[Eg6_analysis] meeting tomorrow

[Dupré Raphaël]

Ok I understand better now, I agree that at first order if everything
is symmetric it should just broaden your distribution. However, the
shift in CLAS
 and in RTPC plus all the dead channel in both CLAS and RTPC make them
not symmetric, so you cannot assume everything will cancel out. This
is particularly true here because of the highly constrained kinematic
and the cross section varying fast. Indeed if energy on both sides is
slightly different (which we seem to observe) the effect becomes very
different from one side to the other.

Such shift should however be also seen in the drift path on overlying
kinematic region, but our precision is not good enough here and we do
not see it. The large intrinsic broadening in the gas plus the size of
the pads might not allow to easily determine the shift. One way might
be to isolate similar kinematics sample in well working detector
region to check this and see if we can get more precision.

The main problem with our calibration method, like for gain
calibration, is that we have a very constrained kinematic and our fits
can easily compensate any error in our assumptions. That is another
reason to look for other channel in 6 GeV data to check what we are
doing and eventually improve it.

Best regards,

On Fri, Mar 8, 2013 at 5:52 PM, Nathan Baltzell <baltzell at anl.gov> wrote:
> Hi Raphaël,
>
>
>> Up to now, we never seen any phi dependence of the drift paths,
>
>
> Sorry, I didn't intend to suggest there should be.  I was just trying
> to understand how you can get a global shift from a radial offset
> when you average over events that have a symmetric phi distribution.
> Take any two regions 180 apart (like the two dominant sectors for
> elastics), and if any effect results from a radial offset it should
> be opposite sign for them.  But ok, probably I'm thinking wrong ....
>
>
>
>> well since we have a phi centered on 180 right now!
>
>
> The difference between the phis for our elastic electron and alpha
> is actually about one/two deg off from 180 on average.  But I was
> more interested in knowing if this amounts to truly just a shift
> in phi and no effect on curvature, then it is not a big deal and I
> should assume no effect on gain calibration.
>
> Regards,
> Nathan
>
>
>
>
> On Fri, 08 Mar 2013 09:41:13 -0600, Dupré Raphaël <raphael.dupre at gmail.com>
> wrote:
>
>> Hello Nathan,
>>
>> Up to now, we never seen any phi dependence of the drift paths,
>> Mohammad will have to check this again now that things are working
>> better, but I do not expect any variation there.
>>
>> The shift in phi when moving the beam comes from the fact that most
>> event comes from two sectors and are basically produced on the same
>> plan. There is no reason that applies while measuring events all over
>> the detector, then any shift would affect us.
>>
>> Yes the shape of dphi gives the curvature, the overall shift is
>> apparently due to the shift during the amplification, we could check
>> that the value is not crazy there, but it does seem to work pretty
>> well since we have a phi centered on 180 right now!
>>
>> Best,
>>
>> On Fri, Mar 8, 2013 at 4:03 PM, Nathan Baltzell <baltzell at anl.gov> wrote:
>>>
>>> Hi Mohammad and Raphaël,
>>>
>>> It looks like maybe a global shift in dphi of about 1deg.
>>> I'm trying to understand, maybe you can help me:
>>>
>>> Is it correct that the drift paths calibration is independent
>>> of phi?  And this would be the same as ageraging over phi.
>>> And then is it understood why moving the RTPC would result in
>>> a global shift in the dphi calibration for a y-offset but not
>>> an x-offset?
>>>
>>> Is it the case that the shape of this dphi vs TDC distribution is
>>> what turns into track curvature, while an overall shift in dphi
>>> would just move the track in phi?  Or is that too naive?
>>>
>>> Sorry if I got this all wrong ....
>>>
>>> Thanks,
>>> Nathan
>>>
>>>
>>>
>>>
>>>
>>> On Fri, 08 Mar 2013 03:33:51 -0600, <hattawy at ipno.in2p3.fr> wrote:
>>>
>>>> Hello all,
>>>>   Following our meeting yesterday, here you can attached file with the
>>>> results. New parameters are needed, so i will start directly, i may not
>>>> finish today.
>>>>
>>>> Best regards,
>>>> Mohammad.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>> Hello everyone,
>>>>>
>>>>> We will have eg6 meeting tomorrow, March 7, at 10am in F227.
>>>>> Agenda can be found at -
>>>>>
>>>>>
>>>>> http://clasweb.jlab.org/rungroups/lowq/wiki/index.php/Meeting_on_March_7_at_10:00am
>>>>>
>>>>> (thanks to Nathan)
>>>>>
>>>>> Calling instructions are at -
>>>>>
>>>>>
>>>>> http://clasweb.jlab.org/rungroups/lowq/wiki/index.php/Agendas_and_Minutes_of_the_meetings#Agendas
>>>>>
>>>>>
>>>>> Stepan
>>>>> _______________________________________________
>>>>> Eg6_analysis mailing list
>>>>> Eg6_analysis at jlab.org
>>>>> https://mailman.jlab.org/mailman/listinfo/eg6_analysis
>>
>>
>>
>>
>> --
>> Raphaël Dupré



--
Raphaël Dupré