Note on preliminary cost estimate for an integrated tracker for the STAR experiment at RHIC Gerrit van Nieuwenhuizen (MIT), Ernst Sichtermann (LBL) and Bernd Surrow (MIT) August 2, 2004 Abstract The following document provides a first preliminary cost estimate for an integrated tracker for the STAR experiment at the Relativistic Heavy-Ion Collider (RHIC) at Brookhaven National Laboratory (BNL). This includes besides an overview of the various cost items in particular a funding profile starting from fiscal year 2005 until fiscal year 2009 to achieve the completion of an integrated tracker for the STAR experiment at RHIC in a timely fashion. Achieving such a funding profile will be crucial to continue with a competitive physics program of the STAR experiment by the end of this current decade for the relativistic-heavy ion program as well as for the polarized proton-proton program at RHIC at BNL. 1
Contents 1 Introduction 3 2 Preliminary cost estimate of the inner silicon barrel detector 6 3 Preliminary cost estimate of the inner forward silicon disk system 7 4 Preliminary cost estimate of the outer forward GEM detector 7 5 Funding profile for fiscal year 2005 until 2009 8 2
1 Introduction Layout of STAR integrated tracking upgrade y (cm) 200 eta=1.0 150 eta=-1.0 eta=1.09 100 eta=2.0 50 0-50 -100-150 Silicon disk area (m 2 ): 1: 0.14 2: 0.24 3: 0.39 (r,z): (8.5,24)cm(r,z): (11,36)cm(r,z): (14,48)cm Total: 0.77 Silicon barrel area (m 2 ): 1: 0.15 2: 0.34 3: 0.60 (r,z): (10,24)cm (r,z): (15,36)cm (r,z): (20,48)cm Total: 1.09 EEMC GEM tracker area (m 2 ): (rin,rout): (75,200)cm Total: 10.8 Pixel area (m 2 ): 1: 0.02 2: 0.05 (r,z): (1.5,20)cm (r,z): (4,20)cm Total: 0.07-200 -300-200 -100 0 100 200 300 z (cm) Figure 1: Conceptual layout of the STAR integrated tracking upgrade (Y Z view). The study of heavy flavor production in Au-Au collisions as well as in polarized protonproton collisions in addition to the study of W production in polarized proton-proton collisions will require an upgrade of the STAR inner and forward tracking system. This program has been documented in the STAR Decadal Upgrade plan and recently presented at the DOE RHIC program review (June 30 - July 1, 2004) and at the NASAC subcommittee review on Relativistic Heavy Ions (June 2-6, 2004) at Brookhaven National Laboratory. A side view (Y -Z view) of the conceptual layout of the STAR integrated tracking upgrade is shown in Figure 1. A close-up view of the inner tracking system is shown in Figure 2. The inner layers are formed by a pixel detector based on two layers of Active-Pixel Sensors (APS) followed by a new silicon tracker in the pseudo-rapidity-range of 1 < η < 1. During a recent internal STAR review of the Silicon-Vertex-Tracker (SVT) it has been concluded that the SVT should not be regarded as a pointing device for the STAR pixel detector resulting in the recommendation to consider the design of a new pointing device linked to the pixel detector upgrade project. This together with the upgrade of the forward tracking system forms the basis of the STAR integrated tracking upgrade. 3
Layout of STAR integrated tracking upgrade y (cm) 40 eta=1.0 eta=-1.0 eta=1.09 20 eta=1.7 eta=2.0 0-20 -40-40 -20 0 20 40 z (cm) Figure 2: Conceptual layout of the inner STAR integrated tracking upgrade (Y Z view). The forward part as shown in Figure 1 consists of three inner silicon disks and a large area forward GEM tracking detector in front of the STAR Endcap Electromagnetic Calorimeter. The production of W bosons provides an ideal tool to study the spin-flavor structure of the proton. W bosons are produced in ū d (u + d) collisions and can be detected through their leptonic decay. Forward scattered e (+) tagged in the STAR EEMC (1.09 < η < 2) off the incoming polarized proton beam moving toward (away) from the STAR EEMC, yield a purity for W (+) coming from ū d (u + d) quarks of about 98% (75%). The discrimination of ū d (u + d) quark combinations requires distinguishing between high p T charged leptons through their opposite charge sign which in turn requires precise tracking information. This forms the main motivation for the forward tracking upgrade. The target date for the first installation of part of the integrated tracker is prior to the foreseen long Au-Au run at the end of this decade. This would include the installation of the APS pixel detector together with a new mechanical support system and a minimal barrel system (to be determined from on-going simulation work. This would be then followed by a completion of the inner forward system based on silicon detectors as shown in Figure 2 using a disk configuration together with a large area GEM tracker in front of the STAR EEMC: Stage 1: Installation of STAR APS pixel detector together with a minimal new barrel tracking detector based on silicon technology 1 < η < 1) (Heavy Flavor Physics) Goal: Proposal by summer 2005 4
Installation of new inner tracking system prior to foreseen long Au-Au run at the end of this decade (2008) Stage 2: Upgrade of the forward tracking system (1 < η < 2) (W physics) Goal: Proposal by summer 2006 Installation of forward system prior to first long polarized pp production at s = 500 GeV (2009) The following three sections will provide a first cost estimate for the inner silicon barrel system, the inner silicon forward disks and the forward GEM tracking detector. It is understood that this is a preliminary cost evaluation which will be replaced by a refined cost analysis for the actual proposal of each proposal stage. Various details on the final layout such as the number of silicon barrel layers, the length and radius of each barrel layers as well as the the inner and outer radii and the distance to the nominal interaction point of the forward disks is subject to on-going simulation work. The last section provides a funding profile starting from fiscal year 2005 until fiscal year 2009 to achieve the completion of an integrated tracker for the STAR experiment at RHIC in a timely fashion. 5
2 Preliminary cost estimate of the inner silicon barrel detector Item Design A Design B Remarks Amount k$ Amount k$ Sensors 894 894 1392 1392 $1000/sensor Sensor R&D 100 100 $50k times 2 types Hybrids 260 130 464 232 $500/berillia substrate thin film Hybrid R&D 50 50 APV25 chips 4470 120 6960 174 $25/chip Cables 260 130 464 232 $500/low mass cable Cable R&D 50 50 FEE 572160 600 890880 900 $1/channel, in house R&D Integration FEE/DAQ 100 100 Power Supply 100 100 Power and bias supplies Cooling 200 200 Under-pressure water cooling Mechanics 1000 1000 Low mass, in house R&D Misc. items 100 100 Total 3574 4630 No contingency and overhead Table 1: Cost estimate for the inner silicon barrel system. This estimate is based on a sensor size of 40 cm 2 and one stereo pair per layer. Design A refers to system of 3 layers with radii of 70 mm, 150 mm and 170 mm whereas Design B refers to a system with radii of 100 mm, 150 mm and 200 mm. The following comments have to be taken into consideration for the cost estimate of the silicon barrel system: The exact layout is not finalized. The occupancy and required performance need to be still resolved through on-going simulation work. For the time being, an inner silicon barrel system of three layers covering the pseudo-rapidity region of 1 < η < 1 is assumed. A conservative design would then consist of silicon strip sensors with each layer consisting of stereo pairs. The size of the silicon sensors, quantified by the the strip length, will determine the occupancy. In principle, one would like to push the first layer as close as possible to the pixel layer. If one restricts the design to one (or rather 2 because of the required stereo angle) sensor type then the closest one can get is to a radius of 70 mm, below that the occupancy for central Au+Au collisions will be above 10%. It is foreseen to use the APV25-S1 chip which leads to a strip pitch of about 50 µm. It has been shown that by introducing an additional floating strip between the active strips, sub-10 µm resolutions are feasible. If a 90 stereo angle is used, each layer would then allow a space point resolution at the level of sub-10 µm precision. This would then result in a double metal sensor design. 6
The current estimate relies heavily on the PHOBOS experience gained during the silicon detector design and construction. The cost estimate does not include any contingency, overhead and any cost for personnel. The last cost item are expected to be already covered by existing personnel at participating institutes. Table 1 provides an overview of the various cost items for the inner silicon barrel system. This estimate is based on a sensor size of 40 cm 2 and one stereo pair per layer. Design A refers to system of 3 layers with radii of 70 mm, 150 mm and 170 mm whereas Design B refers to a system as shown in Figure 1 and 2 with radii of 100 mm, 150 mm and 200 mm. 3 Preliminary cost estimate of the inner forward silicon disk system The comments made in the previous section on the assumption of the preliminary cost estimate applies as well to the following discussion of the inner forward silicon disk system. The respective cost items are shown in Table 2 based on the conceptual layout shown in Figure 1 and 2. It is assumed that the main part of the mechanical support structure is in place after the Stage 1 installation. Item Disk design Remarks Amount k$ Sensors 675 675 $1000/sensor Sensor R&D 100 $50k times 2 types Hybrids 196 98 $500/berillia substrate thin film Hybrid R&D 25 APV25 chips 3370 85 $25/chip Cables 260 130 $500/low mass cable Cable R&D 25 FEE 431360 450 $1/channel, in house R&D Integration FEE/DAQ 100 Power Supply 100 Power and bias supplies Cooling 100 Under-pressure water cooling Mechanics 300 Low mass, in house R&D Misc. items 100 Total 2288 No contingency and overhead Table 2: Cost estimate for the inner forward silicon disk system. 4 Preliminary cost estimate of the outer forward GEM detector The cost estimate of the outer forward GEM detector is based on the conceptual layout shown in Figure 3 using individual triple-gem chambers. Each considered cost item is shown in Table 3. 7
y (cm) 200 150 100 50 0-50 -100-150 -200 89 88 87 64 86 63 85 62 43 84 61 42 41 25 83 60 40 24 82 59 10 STAR EEMC GEM tracker 23 39 9 81 58 22 38 8 80 57 21 37 7 79 56 20 90 65 44 26 11 Ring 1: Modules: 24 2 14 29 47 68 Ring 2: Modules: 30 91 66 45 27 12 Ring 3: Modules: 36 1 13 28 46 67 Ring 4: Modules: 42 158 137 119 104 92 Ring 5: Modules: 50 103 118 136 157 182 159 138 120 105 93 102 117 135 156 181 Total: 182 160 106 94 101 139 121 116 134 155 180 95 100 161 122 107 115 140 96 99 133 154 179 108 97 98 114 162 141 123 109 124 110 111 112 113 132 153 178 131 163 142 125 143 126 127 128 129 130 152 177 164 151 144 165 145 146 147 148 149 150 176 175 166 174 167 168 169 170 171 172 173 36 6 78 55 19 35 5 77 54 18 4 34 76 53 17 3 33 16 75 52 32 15 74 51 31 73 50 30 72 49 48 71 70 69-200 -150-100 -50 0 50 100 150 200 x (cm) Figure 3: Conceptual layout of the outer forward GEM tracker in front of the STAR EEMC based on individual triple-gem chambers. 5 Funding profile for fiscal year 2005 until 2009 Based on the cost estimates provided in the last three sections, a funding profile has been developed from fiscal year 2005 until fiscal year 2009 to achieve the completion of an integrated tracker for the STAR experiment at RHIC in a timely fashion. Achieving such a funding profile will be crucial to continue with a competitive physics program of the STAR experiment by the end of this current decade for the relativistic-heavy ion program as well as for the polarized proton-proton program at RHIC at BNL. A preliminary funding profile is shown in Table 4 and 5. 8
Item Disk design Remarks Amount k$ GEM chamber mechanics 200 100 $500/chamber GEM foils 900 180 $200/GEM foil Hybrids 728 364 $500/berillia substrate thin film Hybrid R&D 25 APV25 chips 1456 40 $25/chip Cables 260 130 $500/low mass cable Cable R&D 25 FEE 186368 190 $1/channel, in house R&D Integration FEE/DAQ 100 Power Supply 100 Power and bias supplies Cooling 30 Air flow system Mechanics 300 Low mass, in house R&D Misc. items 100 Total 1684 No contingency and overhead Table 3: Cost estimate for the outer forward GEM detector based on individual triple-gem chambers. 9
Fiscal year Required funding k$ Remarks 2005 50 Barrel Sensor R&D batches 25 Barrel Hybrid R&D batches 40 Barrel misc. items Total FY05 115 2006 50 Barrel Sensor R&D batches 25 Barrel Hybrid R&D batches 50 Barrel Prototype cables 1392 Barrel Sensor order 232 Barrel Hybrid order 174 Barrel APV25 chip order 500 Barrel mechanics 40 Barrel misc. items Total FY06 2463 2007 500 Barrel mechanics 1532 Barrel cables, FEE, Integration FEE/DAQ, Power supply and Cooling 100 Forward disk Sensor R&D batches 25 Forward disk Hybrid R&D batches 25 Forward disk Prototype cables 375 Forward disk Sensor order 50 Forward disk Hybrid order 45 Forward APV25 chip order 150 Forward mechanics 60 GEM foil order 25 GEM Hybrid R&D batches 25 GEM Prototype cables 120 GEM Hybrid order 15 GEM APV25 chip order 100 GEM chamber mechanics 50 Forward disk misc. items 20 Barrel misc. items Total FY07 3217 Table 4: Preliminary funding profile starting from fiscal year 2005 until fiscal year 2007. 10
Fiscal year Required funding k$ Remarks 2008 300 Forward disk Sensor order 48 Forward disk Hybrid order 40 Forward APV25 chip order 150 Forward mechanics 880 Forward disk cables, FEE, Integration FEE/DAQ, Power supply and Cooling 60 GEM foil order 120 GEM Hybrid order 15 GEM APV25 chip order 150 GEM mechanics 275 GEM cables, FEE, Integration FEE/DAQ, Power supply and Cooling 50 Forward disk misc. items 50 GEM misc. items Total FY08 2138 2009 60 GEM foil order 124 GEM Hybrid order 10 GEM APV25 chip order 150 GEM mechanics 275 GEM cables, FEE, Integration FEE/DAQ, Power supply and Cooling 50 GEM misc. items Total FY09 669 Total FY05-FY08 8602 Table 5: Preliminary funding profile starting from fiscal year 2008 until fiscal year 2009 and total costs using Design B for the silicon barrel system. 11