[Olympus] Things to investigate for the luminosity determination

[Michael Kohl]

Hi,
here are some notes on tasks for the luminosity determination that were
discussed:

Slow-control luminosity:
Based on a microscopic simulation of the conductance in the presence of 
wakefield suppressors and other details of the target, Brian has 
accomplished a new density profile that extends beyond the nominal length 
of the target cell and which provides about 20-30% more target thickness 
for a given flow.

Identified task:
The target simulation should be validated, e.g. using wire chamber tracks 
in a narrow angular bin (to keep the cross section constant) for a scan of 
the vertex distribution. 
The detailed shape of the density distribution can be affected by local 
temperature profile; it should be simulated how much the density can vary 
subsequently.

12-degree luminosity:
Based on MWPC-only tracks, significant progress has been achieved. 
Observed yield distributions agree quite well with MC now. The luminosity 
from Brian's 12-degree MWPC-only analysis agrees with the slow control 
luminosity to 4% (higher than SC). The left-right systematics (luminosity 
difference) is at the level of 2%. The ratio of e+/e- is at the 1% level 
consistent with unity and has been blinded to a +-0.5% level.
SiPM efficiency maps have been determined.

Identified tasks:
Proton tracks
Refinement of cuts

MWPC:
Stripe structure only in right sector (mainly theta) not understood. 
Present in data and MC. Possibly bug in reconstruction.

Small bump at missing energy of 600 MeV only for e+ in left and right 
sector

Further digitization improvements. MC resolution is a bit too good.
Use GEM tracks to verify hits on MWPC


GEM:
Sync error: it is a flag in the code to occur when for the event less than 
12 APVs have data. APV data would show values outside of physical 
range, containing header data or simply random memory data. Happens 
runtime dependent, not always but if it happens then for the same events. 
The current handling is not correct, and may lead to correlated 
inefficiencies. 
-> assess how often it happens
-> either discard the event or make sure that data transfer is done 
correctly

Incorporate pedestal and common-mode noise subtraction, masking of dead 
channels, gain calibration

New GEM plugin 2.0


SYMB luminosity:
The average luminosity from SYMB is about 5% percent lower than the slow 
control luminosity.
The luminosity from the SYMB system shows a charge asymmetry of about 9% 
(higher for positrons), which is not understood. 
No trivial factors of 2 are responsible.
Thecharge asymmetry in SYMB of 9% reduces to 7.8% by different use of the
two-vertex generator.
There is no radiative annihilation generator, however this can only 
produce a percent-level modification.
There are too many events in simulation, or too few in data.
ep in SYMB: underflow bin unreliable, rate off by factor 2, e+/e- is near 
unity by coincidence

Gain studies:
Electron gain (ADC over simulated photoelectrons) changes around run 8200, 
but not for e+
-> digitization issue? How does the number of photoelectrons depend on 
beam position?

Evaluation of LED pulser data for gain monitoring: 10% decrease for 
central crystal over long time. Short-term variations likely due 
to rate dependence

e-(Moller) variation due to field change, beam pos change -> to be 
reproduced in MC

Compare gain for LED on vs. off, compare LED gain with beam on/off


Best regards
   Michael



On Tue, 10 Mar 2015, Douglas K Hasell wrote:

> Dear Colleagues,
>
> 	The meeting at ASU went very well and the workshops concentrating on tracking and luminosity were very productive.  Michael will send some notes regarding the luminosity determination.  Below I list some of the things we agreed to pursue for the tracking.
>
> 	The current tracking is quite slow (or needs a lot of computing resources).  We identified a number of things to investigate to speed up the tracking:
>
> For the jump scans optimise the depth, number, and kinematics of the jumps.
>
> Investigate excluding hits before passing the event to the tracker.  Possible to exclude hits based on an initial fit to the clusters.
>
> Optimise the initial parameters based on a simple (circle) fit to the clusters.
>
> Determine what causes a large number of track candidates to be tracked for the same set of clusters.
>
> Other things to be done for the tracking:
>
> 	- redo the time to distance determination
>
> 	- optimise the lamdba / beta parameters in the EAA
>
> 	- study wire chamber track / hit efficiency
>
> 	- incorporate the pion generator to see if this can explain some of the excess seen in data
>
>                                                    Cheers,
>                                                            Douglas
>
> 26-415 M.I.T.                                 Tel: +1 (617) 258-7199
> 77 Massachusetts Avenue                       Fax: +1 (617) 258-5440
> Cambridge, MA 02139, USA                      E-mail: hasell at mit.edu
>
>
>
> _______________________________________________
> Olympus mailing list
> Olympus at mit.edu
> http://mailman.mit.edu/mailman/listinfo/olympus
>

+---------------------------------------------------------------------
| Dr. Michael Kohl, Associate Professor and Staff Research Scientist
| Physics Department, Hampton University, Hampton, VA 23668
| Jefferson Lab, C117, 12000 Jefferson Avenue, Newport News, VA 23606
| Phone: +1-757-727-5153 (HU), +1-757-269-7343 (Jlab)
| Fax:   +1-757-728-6910 (HU), +1-757-269-7363 (Jlab)
| Email: kohlm at jlab.org, Cell: +1-757-256-5122 (USA)
+---------------------------------------------------------------------