Re: talk today on the pair spectrometer

[Alexander Somov <somov@jlab.org>]

Hello Richard,

I am planning to tilt pair spectrometer counters and
check the energy ranges once again (to be sure that evrth.
is Ok). I will check it in svn shorly.

Note:
-----
At the moment I have converter inside the collimator cave
(2m upstream from the pair spectrometer magnet). If you like
I can move them to the position after the cave (closer to the
converter).

Cheers,
       Sascha




On Mon, 22 Sep 2008, Richard Jones wrote:

> Sascha,
>
> When last we spoke, you were going to make a few tweaks to the pair
> spectrometer detector geometry, and then check it in to svn.  Could you
> let me know when you have done that?  I am planning to do the
> systematics study for the endpoint energy calibration.  Once I have the
> corrected geometry I am able to run the simulations on our cluster to
> get the statistics we need.
>
> Before you check things in, please be sure to update your local gxtwist
> installation with the latest changes from subversion.  I have put in
> numerous fixes, that I would not like to see rolled back.
>
> Richard Jones
>
> Could you let me know when you have checked in theHrachya.Hakobyan wrote:
> > Richard,
> >
> > I don't know if agreement to define the endpoint energy reconstruction
> > is stiil in power.
> > I remember more&less how that was done in past. It was the
> > minimisation of the difference(functional)
> > BremMC(xo)-Integral(BrTh(x0)*Gaussian(x-xo)dx, where BremMC(xo) is a
> > simulated incoherent  bremmstrahlung spectrum with all  factors
> > involved, BremTh(x) is its analytical expression for 12 GeV endpoint
> > and Gaussian is the smearing function. The parameters of the fit are:
> > the scale factor, introduced  into argument of BremTh, i.e.k*x instead
> > of x, and the sigma of Gaussian. As a first step before the fit the x
> > value is defined as a ratio of PS energy scale to approximately
> > evaluated experimental endpoint energy  Eeexp  in the same units.
> >  If the PS energy scale is correct, the scale factor k obtained should
> > be equal to 1 within errors.If not it will prompt you how to  shift
> > the experimental endpoint energy for the next iteration. Rougly
> > speaking Ee=Eeexp/k. Obviously few bins before and after endpoint
> > energy  are really affected by resolution and so it is evident the
> > precision of  scale factor k should be better by factor 2-2.5  than
> > the resolution itself.
> >
> > Hrachya
> >
> >
> > On Thu, 18 Sep 2008, Richard Jones wrote:
> >
> >> Hrachya,
> >>
> >> Do you have a pointer to your talk on the pair spectrometer for the
> >> collaboration meeting?  It is scheduled for today at 4:30pm.  I know
> >> this will be difficult for you because of the time difference.  If
> >> you can pass me a copy of the talk, I will get it uploaded to the
> >> portal and will assist you with the presentation in any way you wish.
> >>
> >> Also please let me know how you would like to connect to the
> >> meeting.  If you want us to call you, I will ask our collaborators at
> >> Jlab to make the call as they have for the last couple of meetings.
> >>
> >> Richard Jones
> >>
>
>