[Clas12_calcom] [EXTERNAL] Re: new timelines for RGB available - ECAL check - feedback needed

Thu Apr 13 12:41:55 EDT 2023

Hi All,

From the software perspective, I think all the tests that have been done on the different software versions and in particular the comparison done on MC (see today's software meeting agenda) have shown that the changes done to reconstruction do not lead to any loss of electrons and lead to a 10% increase of pi0s up to 6-7 GeV.  So running the two reconstruction version with the same constants will not help in understanding the differences between the two timelines. It can still be done for a run for a sanity check but I don’t think more than one run would be necessary. If doing so can help closing this chapter, I would be in favor.

To understand the timeline changes or to confirm that they are coming from the calibration constants changes, cooking one run with 8.7.0 and the two sets of constants will be more relevant in my mind, but again I think one run would be sufficient. There are already all indications from other studies that a few % loss of electrons mostly at the edges of the sectors arise from the lower sampling fraction. Anyway, that's up to CalCom to decide.

I would consider reprocessing all 6 validation runs in case the physics analysis of last week's samples leads to suspicious results that cannot be understood otherwise.

Best regards,
	Raffaella

> On 13 Apr 2023, at 17:53, silvia--- via Clas12_calcom <clas12_calcom at jlab.org> wrote:
> 
> Thanks, Cole.
> I'd just like not to do too many cookings and spend too much time looking
> at the data if there may be still other adjustments to come on
> calibrations or other CCDB constants.
> If you and CalCom think this is the best that can be done with ECAL, from
> both the gain calibration and the SF parametrization, for electrons and
> photons, we can launch the test cookings.
> Another question: I noticed that while the timing for pi- is consistent
> with the previous cooking/calibrations, the timing for pi+ is now out of
> specs for most runs. Is this something that we should be concerned of or
> not?
> Thanks!
> Best regards,
> Silvia
> 
> 
>> Silvia,
>> 
>> In my view it is more important to look at 8.5.0 vs 8.7.0 with the current
>> calibration. There is already a SF parameterization in place for the first
>> and second half of RGB. In any case I donâ€™t believe the PMT gains are
>> stable below the 2% level, due to CAEN HV shifts and luminosity dependence
>> (beam current ramping up and down or off for long periods), in addition to
>> the overall slow gain drifts with time. With this calibration the drift
>> was reduced from 3% to 2%.
>> 
>> Cole
>> 
>>> On Apr 13, 2023, at 9:41 AM, silvia--- via Clas12_calcom
>>> <clas12_calcom at jlab.org <mailto:clas12_calcom at jlab.org>> wrote:
>>> 
>>> Dear all,
>>> new timelines for RGB S19 have been produced after the recent changes to
>>> CCDB for ECAL.
>>> They can be found at this link:
>>> https://clas12mon.jlab.org/rgb/pass0/v28.51/tlsummary/ <https://clas12mon.jlab.org/rgb/pass0/v28.51/tlsummary/>
>>> 
>>> I have made a comparison, restricted to ECAL, with the timelines done
>>> with
>>> the February cooking, please find my slides at the link:
>>> https://urldefense.proofpoint.com/v2/url?u=https-3A__box.in2p3.fr_index.php_s_ENtSbFexnzxfxS8&d=DwIFaQ&c=CJqEzB1piLOyyvZjb8YUQw&r=LNUyx07OPlPP7unJr9bKdw&m=8mgJGEUEQzZ4W9FyipfDVE6UURt0xzTx9OghWPgCsecPkUKICtQu5_hk0Nl2THTu&s=d8Mx6e5HUplxhLqEiIebachl1KrOyw3xxWtGXNW_eFM&e= <https://urldefense.proofpoint.com/v2/url?u=https-3A__box.in2p3.fr_index.php_s_ENtSbFexnzxfxS8&d=DwIFaQ&c=CJqEzB1piLOyyvZjb8YUQw&r=LNUyx07OPlPP7unJr9bKdw&m=8mgJGEUEQzZ4W9FyipfDVE6UURt0xzTx9OghWPgCsecPkUKICtQu5_hk0Nl2THTu&s=d8Mx6e5HUplxhLqEiIebachl1KrOyw3xxWtGXNW_eFM&e=>
>>> 
>>> There is an overall loss of electrons/trigger or roughly 2%, with the
>>> new
>>> reconstruction and CCDB.
>>> 
>>> I need feedback from Cole and CalCom in order to decide how to proceed.
>>> 
>>> Should we cook our 6 test runs to verify the performances? Or are
>>> further
>>> changes expected to happen to ECAL calibration constants or SF
>>> parametrizations (for both electrons and photons)?
>>> 
>>> In our RGB meeting of last week various possible tests were mentioned:
>>> - cook the 6 runs with the same CCDB timestamp (the old one or the new
>>> one, or both?) and the two versions of CJ (8.5.0 and 8.7.0)
>>> - cook the 6 runs with the same CJ and different CCDB timestamp.
>>> 
>>> What should we do?
>>> 
>>> Thanks a lot for your feedback.
>>> Best regards,
>>> Silvia
>>> 
>>> 
>>> _______________________________________________
>>> Clas12_calcom mailing list
>>> Clas12_calcom at jlab.org
>>> https://mailman.jlab.org/mailman/listinfo/clas12_calcom
>> 
>> 
> 
> 
> _______________________________________________
> Clas12_calcom mailing list
> Clas12_calcom at jlab.org <mailto:Clas12_calcom at jlab.org>
> https://mailman.jlab.org/mailman/listinfo/clas12_calcom <https://mailman.jlab.org/mailman/listinfo/clas12_calcom>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/clas12_calcom/attachments/20230413/710a1728/attachment-0001.html>