[Primexd] Random trigger skims

[Alexander Austregesilo]

Hi Igal,
If I understand the method correctly, you have to run the plug-in over all files in order to get a fiducial map for the time span of an entire run.
Cheers,
Alex



-------- Original message --------
From: Igal Jaegle <ijaegle at jlab.org>
Date: 1/23/21 10:18 PM (GMT-05:00)
To: Alexander Somov <somov at jlab.org>, Peter Pauli <Peter.Pauli at glasgow.ac.uk>
Cc: Alexander Austregesilo <aaustreg at jlab.org>, primexd at jlab.org
Subject: Re: [Primexd] Random trigger skims

Peter,

The PS_flux plugin needs to be run on all evio files or just the first 5 are enough for each run?

tks ig.
________________________________
From: Alexander Somov <somov at jlab.org>
Sent: Friday, January 22, 2021 12:35 PM
To: Peter Pauli <Peter.Pauli at glasgow.ac.uk>
Cc: Alexander Austregesilo <aaustreg at jlab.org>; Igal Jaegle <ijaegle at jlab.org>; primexd at jlab.org <primexd at jlab.org>
Subject: Re: [Primexd] Random trigger skims

Thanks Peter,

I will take a look, EPICS interface is more difficult to handle, it's easy to deal with the
PS rate directly.

Cheers,
             Sasha
________________________________
From: Peter Pauli <Peter.Pauli at glasgow.ac.uk>
Sent: Friday, January 22, 2021 8:16 AM
To: Alexander Somov <somov at jlab.org>
Cc: Alexander Austregesilo <aaustreg at jlab.org>; Igal Jaegle <ijaegle at jlab.org>; primexd at jlab.org <primexd at jlab.org>
Subject: [EXTERNAL] Re: [Primexd] Random trigger skims

Hi all,

So here is the method I use.

In order to see if the beam was on or not the counts in the PS are used. They correlate very nicely with the beam delivery. David L. added a few histograms to the PS_flux plugin which are basically filled when there are counts in the PS. All I do is convert the histogram into a string that can be put into the ccdb which contains the entries in the hist for each time increment (bin-by-bin). Then in the DBeamCurrent_factory this database is loaded and used to determine if the beam is on by checking if the count is over a certain threshold. It seems a bit complicated which is owed to the fact that previously the beam_on condition was derived from EPICS. I wanted to keep the changes to the code as minimalistic as possible and have it backwards compatible.

If you want to have a look at the macros I wrote to get the data tables from the histogram have a look at
/w/work/halld/home/ppauli/randomTrigger/

It contains the files as well as some of the tables for the 2020 batches that have already been produced.

If you need more info on this method please get in touch.

Cheers,
Peter


On 21 Jan 2021, at 22:56, Alexander Somov <somov at jlab.org<mailto:somov at jlab.org>> wrote:


Hi,

The fraction of accidental hits in the CCAL is relatively small
so the impact of accidentals on the efficiency of Compton events
is not significant, at least for the event selection I used.

The difference in efficiency was well below 1% (on the level of sensitivity of
the MC sample I used) when the beam trip time was excluded  (for some
runs with the typical fraction of the beam off time )

At any rate, when Peter is done with his method, we can produce new hddm files
for random triggers. The easiest way would be to accept or reject random triggers
in between SYNC events, where we read out TS scalers (I don't know how exactly
Peter is implementing this).

Cheers,
             Sasha



________________________________
From: Primexd <primexd-bounces at jlab.org<mailto:primexd-bounces at jlab.org>> on behalf of Alexander Austregesilo <aaustreg at jlab.org<mailto:aaustreg at jlab.org>>
Sent: Thursday, January 21, 2021 5:25 PM
To: Igal Jaegle <ijaegle at jlab.org<mailto:ijaegle at jlab.org>>
Cc: primexd at jlab.org<mailto:primexd at jlab.org> <primexd at jlab.org<mailto:primexd at jlab.org>>; Peter Pauli (PGR) <p.pauli.1 at research.gla.ac.uk<mailto:p.pauli.1 at research.gla.ac.uk>>
Subject: [Primexd] Random trigger skims

Hi Igal,

We have a plugin called randomtrigger_skim to filter out the random
triggers from any evio input and convert them into the hddm format that
can be mixed into the simulation by mcsmear. To be more realistic, we
only accept random triggers that fall into a time period were the beam
was actually on. The plugin takes this beam fiducial map into account,
but it has to be generated beforehand. Peter Pauli (cc) was just working
on an improved scheme for this extraction, maybe he can help you also
for the PrimEx data.

Once the fiducial map is ready, you just have to run the plugin over all
evio files. Afterwards, we can concatenate the hddm skims from
individual evio files so that we have one file per run.

Before you have the fiducial map, you can also include the option
`RANDSKIM:USEBEAM 0` into your config file to simply write out all
random triggers. This serves already as a good approximation.

Let me know if you have any questions,

Alex

_______________________________________________
Primexd mailing list
Primexd at jlab.org<mailto:Primexd at jlab.org>
https://mailman.jlab.org/mailman/listinfo/primexd

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/primexd/attachments/20210124/8f0aa194/attachment-0001.html>