The secondary mirrors of Tel. A & B were realigned by Marc
and Nat just before this science run. Hence, getting the
telescope beams into the laboratory and realigning the
star-acquisition CCD mounts took almost two nights.
I made some science measurements on the third night but
in the presence of strong laser background light on the
star-tracker CCD. I opened the tank, improved the baffling
and realigned the IOTA relay mirrors. Since then, I didn't
have any laser leak problem.
Then a storm passed by and I had a bad weather for more
than a week. The sky cleared up, but the seeing was terrible
during next two nights. I made lot of measurements, but mostly
without a fringe tracker.
The last four nights were excellent and I recorded good
amount of data during these nights.
Looking at the stellar images generated by scanning with
IONIC fibers, I got the impression that the focusing of
Telescope A & B are much much better than Telescope C. To
make this case stronger, the flux on PICNIC for Tel. A
and B were about a factor of two more than that of
Telescope C. But, later I noticed by looking through the
alignment telescope that the focusing of Telescope C is
not all that bad. Also, I noticed that the focusing of
Telescope A is not as sharp as Telescope B. So, I
suspected the IONIC alignment for Telescope C. I realigned
the flat mirror and the dichroic (since I didn't want to
touch the parabola) on the IONIC table corresponding to
the Telescope C in order to improve the image of
Telescope C as seen by the IONIC fiber. Now, the flux on
the PICNIC and the star-tracker CCD are comparable for all
three telescopes. I estimate the coupling efficiency of
IONIC (under excellent seeing conditions) as 65% for
Telescope B and about 40% for Telescope A & C.
Both Telescope A & B images (as seen by the fibers) are
elongated in one axis. Re-aligning corresponding secondaries
may improve the coupling efficiency of IONIC. However, I
believe that at this level of alignment, equally important
is the alignment of the parabolas on the IONIC table.
I did manage to keep the laser spots out of the star-tracker
CCD, but they are there. It could be a serious problem when
we have a more sensitive star-tracker CCD. So, it should be
Although, the indirect solution suggested by John is attractive,
I think, we should try to find a direct solution to this problem.
probably, any residual leak could be blocked with a Notch filter
as and when required.
Long Delay server:
Seeing the good performance of the ldelay server, I decided
to run the 'server' program on the console. I didn't have to
reload the server program during rest of my run. I killed the
program just before leaving the mountain for security concern.
I noticed that ldelay computer doesn't have 'xlock' application
on it, although there are document files related to 'xlock' in
this machine. I think, it's good to load this application
and lock the display after launching the 'server' program.
Now that the ldelay server is performing well, I don't see any
reason to power it up though CBCOM. I think, ldelay should be
powered directly, running all the time. If so, the existing
CBCOM could be used to recycle power for the LD motors in
order to save its life time. 'vxworks-start' and 'vxworks-stop'
scripts could power on and power off LD motors.
Tracking failure with Telescope C:
Telescope C fails to track at low declinations, while A & B
are tracking well. OT says that the telescope is tracking with
zero 'error' value and it reads the 'position' as 'command'
value. But, the star is drifting at about the sidereal speed
and it's not possible to acquire the target again. Is it an
issue of telescope imbalance or telescope modeling?