[Halld-cal] gamma confidence level plots

[Shepherd, Matthew]

I should have added just a little more explanation.  This is from looking at pi0 photons in very clean omega events.  I make a plot of 

RF time + flight distance/c - shower time

where the flight distance is computed from the center of the target (to coincide with the RF time).

Then I can look at the width of this (and mean) as a function of energy.

It seems the time walk correction that exists now in the code is not optimal (nor is the time determination).  If I go back to cluster time I can derive a new time correction as a function of energy (turns it out is pretty much linear).  Also using the energy weighted time average of hits rather than the present algorithm of taking the time of the largest hit produces some modest improvements in resolution.

Matt


> On Aug 24, 2017, at 4:50 PM, Sean Dobbs <s-dobbs at northwestern.edu> wrote:
> 
> This is really nice to see.  For the record, the MC block level times are smeared by 400 ps, so I imagine that's approximately what we should see for this showers in simulation.
> 
> Cheers,
> Sean
> 
> On Thu, Aug 24, 2017 at 4:46 PM Shepherd, Matthew <mashephe at indiana.edu> wrote:
> 
> Attached is what I see in data for timing resolution of FCAL showers as a function of shower energy.  I think Sean said the MC was smearing at 400 ps.  That seems like a reasonable start.  Obviously I need to make the same plot for MC.
> 
> I think we can improve the timing resolution a little in data also.
> 
> Matt
>