LumiCal mechanical design, integration with LDC and laser alignment Wojciech Wierba Institute of Nuclear Physics PAN Cracow, Poland IRENG07, SLAC, 17-21.09.2007

Outline of talk Mechanical design: Integration with LDC: dimensions, space for FE electronics, cables & cooling. Integration with LDC: LumiCal fixing, cables & cooling patch, beam pipe shape, opening scenario. LumiCal alignment: requirements, solutions, problems

Mechanical design Very challenging project Two half barrels to clamp LumiCal on the beam pipe 30 tungsten/silicon detector layers Odd/even planes rotated by 7.5 degree Total weight of ~250 kg (one LumiCal) Self supporting design of the tungsten structure „C” frames for supporting cables, cooling, alignment Movable support to open LumiCal – temporary support necessary Very challenging project Accuracy in Si sensors placement in order of a few micrometers

Mechanical design „C” frame Cables, cooling FE electronics Si detector Inner det. radius Inner mech. radius Barrel closing Movable support

LumiCal mech. dimensions Inner mech. radius (96 mm) has to be smaller by ~4 mm than inner det. (100 mm) radius because of straight line cutting of Si wafer. Outer det. radius is limited by the 6” Si wafer size.

LumiCal cross section Si detectors glued to ceramic support or directly to tungsten absorber. Si detectors glued to each tungsten plate (better energy resolution) or to every second tungsten plate on both sides (very precision machining of a half of tungsten plates. The gap between plates should be reduced from 1.5 mm to ~1 mm (or less) using bump bonding (better energy res.).

Water cooling Power dissipation ~20 W with cycled electronics Water used for cooling and temperature stabilization Underpreasure cooling system to avoid leakages

Integration with LDC How to fix LumiCal? (~250 kg) Where to put signal, power cables and cooling pipes? (~10-20 cm2) Beam pipe shape – minimize the material in front of LumiCal Opening scenario (Vertex, TPC maintenance)

LumiCal inside LDC LumiCal can be mounted to special support fixed to the square ‘construction’ pipe. Cables and cooling water pipes can be feed out in the gap between TPC and ECal endcap. Space for connectors Access to connectors LumiCal has to be centered on the outgoing beam.

Auxiliary systems Output digital links: max. 360 TP cables LVDS (estimated 180) Power cables: 6 power lines ~5-10 mm2 Control cables (+clock): a few TP cables + 2-4 coax cables Water cooling: ~4 copper pipes ~10 mm dia. Thermal insulation of the LumiCal to prevent temperature changes

Beam pipe centered on detector axis Solution not convenient for BeamCal and vacuum (to small pipe diameter)

Beam pipe centered on outgoing beam Better for BeamCal and vacuum, the LHCal has to be centered on the outgoing beam. The beam pipe diameter between LumiCal and BeamCal has to be discussed more carefully.

Opening scenario It is not necessary to open and take out LumiCal for Vertex and TPC maintenance. But, to pull out TPC, it is necessary to disconnect cables and cooling pipes. Need space for connectors and access to cables and pipes. For LumiCal maintenance (beam pipe exchange) it is necessary to install temporary support with movable cars. Disconnect cables and cooling pipes Unscrew halfbarrels and move both LumiCal parts out.

Two LumiCal’s alignment LumiCal x, y position with respect to the beam (incoming) should be known with accuracy better than ~700 µm (better ~100-200 µm) (LumiCal’s will be centered on outgoing beam) Distance between two LumiCal’s should be known with accuracy better than ~60-100 µm (14 mrad angle)

Alignment measurement based on beam pipe and BPM.s Reflective laser distance measurement – accuracy ~1-5 µm, resolution ~0.1-0.5 µm Mirrors glued to beam pipe Calibration of sensors procedure – detector push-pull solution (?) Calibration of sensors procedure after power fault (?) Beam pipe (well measured in lab before installing, temperature and tension sensors for corrections) with installed BPM (BPM’s also on outgoing beam?) Laser beams inside ‘carbon’ pipe (need holes, but possible)

LumiCal laser displacement measurement Two laser beams, one perpendicular, second with the angle of 45˚ to the CCD/CMOS sensor surface, are used to calculate the position shift. The CCD camera and lasers can be fixed to the LumiCal and beam pipe. We foresee 3 sensors (or more) to measure tilt of each LumiCal. Six (?) laser beams from one to another LumiCal passing inside the ‘carbon support’ pipe can be used to measure the relative position shift (the method described above can be used) and the distance between two LumiCal’s (very challenging, not solved yet). The measurements were performed on a proof-of-principle basis and achieved the accuracy of about ± 0.5 μm in x, y and ± 1 µm in z direction. Thermal stability of the prototype is ~0.7 µm/ºC

Summary Very challenging project: precisely positioned Si sensors (inner radius accuracy < ~4 µm), x & y alignment with respect to the beam < ~700 µm, distance between Calorimeters < ~100 µm, tilts < ~10 mrad. Alignment is a crucial task: distance between LumiCals (z) should be known better than ~100 µm over 4.5 m. A lot of problems to be solved, a lot of work to be done…