by: John M. Hill
Today's milestone was that overnight the 8.4m mirror has cooled below
the 500 C mark. Things happen pretty slowly while the mirror is
annealing, so crossing 500 degrees represents the milestone of being
3/8 through the annealing range. Maybe only oven pilots notice this
kind of milestone, but we have been cooling at a rate of only -2.5 C
per day.
By judicious poking and prodding around the outer wall of the furnace
after we stopped spinning on January 21, we have determined that the
extra ton of glass is on the floor of the oven outside the tub that
surrounds the mold. As physics dictates, this glass is also in a
parabolic shape piled up against the wall of the furnace. We deduce
that there must have been a leak somewhere around the edge of the
mold. So far we have not been able to determine the cause or the
precise extent of the leak, because you can't see much of the outer
part of the mold on the video images. We have been spending our free
time inventing more and more clever ways of peering into the oven
through small holes and trying to estimate the thickness of the
faceplate in the outer part of the mirror. We've got telephoto video
cameras, and laser pointers and a variety of other tricks. So far the
data is inconclusive, and we will probably have to wait until April to
learn what the real story is. In the meantime, the Mirror Lab may have
a new subsidiary named the Laboratory for High Temperature
Proctoscopy.
P.S. Save some "scaramanzia" for the launch of STS-82 on February 11.
This is the HST servicing mission where NICMOS (and STIS) get installed
in HST. NICMOS is all loaded into the shuttle and ready to go, except for
the final cooling of the solid cryogen tomorrow.
What is happening?
It has been two weeks since the last update of casting news. The week
of January 20 I was catching up on my sleep. The week of January 27 we
had a series of LBT Project engineering meetings to prepare the various
drawings and specifications for putting the steel parts of the
telescope out to bid. Besides that --- nothing much was happening to
the casting.What is annealing?
The reason we cool so slowly during the annealing temperature range is
to avoid stresses in the glass after it reaches room temperature. Low
stress makes the blank more stable and more durable. As the blank cools,
the center is always a bit hotter than the edge and thus expands
slightly relative to the edge. If this strain from differential
expansion is large during the whole time that the viscosity increases
(constant rapid cooling), the strain will become essentially
permanent. Because the glass is still relatively soft (low viscosity)
at the beginning of annealing, any stress in the glass from its
previous casting history or differential expansion will relax away to
nothing. When the blank reaches temperature equilibrium at room
temperature the strain field locked in during annealing reappears as a
stress field. This is why we cool so slowly between 530 C and 450 C to
avoid "freezing" in a strain field which appears later as a stress
field. Between 550 C and 400 C the stress relaxation time scale changes
from minutes to months. By cooling so that the temperature gradient is
around 1 degree C, we keep the final stresses due to annealing around
30 psi or less.Where is the "missing glass"?
In the previous report(#8), I mentioned that the glass level in the
faceplate seemed low and that we could not account for about a ton of
glass that we expected to be in the faceplate. Based on more careful
analysis of the Camera B video images, we now believe that the
faceplate thickness at the center of the mirror is 30 +-1 mm. Since
the final faceplate is ground and polished down to 27 or 28 mm, we have
enough glass in the center to be happy. This is why we put in extra
glass in the first place --- to account for the various uncertainties
of mold volume, rotation speed and small leaks. The thinner faceplate
even saves us a few weeks on the generating schedule since we have
less glass to grind off. Is there any excitement?
The oven pilots like to say that "boring is bliss". The biggest
excitement of the last two weeks happened on Friday January 31 when the
power in the polishing lab next door went off. This immediately
brought the UA police department since it tripped the alarm system.
There seems to have been an "unexplained" ground fault that tripped a
main breaker. "unexplained is bad" may also become a popular pilot
saying. The power was off for about two hours. There was no effect on
the oven since the casting lab is on different circuits than
polishing besides having backup generators. Because it was early in
the morning, there was also no effect on generating the back plate of
the Magellan 6.5m mirror which started last Wednesday.Just what are scaramantic gestures?
OK, here's what the word used at the end of Casting News #1 means:
"scaramantic" may not be an English word at all --- at least many of
you could not find it in the dictionary. "scaramanzia" is an Italian
word relating to behaviors or gestures that avoid having bad luck.
Some typical gestures range from double horns (like U. Texas Longhorns)
to gestures that I can't make in polite company. I'm not sure that the
horns work as well at Arizona Stadium as they do in Texas? Since
breaking an 8.4m mirror is really bad luck, keep on making those
scaramantic gestures. Sometime during annealing the glass actually
gets to a point where the viscosity is high enough that we could break
it with a thermal or mechanical shock.
Back to LBT Casting Updates