LHC CMS Detector Upgrade Project Design of the

LHC CMS Detector Upgrade Project Design of the He and HB ODU M. Mc. Kenna, M. Wayne Notre Dame 1

LHC CMS Detector Upgrade Project ● Schedule Highlights The ODU project is done in two stages ● ● Preproduction of up to 32 ODUs (16 HB and 16 HE) ● Mechanical unit design for HB/HE completed by 26 -June-2014 ● Assembly of HE begins 28 -July-2014, assembly of HB begins 25 -Aug-2014 ● HE (HB) units at CERN and ready for mounting in RMs by 15 -Feb-2015 (15 -Mar-2015) Full production of 360 ODUs (180 HB and 180 HE) ● Design modifications based on test beam results completed by 04 -Sep-2015 ● Assembly (done in 5 batches) of HE begins 28 -Dec-2015, ends 18 -May-2016 ● Last batch of HE checked out at CERN, ready to mount on 12 -Sep-2016 ● Assembly (done in 5 batches) of HB begins 19 -May-2016, ends 10 -Oct-2016 ● Last batch of HB checked out at CERN, ready to mount on 8 -Feb-2017 2

LHC CMS Detector Upgrade Project Initial design assumptions ● ● Mechanical dimensions for both HE and HB ODU does not change from 2011 -2012 test beam designs Patch panel, brackets, etc. are not changed, cookie design accommodates different numbers of Si. PMs on HE/HB and new Si. PM package dimensions ● Threading of fiber pigtails changes accordingly ● HE ODU will have 48 channels ● ● ● Depth segmentation based on scheme developed by Vitaly Smirnov HB ODU will have 64 channels ● Depth segmentation based on 4 depth segments per tower ● Interleaving of back two segments for some redundancy Optical mixers – not sure if needed, would simplify design and fabrication if these are not necessary (more on this later) 3

LHC CMS Detector Upgrade Project First prototype of the HE ODU Up to 7 layers in depth 96 channels for each pair of ODU Threading a bit tricky but achievable With drawings and color-coding, can be done in “mass production” by students and techs But there a few issues to resolve 4

LHC CMS Detector Upgrade Project HE ODU design issues ● ● ● With the current design for depth segmentation, one ODU of each pair requires 49 Si. PMs while the other requires 47 Si. PMs ● Consequence of how cables bring in the fibers to the patch panels ● We propose combining L-1 with L 0, L 1, L 2 in tower 18 - open to other suggestions ● Will result in 2 unused Si. PM channels per pair of HE ODU The patch panel has a connector for “Layer 00” ● Second sigma loop in the 0 th layer (? ) ● Not clear if these fibers are actually brought to the ODU – eliminate? ● We propose threading fibers from Layer 00 together with Layer 0 ● There is no input for tower 17, L 0 – we will use L 00 instead In 44 of the 48 channels the Si. PM receives 4 fibers or less (2. 8 mm device), the other 4 channels receive 5 -7 fibers (3. 3 mm device) vs vs We propose this 5

LHC CMS Detector Upgrade Project Other HE ODU points, questions ● ● ● The “cookie” will have 44 holes with 2. 37 mm diameter (for 2, 3, 4 fibers) and 4 holes with 2. 87 mm diameter (for 6, 7 fibers) The holes for attaching the patch plate and bottom plate to the cookie holder have been moved to provide increased clearance for a v-groove needed for the polishing fixture Calibration inputs ● We propose 7 calibration channels with 8 fibers per channel ● Calibration fibers are 0. 25 mm clear PMMA, one fiber to each Si. PM with 8 extra fibers ● Will springs be used to press the cookie to the Si. PMs? ● Will black anodizing be required? ● ● If “yes” we need to know which areas to keep clear for electrical ground contact There are unused connector slots – do we need blank connectors to cover them up? 6

LHC CMS Detector Upgrade Project Optical Mixers Optical mixers are designed to spread large amounts of light across the Si. PM active area Effectively increases the dynamic range at the high end, but fast recovery time can also increase the dynamic range Do we need the mixers for HE and/or HB? We propose instrumenting half the preproduction ODU with mixers, test in the beam and on the bench, then make the final decision for full production Optical Mixer plate 4 mm thick – 3. 2 mm diameter fiber elements Note: overall design must account for additional thickness of mixer plate 7

LHC CMS Detector Upgrade Project HB ODU Design issues ● To our knowledge, no real study of the depth segmentation has been done yet – only design is the following: 4 segments in depth per tower Identical for towers 1 -15, tower 16 a bit different Interleaving of back two segments, then combined in the electronics ODUs will require 64 Si. PM each – 32 small, 32 large diameter Threading looks to be challenging because of the interleaving 8

LHC CMS Detector Upgrade Project HB ODU – 1 st attempts at threading This is doable but fibers are tight once the ODU is compacted to appropriate size The threading above was done using the same “hybrid” package as in the HE ODU Might be easier with 4 packages with all smaller diameter, 4 with all larger diameter 9