Performance of the Opal Si-w Luminometer at Lep I-ii

The LEP ee collider at CERN operated for more than a decade: in 198995 at center of mass energies close to the Z peak (LEP-I); in 1996-2000 at higher energies, up to 209 GeV (LEP-II). In the first phase a large number of Z events were collected, of the order of 5 × 10 events per experiment. To match the inherent precision of this data sample, the error on the integrated luminosity had to be better than 10. At LEP the relevant process for the luminosity measurement is Bhabha scattering at small angle, which delivers a counting rate higher than the Z event rate at resonance. The Bhabha angular spectrum falls like 1/θ, implying a high sensitivity to the definition of the minimum polar angle of the acceptance. For example an uncertainty δθ = 10 μrad (which in our configuration is equivalent to 25 μm in radius at the face of the detector) would give an unacceptable systematic error of 10. Precision luminosity measurement was thus a demanding task, dictated by interest in measuring absolute cross sections at the Z peak. In particular,