Performance evaluation and optimization of the luminosity detector ALFA

The startup of the LHC (Large Hadron Collider) has initialized a new era in particle physics. The standard model of particle physics has for the last 40 years with tremendous success described all measurements with phenomenal precision. The experiments at the LHC will test the standard model in a new energy regime. To normalize the measurements and understand the potential discoveries of the LHC experiments it is often crucial to know the interaction rate - the absolute luminosity. The ATLAS (A Toroidal LHC ApparatuS) detector will measure luminosity by numerous methods. But for most of the methods only the relative luminosity is measured with good precision. The absolute scale has to be provided from elsewhere. Therefore ATLAS plans to measure the flux of protons scattered under very small angles as this flux relates directly and with good precision to the absolute luminosity. This will be done by the ALFA (Absolute Luminosity For ATLAS) detector. The detectors will be positioned about 240 m from the interaction point to make it possible to measure sufficiently small scattering angles. To approach close enough to the beam (few millimeters) the detectors will be moved inside the beam pipe contained in Roman Pots with a secondary vacuum. The detector itself is build up from 500 µm squared scintillating fibers. To improve resolution the fibers are staggered. The fibers will be readout by MAPMTs (MultiAnode PhotoMultiplier Tubes ). To eliminate background two stations containing an upper and a lower detector will be positioned at each side of the interaction point. The combined tracks for 4 stations will give the flux and angle of the scattered particles, from which the luminosity can be calculated with good precision. It is essential that each scintillating fiber delivers enough light to make a clear signal. Therefore tests using standard electronics for readout of a MAPMT connected to a prototype detector have been performed. The setup was primarily used to determine the light yield of the scintillating fibers which was measured for both low energy electrons from a source and high energy muons from cosmic rays. After the method was verified an upgrade was made to custom electronics which allowed readout of a full detector. This made it possible to perform light yield measurements of all fibers by using cosmic muons. As the cosmic test used almost all the parts that eventually will be installed at LHC/ATLAS it thereby also provided a stability test for all components involved. To verify the performance in conditions as close as possible to the real use in the LHC, two detectors were installed in an ALFA station and tested in a beam. The test beam have the clear advantage that the particle rate is much higher than with cosmic rays and that the particles always comes from the same direction. This makes studies of resolution, cross talk etc feasible. At the test beam a high precision silicon pixel telescope was used to improve the resolution studies and make internal alignment of the detectors.