Fig 1 is the comparison between left and right arm PVDIS bcm asymmetry of u1,u3,d1,d3. The raw scaler counting is even convincing that there is just one scaler counting diff at most between left and right in all u1,u3,d1,d3.
Fig 2 is the comparison between the asymmetry of happex u1 and our left u1,u3,d1,d3. The result shows the clear correlation already, and it is because we haven't done pedestal subtraction of bcm scaler yet that leads to a little fat pattern you see here. Usually, when there is no beam, it will still have several hundred hz reading from scaler, and u3,d3 is roughly three times as large as u1 and d1.
So I think the previous noncorrelation problem is basically solved. It is because the bcm 740 fanout in the counting house that brought us trouble. After I set up the new module to fanout them, both test pulse result and real beam result is consistent and good.
PS: I plug back the d3 to its original d3 path, and restore the d10 path
to avoid confusion. Please, Xiaochao decide we should find reason of
missing d3 on right arm in next access which is probably just cable
unpluged or we use original d10 path, which is not very useful, to go
through d3 signal.
A copy of this log entry has been emailed to: xiaochao, rom, rsubedi, deng, dwang, pkpan
Figure 1
Figure 2