[Rgc_analysis] [EXTERNAL] RE: Raster Calibration Constants Plots

[Sebastian Kuhn]

Hi Derek,

nice results. Some observations:

1) Curiously, there is more run-to-run fluctuation in the x-parameters than the y-parameters - in BOTH run sets. Of course, that could be an accident.
2) P0 and P1 are strongly anticorrelated. It MIGHT be better to plot P0 + P1*meanadc instead of P0, where meanadc = 2300 for x and =1950 for y. This quantity would have a direct interpretation as the location of the CENTER of the raster ellipse, while the offset in ADC values is meaningless.
3) I do not agree entirely that the 2 run groups are consistent - for instance, the x-slope varies from 9.5 for the first set to 9.2 for the second. More surprisingly, the CCDB values for the OFFSETS (P0) are a bit different, while they are the same for the 2 slopes. I suspect this may have to do with the alignment procedure.
4) In any case, guess we could use a single set of calibration constants for the entire data set and wouldn’t be too far off. I still would prefer to use YOUR results to replace the CCDB values for Pass1 cooking - maybe we can cook one run for both choices and see if there is any difference in the Physics quantities (e.g., theta vs. vz for electrons). If not, it doesn’t matter; but if there IS a difference, I would argue that your results are better than what’s in the CCDB.
5) No matter how we will finally cook the Summer 2022 data, I hope that now that your framework exists, we can make sure that all calibration validations (Pass0) etc. for the remaining to parts of the experiment can be done with the raster parameters from our “new” method. To make sure, it might be good to check whether there is really an issue with using other targets than Carbon - ideally we want to use all but empty target runs.

- Sebastian

On Jan 9, 2024, at 7:57 PM, Holmberg, Derek via Rgc_analysis <rgc_analysis at jlab.org<mailto:rgc_analysis at jlab.org>> wrote:

Hi All,

I attached an updated version of the plots I had made. These contain the same content as the previous email, but I cleaned up the graphs so it’s easier to compare each of the graphs to each other.

The calibration constants for the 16292-16297 run period seem to be relatively consistent with those from the 16700-16704 period, but I wanted to see what you all thought about it.

I hope you all have a great week!

Sincerely,
Derek Holmberg

From: Holmberg, Derek via Rgc_analysis <rgc_analysis at jlab.org<mailto:rgc_analysis at jlab.org>>
Sent: Tuesday, January 9, 2024 4:56 PM
To: rgc_analysis at jlab.org<mailto:rgc_analysis at jlab.org>
Cc: Griffioen, Keith <griff at wm.edu<mailto:griff at wm.edu>>
Subject: [Rgc_analysis] [EXTERNAL] Raster Calibration Constants Plots

Hi Everyone,

Apologies for the delay in getting these back to you; I made a time-costly mistake when I was trying to run the vertexing code on the skims.

I plotted each of the calibration constants as a function of run number on each of the attached slides. Some of them were for the 16292-16297 run period, and the others were for runs 16700-16704.

The fit parameters are defined as such: V = P_0 + P_1*ADC

Where V is the reconstructed x or y vertex, ADC is the ADC signal from the raster, and P_0, P_1 are the fit raster calibration constants. I also included the current CCDB values as the green lines on the graphs. Please let me know if you have any questions or what you think. The constants appeared to be consistent within each run period.

Sincerely,
Derek Holmberg
<Raster_Calib_Plots_V2.pdf>_______________________________________________
Rgc_analysis mailing list
Rgc_analysis at jlab.org<mailto:Rgc_analysis at jlab.org>
https://mailman.jlab.org/mailman/listinfo/rgc_analysis

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/rgc_analysis/attachments/20240110/89e19062/attachment.html>