Date: Wed, 20 Dec 2000 13:59:18 -0800
From: Stan Whitcomb <stan@ligo.caltech.edu>
Subject: problem with the E2 data

Keith,

We have discovered a problem with some of the E2 interferometer signals that
people working on the E2 investigations should be alerted about.

The key error signals for locking the interferometer are derived from the 
various
photodiode RF signal by demodulating with two arbitrary (but quadrature) local
oscillators.  After digitization, these are rotated in phase to give the 
correct
phases (.eg., AS_I, AS_Q).  What we discovered was that the rotation matrix
was not implemented correctly, so that the the error signals (AS_I, AS_Q,
REFL_I and REFL_Q),  while linearly independent, are not as orthogonal as they
should be.  It isn't as bad as it could be since the rotations that we were 
using
were typically 5-10 degrees.

This won't affect many of the investigations; for example any investigation
(e.g., tidal) using the feedback to the arms would be unaffected by a little
cross-coupling in the error signals.

Let me try to assess the importance of this on each of the investigations:

Lock Loss --  probably no effect
Seismic -- no effect, provided they use control signals, not error signals
Correllations -- there will be an effect, I will be sending an email to 
Nelson and Adrian
Noise stationarity -- probably little effect
Calibration -- some effect, probably not major.
Nongaussianity -- probably not major
Angular fluctuations -- minimal effect
Transients and bursts -- minimal effect
Bursts and chirps -- minimal effect
Tidal -- no effect
Timing -- no effect
Data integrity -- no effect
Line characterization -- little effect on analysis, maybe a little effect 
on interpretation
Data compression -- no significant effect
Frequency noise -- significant effect, Rana is aware of the problem and has
some ideas about how to deal with it

I suggest that we forward this message to the E2 investigators and then I will
briefly describe it at the next DC telecon.  In the meantime, Iwould be 
happy to
answer any questions that the E2 investigators might have

stan