Wire-scanner recentering experiment error estimates

In the wire-scanner recentering experiment, the effective center of a wire scanner is determined by turning the previous quadrupole magnet on and off. The beam is steered into the quadrupole such that it drifts to the same point somewhere on a second wire scanner downstream, for the quadrupole both on and off. With this steering, the beam is assumed to be centered at the quadrupole and first wire scanner. Approximate formulas and more exact numerical calculations show that for any incoming position at the quadrupole (not just the center), there is a corresponding incoming angle such that the beams for the quadrupole on and off will drift to the same point. Errors in the recentering experiment of 9 to 13 February 1977 due to the beams not converging to exactly the same point and due to the convergence point being off axis are estimated to be 0.8 mm or less. Errors from quadrupole angular misalignments are estimated to be 0.5 mm or less.