2009 Run QA

Here I summarize the issuse that I found during my run by run QA of days 160-166. I also post the finished run list with comments and error codes. A description of the procedure and error codes used can be found here.

The final run list for days 160-166 can be found here.

Summary:

Day 160:

• Day 160 had 30 runs to check
• Starting with R10160023 there is a stuck bit in JP 6 in the HT input. No error code is used
• For runs R10160069-R10160071 the BSMD and BPRS are unusable. Error codes Px and Sx are used
• Starting with R10160072 BSMD RDO 7 is masked out. Error code Sc is used

Day 161:

• Day 161 had 19 runs to check
• For runs R10161022-R10161024 half the barrel seems to be dead. Error code Xx is used
• Starting with R10161025, a high fraction of events have TOF electronics errors. Error code Fc is used

Day 162:

• Day 162 had 22 runs to check
• By run R10162006 the BSMD has been fixed and the TOF electronics errors have been cleared
• Starting with R10162006 Endcap crate 6 has been taken out of the trigger, but is not masked out at L2. Error codes Ec and Ej are used. (Shift log indicates crate 6 taken out for R10161055, which is not listed in the run list)
• R10162028 stopped due to many TOF bunchID shift errors. Error codes Fx and Do are used
• R10162029 TOF is taken out of run
• R10162030 TOF is put back into run

Day 163:

• Day 163 had 16 runs to check
• Starting with R10163007 the BSMD is out of the runs
• Run R10163011 has a completion status of Run Crash and no QA plots are available, therefore the error code is left as ??
• Starting with R10163048 the BSMD is back in the runs
• Starting with R10163048 endcap crate 6 is now masked out at L2. It is still out of the trigger. This state persists through the end of my week. Error codes Ec and Ej are still used

Day 164:

• Day 164 had 28 runs to check
• For run R10164034 Endcap MAPMT box 5S3 is 90% dead. Error code Mb is used
• For run R10164078 large chunks of the BPSD appear to be stuck on. Error code Px is used

Day 165:

• Day 165 had 49 runs to check
• Runs R10165011, 13, and 14 were taken without the endcap
• Starting with run R10165036 ESMD crate 75 has crate ID errors
• Starting with run R10165041 all events have TOF electronics errors. Error code Fc is used
• By run R10165046 the TOF electronics errors have been cleared
• For runs R10165047, 48, and 54 half of the barrel seems dead for part of the run. Error codes Bx and Bj are used
• For run R10165068 several endcap trigger patches appear to be dead. Error code Ex was used

Day 166:

• Day 166 had 28 runs to check
• Starting with run R10166012 BTOW crate 0x1A (softID 580-660) appears dead. Error code Bc used
• Starting with run R10166021 all events have TOF bunchID shift error. Error code Fc used
  
• Starting around run R10166022 TPX dead time is 100% at the start of runs leading to corruption in the Endcap
• Run R10166025 the TOF is taken out of the run and the bunchID shift errors cleared
• Starting with run R10166052 ESMD crate 75 no longer shows any corruption
• Starting with run R10166060 the BTOW crate 0x1A problem is fixed
• Starting with run R10166060 all events have 2 TOF bunchID shift errors. Error code Fc is used
• Starting with run R10166067 TOF bunchID shift errors are cleared

Questions on Error Status

• Runs R10161022-024: Oleg recommended that these runs be marked bad because of barrel problems, however the endcap looks fine (except for the stuck bit problem). Should these be marked Xx or should the barrel just be marked bad
• Run R10165068: This is a zdc_polarimetry run and sever endcap trigger patches look dead. Is it worth keeping the run with bad endcap data
• Runs in the _noendcap configuration have odd barrel L0 spectrum. The show prominant horizontal bands (L0 plots P6). Examples are runs R10165072, R10166051, R10166054, among others. Is this  a problem
• Runs R10166051 - R10166060 alternate between regular production and _noendcap and successful, questionable and bad. I want to make sure I have marked these correctly
• Run R10166063 is marked as a run crash and has a long period of TPX dead time at end of run, but everything else looks fine. Should this still be marked bad