[Frost] phi offset

[Franz Klein]

Folks,
the 'beam' position makes only an impact to the track momentum if there 
is a strong field near the target.
However, the 'beam' position is not known quite well for photon beams 
(width ~7-10mm), thus the so-called beam position is just the center of 
the fitted multi-track vertex (assuming no magnetic field at the target 
position).
The so-called 'beam' or 'vertex' position is the intersection of the 
extrapolated track with a plane along the nominal beamline perpendicular 
to the sector in which the track was found. I is simple to move this 
plane to the 'beam center' line - but we didn't do that in the cooking. 
The other thing missing in the cooking is the holding coil field (which 
causes a small shift in the polar angle and azimuth for g9a, and a shift 
in the polar angle for g9b for low-momentum tracks).
The other unfortunate component is energy loss in dense material: 
currently we force tracks to loose energy in C, CH2 when they come from 
the neighbourhood of C and CH2, making assumptions about the pathlength 
inside the thin targets (it is in principle not different for butanol: 
we make some mistake when the track originated 5mm away from the beam 
center, etc).

To make it short: 'beam' position could NOT affect momentum 
reconstruction in the cooking!

What can be done:
(a) a corrected vertex fit routine including magnetic field + energy loss
(b) do empirical corrections

We have to do empirical corrections anyway since the torus field is only 
calculated using Biot-Savart and since I couldn't do a great job with DC 
alignment (less holes in the DC would have helped to get better 
convergence but I admit that the job could have been done better!).

My 2+ cents,
Franz

On 07/08/2013 10:27 AM, Eugene Pasyuk wrote:
> The beam offset for g9 a and b was of the order 1-3 mm. It is in caldb on the run number basis. I don't think that such an offset would affect momentum reconstruction.
>
> -Eugene
>
> ----- Original Message -----
>> From: "Michael Dugger" <dugger at jlab.org>
>> To: "Eugene Pasyuk" <pasyuk at jlab.org>
>> Cc: "frost FROST" <frost at jlab.org>
>> Sent: Monday, July 8, 2013 9:37:44 AM
>> Subject: Re: [Frost] phi offset
>>
>>
>> Eugene,
>>
>> Will a small shift in beam position matter much in determination of
>> momentum? It look like (my very rough estimate) that the shift in phi
>> is
>> much less than a degree. It only shows up as something that looks
>> large
>> when we look at the lambda pulls because those are being divided by
>> the
>> lambda resolution.
>>
>> -Michael
>>
>> On Sun, 7 Jul 2013, Eugene Pasyuk wrote:
>>
>>> We know that the beam was off the CLAS center in g9a and g9b. I
>>> suspect that beam offset might not properly accounted for in
>>> tracking. Could it be a problem in translation from sector
>>> coordinates to CLAS coordinates? May be instead of translating to
>>> CLAS coordinates we need to translate to "beam" coordinates?
>>>
>>> -Eugene
>>>
>>> ----- Original Message -----
>>>> From: "Michael Dugger" <dugger at jlab.org>
>>>> To: "frost FROST" <frost at jlab.org>
>>>> Sent: Friday, July 5, 2013 3:20:49 PM
>>>> Subject: [Frost] phi offset
>>>>
>>>>
>>>> Hi,
>>>>
>>>>>  From CLAS-Note 2003-017 (Mike William's kinematic fitting
>>>>> CLAS-Note), the
>>>> lambda resolution of CLAS from the TBER bank (using data with
>>>>   0.5*B_max)
>>>> is about 2.5 mrad. The value of lambda used in the kinematic
>>>> fitter
>>>> was
>>>> 1.5 times as that given in TBER.
>>>>
>>>> This gives us a resolution of about 0.21 degrees for lambda.
>>>> Priya's
>>>> pull
>>>> distributions for lambda of the proton (momentum < 1.0 GeV) have
>>>> the
>>>> pull
>>>> maxing out at about 0.5 for angles near -90 degrees.
>>>>
>>>> If I take the standard deviation for lambda to be 0.21 degrees and
>>>> assume
>>>> a transverse path-length of 1 meter, I can get the same sort of
>>>> results as
>>>> Priya if the coordinate system of CLAS was displaced from the
>>>> "true"
>>>> coordinate system by about 1.8 mm in the x-direction.
>>>>
>>>> It is starting to look like the lambda pulls might be explainable
>>>> by
>>>> simply shifting the reaction vertex by a small amount.
>>>>
>>>> Another way of looking at this, is that our systematic error in
>>>> phi
>>>> is
>>>> about 0.1 degrees.
>>>>
>>>> Take care,
>>>> Michael
>>>> _______________________________________________
>>>> 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
>>
> _______________________________________________
> Frost mailing list
> Frost at jlab.org
> https://mailman.jlab.org/mailman/listinfo/frost

-- 
===============================================================
                   Franz J. Klein, Associate Professor
                   CUA, Department of Physics
                   Washington, DC 20064
   office: Hannan Hall 206          phone: 202-319-6190
   or: Jefferson Lab,CC F-243       phone: 757-269-6672
---------------------------------------------------------------