There is a new cone in STAR to hold the FGT and the SVT.  This has been implemented in GEANT, see http://www.star.bnl.gov/HyperNews-star/protected/get/fgt/931.html  and http://www.star.bnl.gov/cgi-bin/protected/cvsweb.cgi/StarVMC/Geometry/IdsmGeo/

There are two places where there is significant material, where the inner and outer tubes are joined to the (sharp) cone structure.  For the inner, this is at (y,z)~(21.5,55) cm, while for the outer at ~(40,60) cm.  These are the two rectangles in the plot below; they are actually built out of a bunch of bolts, but are m modeled in Geant as an aluminum ring.

  I decided to map these locations into eta, vz space, where eta is the actual physical eta (not detector eta), since electrons passing through this area may see strong Bremsstrahlung and photonic backgrounds.  Below is the mapping; I only drew the points if the electron would project to the Barrel EMC (taken actually as 200,200, which is very approximate for the outside edge of the TPC).  The cone starts to shadow the TPC at vertex z > 20, and has somewhat maximal shadowing (i.e. mid-rapidity shadowing) for 40 < vz < 80 cm.  Note that at positive z there will be a rather large amount of material above the red line, from the actual FGT structure, the first disk of which starts at 70 cm.  At negative z, negative eta, the area is essentially empty, so these are rather narrow lines of extra material that could be avoided if necessary.