Question about Timing recovery

    I want to simulate a circuit of timing recovery in a QPSK digital
demodulation system . In the receiver,the upsampled signals are match
filtered and tranmitted to a interpolation filter ,then followed a TED and
loop filter. 
    In order to determine the parameters of the loop filter,the error
detector sensitivity Kd should be known first,Kd is the slope of the
detector around zero error.Kd is computed from  open-loop simulation.What
confused me is that, the interpolation filter should be included in the
open loop or not,if yes,how to determine the error control signal ,which
is the input of the interpolation filter?

Regards,
Michel