The parameterization of guiding errors is largely a control engineering
task the details of which are beyond the scope of this report.
Briefly, equation (
) is used with
the mean speed U and turbulent intensity 
as provided by the flow
measurements described in section 4.1.3. The wind
spectrum is modeled by a Von Karman spectrum () in
which the
experimental values of length scale 
are input. The
wind tunnel tests ([Zago 89b] and [Alexandrou])
provided also the torque coefficients of
the telescope for the different orientation and enclosure
configurations.
A series of simulations allowed to parameterize the response
functions about both axes as a function of the factor 
.
Two distinct control cases were considered:
for observations in the visible wavelength
the active loop tilt control (field stabilization) by the
secondary mirror (M2) was evaluated to be so effective with a bandwidth
Hz that residual errors are negligible [Quattri 88].
At the time of writing this report the development of the M2 is
still in progress and it is not possible to know if this
analysis will be confirmed. It may be that the bandwidth of the
tilt control loop will be in fact limited by the need to integrate out
atmospheric image motion in the input signal.
For observations in the infrared wavelength it is planned to use the secondary mirror in the chopping mode. In this case the active loop tilt control cannot be used and the guiding accuracy relies only on the servo loops of the main drives of the telescope. Only results for this latter case are presented here below.