These results have been obtained with one event from a set of standard density cocktail events. For simplicity, only tracks leaving a hit in each SVT and SSD layer are considered. The fast SVT/SSD simulator has been used to include the detector position resolution.
In order to improve the track projection accurary, the TPC tracks can be refitted with the main vertex. Below is a figure which shows the ideal hit to track projection distance in the four layers with and without the main vertex in the fit (stat. information are for without the main vertex). When the main vertex is not included in the fit, the track parameters are taken as the are estimated by the tpc refitter. When the main vertex is considered into the fit its weights are 100, 100 and 500 microns in x,y and z.
As expected, without the main vertex, the distribution gets worse when going from the outermost to the innermost layers since the projection distance increases. The inclusion of the main vertex significantly improves the quality of the projections. Moreover the quality improves when projection on innermost layer which can explained by the a shorter distance to the main vertex. As a matter of fact, the weigth associated to the main vertex tends to force the refitted tracks to pass near the main vertex. This are averaged results and should depend on the tracks transverse momentum and their type (primary or secondary).
