Make New Entry, Make Followup Entry
User name X. Zheng
Log entry time 00:15:14 on November 12, 2009
Entry number 299064
Followups:
keyword=PVDIS DAQ - Right arm spectrum after the access at 8pmToday we were baffled by the wierd PID spectrum on the left arm: That the lead glass 2D histogram and the Cherenkov histogram do not look like clean electrons. There are events (about 5%) that look like pions. It is not clear whether the background seen in the cerenkov histogram is a software artifact caused by cherenkov ADC overflow. And the origin of the background in the PS vs. SH plot is completely unknown (not even a guessed answer). While on the right arm we do not see this behavior.
During the access at 8pm, the analog sum module PS740 were replaced by LeCroy428F. The reason is because PS740 has nonzero offsets that drifts over time, up to several 10s of mV. This was seen from today's owl shift data. While LeCroy 428F do not show the same bahavior. After swapping modules, however, the right arm PID shows the same wierd behavior as left arm. Below are two sets of plot for global electron and pion triggers, from run 4921 (yesterday) and 4941 (today, first run after the swap).
Interestingly, even though we did not touch the Cherenkov (did we???), the cerenkov spectrum with the electron trigger cuts also changed.
It looks like some hardware change during the access has caused this to happen. Could it be subtle timing change due to different response time of the modules? Could it be some effects of the module itself? Or could it be my software problem? (Some double-check would be nice, maybe I am imagining things). I am out of wits here and will continue thinking about it tomorrow.
Run 4941
Run 4921
A copy of this log entry has been emailed to: rom,rsubedi,deng,dwang,pkpan