Parameterization of deflections/preliminary momentum resolution results

From GlueXWiki
Revision as of 19:07, 22 May 2007 by Staylor (Talk | contribs)

Jump to: navigation, search

Using the 1500 Amp TOSCA map provided by Paul Brindza, I ran Garfield calculations over a grid in z(along the beam) and r corresponding to the dimensions of the FDC to compute the deflections of the avalanches along the wire due to the Lorentz force. The Garfield coordinate system, shown below, is unfortunately not the same as the GlueX coordinate system. The deflections in the Garfield coordinate system are in the +/-z direction. This corresponds to deflections in the +/-y direction in the GlueX coordinate system.

Garf coord.gif

The behavior of this deflection for ionizations throughout a given FDC cell is well-described by a plane:

Deflection example2.gif

The two slope (tangent) parameters describing the deflection planes are shown as a function of r and y=z(beam direction in GlueX coordinate system) here:

Tan def angles.gif

Here r is the same as x in the GlueX coordinate system. For interactions in the FDC for other phi values, the tan(theta_y) component is multiplied by cos(phi).

The hitFDC.c routine in HDGeant has been modified to read in a table of these tangent values and interpolate the amount of deflection as a function of the point of closest approach to the avalanche wire on a wire-by-wire/event-by-event basis.

Next I generated npi+ events with 9 GeV photon beam and t-slope of 0.25 GeV^-2 to mostly populate the forward region and ran the events through HDGeant. One caveat is that the current version of the B-field in the repository has slightly less magnitude (~2T in the central region of the bore as compared to ~2.2 T) than the one I used to generate the deflection table. Here is a comparison the momentum resolution as a function of theta for the case where the deflection is turned on (red) vs the nominal case (black). No smearing was added.

Dp p v2.gif