Let us consider an experiment where the time at which the centroid arrives at the collector is measured. This is not the same thing as measuring the time at which the peak collector current occurs. We will deal with the experiment in which the peak collector current is measured in the next section.
The distance travelled by the centroid in time t can be found
by integrating 
:
If we measure the times 
and 
at two
different drift distances 
and 
, and then extract the
drift velocity from the ratio 
, as
Løvaas et al. [1988] do, then the non-hydrodynamic
effects are completely eliminated. However, if we just take the
ratio of drift length to drift time, then a correction needs to
be made to take into account the 
term.
Inverting (
) to get t as a function of 
, the measured value of the drift velocity is obtained by
dividing 
by the time (since all vectors in the system
must be proportional to 
we can treat them as scalar quantities):
England and Elford [1987] supposed that the correction due
to all the end effects is a power series in 
:
Equation () is exactly this form, the correction being
just the sum of a geometric series:
