STAR QA Board

[videbaek <videbaek@bnl.gov>]

Hi Daniel,

interesting list. A couple of comments:

- Are the any injections runs that have been marked as good. I agree 
they should not be used but would be good to know.
- I agree it does not hurt much t throw out short runs, but if its only 
because they are short it indicates to me that the statistics errors are 
not taken properly into account when rejecting.
- I checked a few of the long runs in beginning of list. At least two 
had a peculiar bunch distributions, but I don't see how this could 
effect quality, online plot looks good. I might be good to reruns the 
auto QA with the obvious bad runs taken out, particular all injections 
runs and look at the summary plots on Friday.

best Flemming



On 2021-11-02 13:40, Daniel Cebra via STAR-QAboard-l wrote:
> Hello QA board,
>    I have followed up on item 2 of the "to do" list for the 19.6 GeV
> Bad Runs list - i.e. I have gone back and checked the 76 runs that
> Ashik had flagged as bad and looked them up in the Log Book and Run
> Log Browser. I have attached an excel file which contains an entry for
> each of these 76 runs and records the minbias events, HLTgood events,
> and the relevant comments from the Log Book.
>
>    There are a few findings:
> 1) 38 of the Runs flagged as bad were injection runs - I had thought
> that these had not been produced?
> 2) 14 of the flagged runs are short (i.e. <10 minutes). It is not
> uncommon to flag short runs as bad, but they don't have much data so
> it is not really an issue.
> 3) 15 of the flagged runs were good, long data runs for which there
> were no comments in the log book about why the run might be bad.
> 4) The remaining 9 runs had a variety of issues that were noted in the
> log and could explain why the runs were rejected.
>
>    I will follow-up with the 15 long data runs which appear outwardly
> to be good to try to determine why they were flagged as bad from the
> run-by-run QA.
>
> Regards, Daniel
>
> On Fri, Oct 29, 2021 at 10:45 AM 林挺 via STAR-QAboard-l
> <star-qaboard-l@lists.bnl.gov> wrote:
>
> > Hi all:
> >
> > A brief minutes for today's meeting:
> >
> > https://drupal.star.bnl.gov/STAR/event/2021/10/29/qa-board-meeting
> >
> > 1. 19.6 GeV Run-by-Run QA:
> >
> > Comments:
> >
> > 1). RunID 450 to 500, the jump in eta/phi plot likely due to the
> > RDO problems during this runing period.
> > 2). At the beginning of the run, the beam is not centered, so there
> > is shift in eta;
> >
> > Things to do:
> >
> > 1). Compare the bad run-list to those points that outside the
> > normal ranges, see if there is plotting issues;
> >
> > 2). Check the RunLog to find out the possible reasons for the bad
> > runs;
> >
> > 3). Check other variables like Q;
> >
> > 4). Change the tolerance from 5sigma to 3sigma, to check the
> > sensitivity of cuts.
> >
> > Conclusion:
> >
> > Need to do some checks, but the run list after this QA is good
> > enough for centrality.
> >
> > 2. Discussion: 1). Centrality group is fine with using DCA, but
> > will also monitor the sDCA. May change DCA to sDCA after having a
> > good understanding of this variable.
> >
> > 2). Yu Hu will update the code in Github.
> >
> > 3) We are 80% of the way through 14.6 GeV production. 14.6 GeV QA
> > will start ASAP after the data production is done.
> >
> > 4) For FXT run QA, need a few changes eg. need to change RefMult to
> > number of primary tracks, shift vr cuts. Each working group should
> > think about the changes on QA.
> >
> > Best,
> >
> > Ting _______________________________________________
> > STAR-QAboard-l mailing list
> > STAR-QAboard-l@lists.bnl.gov
> > https://lists.bnl.gov/mailman/listinfo/star-qaboard-l
> _______________________________________________
> STAR-QAboard-l mailing list
> STAR-QAboard-l@lists.bnl.gov
> https://lists.bnl.gov/mailman/listinfo/star-qaboard-l

--
Flemming Videbaek
senior scientist
videbaek @ bnl.gov
Brookhaven National Lab
Physics Department
Bldg 510D
Upton, NY 11973

phone: 631-344-4106
cell     :  631-681-1596

Attachment: 19.6 GeV Bad Runs List.xlsx
Description: application/vnd.openxmlformats-officedocument.spreadsheetml.sheet