[Frost] New findings for FROST momentum problems

[Michael Dugger]

Hi,

Is it possible that a shift in the vertex time is causing problems in 
matching the photon to the event in COBRA?

For GPID the photon picking requires that the 
photonTime+propTime-vTime be within the ST_TAG_COINCIDENCE window of 5 ns.

For Cobra, it looks like there is vertex timing cut in Event.C 
(TimingCut(float __t)) that takes as an argument the coincidence window 
for photon picking. I am having difficulty finding where that window is 
set. There is some code in ClasEvent.h (_index.SetTimingCut(1.5)) that 
sets the default timing cut to 1.5 ns, but I am not sure that this is the 
coincidence window for photon picking.

If the coincidence window for photon picking within COBRA is set to 1.5 
ns, that might cause problems when the vertexTime is shifting on the order 
of 2 ns.

When I get the chance I'll check to see if this momentum problem is seen 
with GPID.

-Michael

On Thu, 18 Nov 2010, Eugene Pasyuk wrote:

> Hi Brian,
>
> You did interesting study.
> The shift in vertex time for different ST paddles by itself is not
> alarming. All times in CLAS are measured with respect to the trigger
> time. If trigger time shifts all times shift. Trigger time depends on
> which detector fired the trigger. For that reason we never use vertex
> time for event selection and PID. What we us is vertex time difference
> obtained from different detectors. Particularly, time difference between
> track vertex time and beam vertex time. When we look at the time
> differences those shifts due to the different trigger time cancel out.
> The differences between g11 and g9a can be explained by the differences
> in electronics. It was not exactly the same.
> So, the time shifts you observe are not necessarily bad thing.
> However, the difference in occupancies may indicate some problems with
> trigger efficiency. We need to do more studies of this. Particularly
> sectors 6 looks odd.
>
> Having said that I don't see how those timing shifts can affect momentum
> reconstruction.
>
> -Eugene
>
>
>
>
> On 11/17/10 2:34 PM, Brian Vernarsky wrote:
>> Greetings all,
>>
>> I have been continuing to work on finding the source of the momentum
>> discrepancies we see between the butanol and CH2 targets.  I tried to
>> look into the tracking code, but could not find any problems there.
>> So I started to look at some other reactions than the gamma p ->  p pi+
>> pi- () reaction, and when I looked at the single track reactions I
>> found what might be the source of our problem, the start counters.  I
>> have written up all of my findings in a presentation you can find at:
>>
>> http://www.andrew.cmu.edu/user/bvernars/FROST_update_2010-11-17.pdf
>>
>> It is a somewhat large file, around 10 Mb.  Please let me know if you
>> have any questions or ideas.  Thanks
>>
>> Brian
>> _______________________________________________
>> Frost mailing list
>> Frost at jlab.org
>> https://mailman.jlab.org/mailman/listinfo/frost
> _______________________________________________
> Frost mailing list
> Frost at jlab.org
> https://mailman.jlab.org/mailman/listinfo/frost
>