The last two weeks I spent working on the new optics model to find if an
effective field boundary model would present an easier approach to do the
corrections necessary to fit the extreme regions of BigBite.

The effective field boundary model is present on page 1.  There is an additional
correction that can be applied to account for the fringe fields given in the 
form supplied by Penner (as I refer to it as the Penner correction).

In the following plots for the models I applied a set of coefficients optimized
by hand for the particular run type.

For the optics I found that there wasn't much to be gained going to the
effective field boundary model in the way of resolution and that the Penner
correction added in terms that not easily corrected for by first order terms.
This presents itself to me as an unphysical correction in the form given.  I
believe that this could be due to the fact that at the kinematics we are running
at, there are approximations made that do not apply to our situation and
therefore higher order terms will need to be applied for the fring field.

Also, these types of corrections did not improve the fit in the extreme regions.


The inplane angle difference dependence on p_diff was also examined.  Page 6
has the dependence plotted and a correlation can be seen very clearly.  On page
7 is an equivalent quantity, the difference in Y position on the neutron arm
and the position empirically prediced by BigBite.  Here, no correlation can be 
seen.  This suggests (though does not prove) that there may be a natural 
correlation between these two variables.  Esentially, for elastics the momentum
and scattering angle in BigBite are redundant variables.  How much the 
momentum is off can strongly influence the predicted position on the neutron arm
and therefore there could be a dependence.  I intend on doing a calculation to
show that this is or is not the case.

Seamus