro, P. Moreira, CERN, 2005 Opto 2 3-6Gbps Versatile Opto Startbidi from or 3reuse fiber plant TRx scratch 28-Sep-06 [email protected] 12 Proposal 4: Flipped OE devices on SoS substrate Ping Gui, Jingbo Ye, SMU, 2005 Start from scratch or reuse fiber plant UTSi integrated photo detector UTSi integrated circuitry receiver circuitry VCSEL driver circuitry quad VCSEL array flip chip attachment quad PIN array 200 um transparent sapphire substrate (UTSi) MMF ribbon fiber 28-Sep-06 [email protected] 13 active CMOS layer The Joint Optoelectronics Working Group System FE 28-Sep-06 TRx1 Optical Link 1 TRx1 Rx2 Optical Link 2 Tx2 Tx3 Optical Link 3 Rx3 [email protected] BE 14 Joint Optoelectronics Working Group Joint Optoelectronics Working Group ResourcesSurvey Distribution Result 5 Series1 CERN Ljubljana Minnesota Ohio Oklahoma Oxford Polytechnique RAL SMU Strasbourg 8 7 6 5 FTE 4 3 2 1 0 11.4 13 institutes, 23FTE 2 development projects (SMU, CERN) Collaborative proposals in preparation 28-Sep-06 CERN [email protected] EU US 6.6 10.2 ATLAS 12.8 15 CMS Conclusions Options for Optical Links at SLHC are: Reuse existing system (open to CMS only) Reuse fiber plant only (open to CMS and ATLAS Pixels) Start from scratch (possibly using components identical to existing ones). Check margins of existing components thoroughly Investigate Laser and PIN diodes in depth Will we reach the total dose limit at SLHC? Draw a line of usability at SLHC, with safety margin. Investigate electronic mitigation techniques (total dose and SEE). For new components, concentrate on only very few options. Generic and versatile The joint optoelectronics working group has the potential to coordinate a few parallel projects across experiments. System level aspects to be addressed by the experiments. 28-Sep-06 [email protected] 16