Lowering the vertical emittance in the ler ring of pep-ii

The low energy ring (LER) in the PEP-II B-Factory has a design emittance of 0.5 nm-rad in the vertical, compared to nearly 0.1 nm-rad for the HER ring. This was thought to be caused mainly by the "vertical step" of 0.89 m in the interaction region straight, where the LER beam after horizontal separation gets bend vertical so it sits on top of the HER in the rest of the ring. Since the program MAD does not easily reveal the location of the major emittance contribution, a program was written to calculate the coupled "curly H" parameter of mode 2 (mainly vertical) along z. Weighting it with the magnet bending revealed that the weak long bends inside the "vertical step" did less than 20% of the emittance growth. More than 80% comes from the last quarter of each of the adjacent arcs with strong bends. This is caused by the coupling cancellation of the solenoid starting already there with the skew quadrupoles SK5 and 6. By introducing additional skew quadrupoles in the straight instead of SK5 and 6 the emittance could be reduced by a factor of ten in simulations, but with very strong skews. Reasonable strong magnets might generate a workable compromise, since a factor of two in emittance promises 50% more luminosity in beam-beam simulations.