Effect of PS leakage magnetic field on KEKB accelerator

It was found on April, 1999 that as beam of the KEKB accelerator oscillates under a time cycle, a stable beam collision could not be carried out. However, observed beam oscillation was remarkable in vertical direction of positron ring but was not so large in horizontal direction of both beams and in vertical direction of electron ring. By analysis of logging data and so forth of the KEKB accelerator it was found that frequency of the oscillation was about 0.5 Hz similar to pattern frequency of the PS accelerator and that it formed only when operating with PS main ring. As mechanism receiving power near the beam collision point was unknown yet, it was found that it was due to leakage magnetic field from source cable for electromagnet and preliminary electromagnet for monitoring of PS main ring to receive power under M5 source room. Effect of these magnetic fields was 280 mG-m in integral value of the field, and 2.4 micro-rad in kick angle. As magnitude of the beam oscillation at collision point was about 1 to 1.2 micron (peak to peak) at the first observation, it reduced to about its half after rolling into a Belle detector and was controlled further to 1/3 to 1/4 by countermeasure around the M5 room. (G.K.)