Date: Mon, 19 Mar 2001 10:17:22 -1000
From: Dean Shutt <d.shutt@jach.hawaii.edu>
To: Ian Robson <i.robson@jach.hawaii.edu>
Cc: Per Friberg <p.friberg@jach.hawaii.edu>,
     Iain Coulson <i.coulson@jach.hawaii.edu>,
     Tomas Chylek <t.chylek@jach.hawaii.edu>,
     Nick Rees <n.rees@jach.hawaii.edu>, Tim Chuter <t.chuter@jach.hawaii.edu>
Subject: Re: tracking/pointing

Ian,

Our current engineering explanation of this effect is outlined below:

1.  The transit pointing effect (a.k.a. the "S" correction) is due to
windup

2.  Windup is due to the shaft between the elevation bearing and the
encoder having a non-infinite torsional stiffness and the presence of
friction in the elevation bearing system

3.  The CHANGE in the magnitude of the S correction is due to CHANGE in
the elevation balance state of the telescope.  i.e.  The telescope is
slightly more out of elevation balance now than when Iain made the
previous measurements.  Hence the torque applied to the shaft for
elevation changes is slightly larger, thus the torsional deflection of
the shaft is greater resulting in a larger S correction.

4.  The change in shape of the S correction is postulated to be due to a
lot of second order effects which I am not smart enough to explain just
now.


The suggested proof of the above (other than the I am not smarter part
which folks currently accept) is to purposely unbalance the telescope a
reasonably large amount and track an object through transit.  The next
day balance the telescope as precisely as possible and repeat the
experiment.  Measure motor current vs. elevation for both final balance
states.  The magnitude of the S correction should correlate linearly
with the elevation unbalance and hence the motor current.

We currently have scheduled telescope engineering time in June when we
may be able to arrange such a test.  In the meantime the Chief Engineer
and Sr. Mechanical Engineer are convinced that the telescope is not
falling apart but that the structure is linearly elastic.

There is always a good deal of discussion surrounding these puzzles but
the suggested test should give a definitive answer as to the validity of
the explanation.  Any additional transit tracking tests that can be
arranged will provide further info as well.

Dean