[Halld-cpp] Fwd: [EXTERNAL] BCAL z calibration for CPP/NPP

[Elton Smith]

I am forwarding these email exchanges primarily from Naomi regarding the BCAL z calibration for CPP/NPP.

Begin forwarded message:

From: Naomi Jarvis <nsj at cmu.edu>
Subject: Re: [EXTERNAL] BCAL z calibration for CPP/NPP
Date: February 23, 2024 at 11:17:37 AM EST
To: Elton Smith <elton at jlab.org>
Cc: Rory Miskimen <miskimen at physics.umass.edu>, David Lawrence <davidl at jlab.org>, Benedikt Zihlmann <zihlmann at jlab.org>, Ilya Larin <ilarin at jlab.org>, Sean Dobbs <sdobbs at jlab.org>, Mark Macrae Dalton <dalton at jlab.org>

Hi,

the monitoring data version browser is at https://halldweb.jlab.org/cgi-bin/data_monitoring/monitoring/dataVersions.py

Best regards,
Naomi.

On Thu, Feb 22, 2024 at 7:36 PM Elton Smith <elton at jlab.org<mailto:elton at jlab.org>> wrote:
Hi Naomi,

Thanks for passing on the sets of plots that illustrate the “BCAL issue” mentioned in the talks yesterday. I took a look at the dates of the monitoring launches and updates to the effective velocity, which I believe is one of the parameters that Mark determines during his calibration. See. screen shots below.  The relevant calibration for runs 100000-10999 (CPP/NPP) was on Aug 14, 2023.

Rough Sequence:

Calibration Mar 23, 2020 [prior GllueX calibration, still valid]

CPP Monitoring period 2022-05 Ver19- July 17, 2023

Calibration Jul 20, 2023 (var=bcal_cpp,  which I assume was not used for any monitoring launches)

Calibration Aug 14, 2023

GlueX Monitoring period 2023-01 Ver07- Jan 23, 2024

CPP Monitoring 2022-05 Ver25-  Jan 30, 2024

The following is my take, but I might be missing something :

It appears that the calibration on Aug 14, 2023 is the one causing the systematic differences between determinations of positions from the BCAL and from tracking. It seems that the [GlueX] calibration from Mar 23, 2020 is still valid for the GlueX period 2023-01 from Ver07 taken in January 2024. From this we can conclude that the BCAL did not change during CPP, as expected since the hardware has been untouched. The changes during CPP directly touched the target position, average track momentum (due to lower beam energy) and CDC (1/4 off). Therefore, as Mark and Naomi have speculated, it is likely Mark’s calibration procedure that relies on tracking that has affected the calibration constants in Aug 14, 2023.

I concur with Naomi with restoring calibrations to the Mar 23, 2020 set [or the precise calibrations used for Monitoring Ver19] and checking that they are ok after updating the CDC gains. I will note that Ilya has already said that he would like to revisit the BCAL calibrations in detail, but this should not delay the REST file production. This could be done together with investigating the BCAL position parameters.

BTW, where can we find the detailed description of each Monitoring Launch? (I’m loosing track of where everything is).

Thanks, Elton.




[Version_dates.png]


[effective_velocities.png]


On Feb 22, 2024, at 11:43 AM, Naomi Jarvis <nsj at cmu.edu<mailto:nsj at cmu.edu>> wrote:

Hi,

As you heard yesterday, Mark has been having trouble with the BCAL position calibrations, he adapted his script to cope with the different target position and missing CDC/4, and must have inadvertently introduced a bug, as the results are worse.  There is a set of calibrations from October that looks okay to me, the non-expert.  I wondered if these would be good enough for your REST production.  I will paste a selection of relevant plot links below so that you can see the problem.  Mark is more than welcome to add better-informed comments.

Example plot from GlueX 2023, not affected by the bug
https://halldweb.jlab.org/work/halld2/data_monitoring/RunPeriod-2023-01/mon_ver07/Run120400/HistMacro_Matching_BCAL.png

Plot from CPP/NPP launch 19, the last 'good' one before the problem arose
https://halldweb.jlab.org/work/halld2/data_monitoring/RunPeriod-2022-05/mon_ver19/Run101000/HistMacro_Matching_BCAL.png

Plot from CPP/NPP launch 25, the most recent
https://halldweb.jlab.org/work/halld2/data_monitoring/RunPeriod-2022-05/mon_ver25/Run101000/HistMacro_Matching_BCAL.png

My time-saving proposal is to retrieve & reinsert the CCDB constants used for ver 19, run another monitoring launch, and use that to update the CDC gains, which were disturbed by BCAL's recalibrations.

Of course it would be nicer to find and fix the bug, and then see if the intended improvements really do help.  I am asking if you need this.

Best regards,
Naomi.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-cpp/attachments/20240228/2c172566/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Version_dates.png
Type: image/png
Size: 382211 bytes
Desc: Version_dates.png
URL: <https://mailman.jlab.org/pipermail/halld-cpp/attachments/20240228/2c172566/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: effective_velocities.png
Type: image/png
Size: 44980 bytes
Desc: effective_velocities.png
URL: <https://mailman.jlab.org/pipermail/halld-cpp/attachments/20240228/2c172566/attachment-0003.png>