The shift workers reported a suddenly trigger rated increase on 3rd arm (>10k) at 6:30am. I took a 10 min test on beam position and slow raster size. I found that if we move slow raster size back to what we used on Saturday and move the beam position down by 0.5 mm, it will help us control the 3rd arm rate to about 5k. But the spot_3 plots do not change much. I suggest taking production data at this setting but once we got a beam down for more than one hour, we should repeat the beam recovery procedure.