The gross features of the electron drift, shown in the plots below, are unsurprising: the isochrones roughly parallel the equipotential lines and the drift lines are compressed into four drift channels converging at the anode wire. Drift through the pair of channels furthest from the anode wire requires an extra 0.020-0.030 us, reflecting the longer drift paths. This effect has been studied in detail elsewhere. [Betts, 1996]
Electron drift past the gating grid and ground wires to the anode wires for the inner subsector (left) and outer subsector (right); isochrones are spaced at 0.010 us.
The field is nearly uniform and constant 0.2 cm or more before the gating grid, so one can simulate the drift of ionization from there to the anode wires to estimate the difference between the inner- and outer-subsector drift times. The results appear in the plots below, predicting that drift in the outer subsector requires an additional 0.083 us.
Times for electrons to drift from a track 0.2 cm before the gating grid to an anode wire in an inner subsector (left) and an outer subsector (right). Times are in microseconds.
At the nominal TPC drift field (E=148.05 V/cm) with E and B parallel, Magboltz predicts a drift velocity V=5.396 cm/us. This is a bit below the ~5.506 cm/us value seen in the STAR TPC, but the calculation here will be internally consistent and give correct values for Z'_d,in and Z'_d,out, if the ratio of real to calculated drift velocity is constant.
The drift numbers are summarized in the table below, showing the prediction that Z'_d,out-Z'_d,in=0.47 cm for the alignment of the inner and outer subsectors - as opposed to Z_d,out-Z_d,in=0.20 cm, based on geometry alone.
| ||Z_d - Z_0||t_d - t_0||(t_d - t_0) x V_drift||Z'_d - Z_d|
Summary of the drift time results from Garfield.