The tg.dp distributions of both L and R HRS are centered at zero when tuning p0 to 1.7028 GeV, about -1% from the real elastic energy. This may be due to an offset in the reconstruction. Run1821--1821 and 20975--20978 are all with this setting. First attachment is the tg.dp and tg.th plot.
After fixing this dipole component, I tuned the septa current. When decreasing to 380A, the tg.y is almost centered at zero, but tg.ph is still shifted. But I felt this current is too low to decrease more, so just stopped here. Second and last attachments are 480A and 380A plots, respectively.
Figure 1
Figure 2
Figure 3