[Rgc_analysis] [EXTERNAL] Re: RGC Beam Charge Asymmetry & Scaler Bank QA

Sat Sep 24 13:34:33 EDT 2022

Hi Gregory,

Thanks for looking into this!

I think we know now that it's not a bug in RAW::scaler, but in how its contents is being interpreted and results in HEL::scaler.  Some assumptions were made in previous years, that were probably/maybe correct at that time, but things have evolved ...

Others have been looking into this too, and there's a pull request in that should address it (along with various other scaler readout quirks):  https://github.com/JeffersonLab/clas12-offline-software/pull/949

Can you point me to a particular run/file where you see a large effect from your particular issue?  I've tested on a few but would like to test on that one too.

-Nathan

On Sep 18, 2022, at 6:52 PM, Gregory Matousek <gregory.matousek at duke.edu<mailto:gregory.matousek at duke.edu>> wrote:

Hi Maurik,

This is great to hear. I looked deeper into the FCup issue today and stumbled across this RAW::scaler bug you are mentioning. It seems that, arbitrarily, the channel id's for the 33ms window and 500microsecond window get swapped mid-run, leading to a ~70x drop in FCup reported per scaler bank entry. I corrected for this in my attached slideshow, and I intend to talk about it more in depth on tuesday.

https://cc.jlab.org/system/files/users/gmat%40jlab.org/Presentations/Bug%20Fixes%20in%20RAW__Scaler%20Bank.pdf<https://cc.jlab.org/system/files/users/gmat@jlab.org/Presentations/Bug%20Fixes%20in%20RAW__Scaler%20Bank.pdf>

Best,
Gregory
________________________________
From: Maurik Holtrop <holtrop at jlab.org<mailto:holtrop at jlab.org>>
Sent: Sunday, September 18, 2022 6:31 PM
To: Gregory Matousek <gregory.matousek at duke.edu<mailto:gregory.matousek at duke.edu>>
Cc: Sebastian Kuhn <kuhn at jlab.org<mailto:kuhn at jlab.org>>; pilleux at jlab.org<mailto:pilleux at jlab.org> <pilleux at jlab.org<mailto:pilleux at jlab.org>>; Mohammad Hattawy <hattawy at jlab.org<mailto:hattawy at jlab.org>>; Anselm Vossen, Ph.D. <anselm.vossen at duke.edu<mailto:anselm.vossen at duke.edu>>; Silvia Niccolai <silvia at jlab.org<mailto:silvia at jlab.org>>; Florian Hauenstein <hauenst at jlab.org<mailto:hauenst at jlab.org>>; Harout Avakian <avakian at jlab.org<mailto:avakian at jlab.org>>; Daria Sokhan <daria at jlab.org<mailto:daria at jlab.org>>; rgc_analysis at jlab.org<mailto:rgc_analysis at jlab.org> <rgc_analysis at jlab.org<mailto:rgc_analysis at jlab.org>>; Nathan Baltzell <baltzell at jlab.org<mailto:baltzell at jlab.org>>; Raffaella De Vita <devita at jlab.org<mailto:devita at jlab.org>>; Serguei Boiarinov <boiarino at jlab.org<mailto:boiarino at jlab.org>>
Subject: Re: [Rgc_analysis] [EXTERNAL] Re: RGC Beam Charge Asymmetry & Scaler Bank QA

Hello Gregory, Noémie, Sebastian,

I am sorry I could not make the 2:15 meeting today.

I have been going back and forth a bit with Nathan about the HEL::scaler bank. The problem seems to be in the decoding of the EVIO bank to the HEL::scaler bank and the issue where the bit that should indicate the difference between the stable and the settle period is not reliable. Nathan pointed out that you can look at the RAW::scaler bank, which contains both the stable and settle information, so you can untangle where the “bit flip” happens and pick the correct one, but this will be a bit more coding than just looking at the HEL::scaler bank.

It is too early to come to a conclusion, but from some cursory looks at the data it appears that we will be able to write some code in the JAVA tools that will correct for this problem. Nathan is already working on some of the required changes to the code.

Best,
Maurik

On Sep 14, 2022, at 1:22 PM, Gregory Matousek <gregory.matousek at duke.edu<mailto:gregory.matousek at duke.edu>> wrote:

Yes, I will be there. I'm putting together an abridged talk which has some interesting conclusions. Hopefully I'll be able to explain things coherently enough.

See you at 2:15pm
________________________________
From: Sebastian Kuhn <kuhn at jlab.org<mailto:kuhn at jlab.org>>
Sent: Wednesday, September 14, 2022 11:41 AM
To: pilleux at jlab.org<mailto:pilleux at jlab.org> <pilleux at jlab.org<mailto:pilleux at jlab.org>>
Cc: Gregory Matousek <gregory.matousek at duke.edu<mailto:gregory.matousek at duke.edu>>; Anselm Vossen, Ph.D. <anselm.vossen at duke.edu<mailto:anselm.vossen at duke.edu>>; rgc_analysis at jlab.org<mailto:rgc_analysis at jlab.org> <rgc_analysis at jlab.org<mailto:rgc_analysis at jlab.org>>; Harut Avagyan <avakian at jlab.org<mailto:avakian at jlab.org>>; Mohammad Hattawy <hattawy at jlab.org<mailto:hattawy at jlab.org>>; Silvia Niccolai <silvia at jlab.org<mailto:silvia at jlab.org>>; Raffaella De Vita <devita at jlab.org<mailto:devita at jlab.org>>; Daria Sokhan <daria at jlab.org<mailto:daria at jlab.org>>; Florian Hauenstein <hauenst at jlab.org<mailto:hauenst at jlab.org>>; Nathan Baltzell <baltzell at jlab.org<mailto:baltzell at jlab.org>>; Sergey Boyarinov <boiarino at jlab.org<mailto:boiarino at jlab.org>>
Subject: Re: [Rgc_analysis] [EXTERNAL] Re: RGC Beam Charge Asymmetry & Scaler Bank QA

Dear Gregory, Carlos and Noemie,

these are very interesting findings - and I can’t say I understand exactly what you are showing and what it could mean. I tend to think that it would be important enough to discuss this today during the daily RC meeting (unfortunately starting at 2:15 p.m.), since we may actually have the opportunity to think of some kind of tests in the Hall (beam is not expected back until later this evening). If it turns out that we cannot normalize our data to helicity- and lifetime-gated FC counts, we need to act rather urgently, so I would prefer not to wait until next Tuesday’s RG-C analysis meeting. Let me know what you think and if you can attend.

- Sebastian

On Sep 14, 2022, at 6:04 AM, pilleux at jlab.org<mailto:pilleux at jlab.org> wrote:

Hello Greg,

I have been wondering about the issues you show, and I asked Raffaella and
Nathan about it, I don't know if they had time to look into it yet.

As you show, for many runs there are weird HEL::scaler readings in a
number of files. For these files, the clock information gives crazy
numbers, and the accumulated charge is almost 0. Please find attached a
plot for run 16889 where I plot the time between two consecutive readings
(that I compute from the accumulated charge using the beam current). It is
supposed to be close to 33ms. As you show in slide 10, the first 86 files
look fine but the last 214 don't.

A problem I have with that is that the number of readings which are fine
change run by run, and it makes the normalization of yields, to compute
DSA for example, difficult. For now, all the normalizations of my dvcs
asymetries look crazy (even when I try to include the proportion of 'good'
readings in my scaling).

Best regards, Noémie.

One more word,

The analysis can be expanded upon by not just looking at file-by-file
scaler bank observables (for example, what was the fcup accumulation for
file 'n') , but entry-by-entry observables. In that case, we could ask
"how does the fcup accumulation evolve within the recon file itself?".
Perhaps this could help us isolate more outliers or anomalies in the data.
If any one has an idea of what we could investigate, feel free to let me
know.

Gregory
________________________________
From: Gregory Matousek <gregory.matousek at duke.edu<mailto:gregory.matousek at duke.edu>>
Sent: Tuesday, September 13, 2022 6:31 PM
To: rgc_analysis at jlab.org<mailto:rgc_analysis at jlab.org> <rgc_analysis at jlab.org<mailto:rgc_analysis at jlab.org>>
Cc: Anselm Vossen, Ph.D. <anselm.vossen at duke.edu<mailto:anselm.vossen at duke.edu>>
Subject: [Rgc_analysis] [EXTERNAL] RGC Beam Charge Asymmetry & Scaler Bank
QA

Hello all,

From today's RGC meeting, we were talking about putting in more effort
into understanding the beam charge asymmetries from the faraday cup
charge accumulation. I put together a presentation here which shows some
work on just that
(https://cc.jlab.org/system/files/users/gmat%40jlab.org/Presentations/Scaler%20Bank%20QA%20for%20Target%20Polarization%20%40%20RGC.pdf<https://urldefense.com/v3/__https://cc.jlab.org/system/files/users/gmat*40jlab.org/Presentations/Scaler*20Bank*20QA*20for*20Target*20Polarization*20*40*20RGC.pdf__;JSUlJSUlJSUl!!OToaGQ!poBSft7q1QE-VjblFkLfXh9HGcjN1frfU9VC6AYrvF5WWO1CO_fCjHLu0CxfWhleR5s-DA8btLudX_LpgJ-8PlunFRdZ-sWwLy3v$><https://urldefense.com/v3/__https://cc.jlab.org/system/files/users/gmat*40jlab.org/Presentations/Scaler*20Bank*20QA*20for*20Target*20Polarization*20*40*20RGC.pdf*3Chttps:/*urldefense.com/v3/__https:/*cc.jlab.org/system/files/users/gmat*40jlab.org/Presentations/Scaler*20Bank*20QA*20for*20Target*20Polarization*20*40*20RGC.pdf__;JSUlJSUlJSUl!!OToaGQ!poBSft7q1QE-VjblFkLfXh9HGcjN1frfU9VC6AYrvF5WWO1CO_fCjHLu0CxfWhleR5s-DA8btLudX_LpgJ-8PlunFRdZ-sWwLy3v$*3E__;JSUlJSUlJSUlJS8vKioqKioqKioqJQ!!OToaGQ!sL5RlMuLSrOHoZllSwaEIA3vCLC9hoNM93etcS4QfXD-fOT5Qr2MF3PpQbcOpUrlw8CXucpgF1HrKYw-x2I$>)

In this presentation, I look into the HEL::scaler and RUN::scaler banks
for Run 16889 and Run 16918 , the former having an apparent 0.6% BCA and
the latter -0.02% BCA (slide 16 reports +0.02% but this is because FC+ is
defined by beam/target spins parallel, where in fact it should just be
helicity = +1, accounting for just HWP status). We can see anomalies in
the scaler banks for both runs, which needs discussing.

Please feel free to ask any questions!

Gregory
_______________________________________________
Rgc_analysis mailing list
Rgc_analysis at jlab.org<mailto:Rgc_analysis at jlab.org>
https://mailman.jlab.org/mailman/listinfo/rgc_analysis<https://urldefense.com/v3/__https://mailman.jlab.org/mailman/listinfo/rgc_analysis__;!!OToaGQ!sL5RlMuLSrOHoZllSwaEIA3vCLC9hoNM93etcS4QfXD-fOT5Qr2MF3PpQbcOpUrlw8CXucpgF1HrKWLRrMQ$>

<readscaler.png>_______________________________________________
Rgc_analysis mailing list
Rgc_analysis at jlab.org<mailto:Rgc_analysis at jlab.org>
https://mailman.jlab.org/mailman/listinfo/rgc_analysis<https://urldefense.com/v3/__https://mailman.jlab.org/mailman/listinfo/rgc_analysis__;!!OToaGQ!sL5RlMuLSrOHoZllSwaEIA3vCLC9hoNM93etcS4QfXD-fOT5Qr2MF3PpQbcOpUrlw8CXucpgF1HrKWLRrMQ$>

_______________________________________________
Rgc_analysis mailing list
Rgc_analysis at jlab.org<mailto:Rgc_analysis at jlab.org>
https://mailman.jlab.org/mailman/listinfo/rgc_analysis<https://urldefense.com/v3/__https://mailman.jlab.org/mailman/listinfo/rgc_analysis__;!!OToaGQ!qBcUz5jlv4anI3nphlhxfcCBbKDY_dC7111187m1jRzF_rig0GQWrDc0-XA-AhZGTcvkAmWxX75caDW2z95q7lY$>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/rgc_analysis/attachments/20220924/b96d521f/attachment-0001.html>