A Casting and Re-Dedication!

At the Mirror Lab where I work, we're casting another 8.4 meter diameter (nearly 28 feet!) mirror substrate for the Giant Magellan Telescope. This one is #4, which will be the center one in the 7-mirror array that makes up the 25+ meter diameter telescope. On the occasion of the "hot hold" at 1165C as the glass melted into the light-weighting mold, the lab was renamed the Richard F. Caris Mirror Lab in honor of our newest benefactor. So at the end of a busy week of activity, on Friday we gathered to celebrate the renaming and another casting. Of course, there were speeches and congratulations all around, especially with the recent delivery of the Large Synoptic Telescope mirror a few months back. Shown at left is a photo-shopped image of the new sign on the building modified as an announcement f the rededication. And at right are a couple of the speakers at the short ceremony, Joaquin Ruiz, Dean of the College of Science, and Ann Weaver Hart, President of the University of Arizona.

Richard F. Caris, the new namesake of the Mirror Lab formed the company Interface Inc, which manufactures load cell force sensing instrumentation. We've been using devices from his company since our first mirrors, so evidently he felt we were something in which he desired to invest. At left, Head of the Astronomy Department and Director of Steward Observatory Buell Jannuzi at center reveals the photo that will hang at the entrance of the Lab, here assisted by President Hart and Associate Director of Steward Jeff Kingsley.

There was a nice crowd, a good fraction of which I knew!  Interestingly, as each mirror cast has its own t-shirt design, only the latest version reflects the name-change.  At right is the black design of GMT3 from 2 years ago, and the new powder blue reflecting the new name and artwork.

After the ceremony, we were all queued up to tour the Mirror Lab, and watch the spinning oven with 18 tons of molten glass within. Of course, the VIPs were allowed in first - interestingly, though I work there, I was at the end of the line in group 29! I walked in regardless to grab some photos - you never know when you might want to do a blog post! Melinda attended, but wasn't up for too much walking, so found an office to sit in for a while. I sneaked past the volunteer security staff to go catch Mr. Caris next to the spinning oven, shown at left chatting with another guest (note the new signs on the wall behind him!). The tour exited down past the base of the oven and out under the handling ring, which dwarf the guests walking past it at right.

We didn't see much of the casting crew - they were busy monitoring the oven.  The temperature was approaching its peak, so they had plenty to do.  The oven, spinning at 4.8 rpm, was radiating heat, and the interior orange glow could occasionally be glimpsed as it spun.

I connected with some friends in the crowd, and walked them into one of the labs so I could pull up the time-lapse of the glass melting that was live on the internet. About the time we finished that, and my explanations of some of my activities, the staff was turning out the lights trying to clear visitors out. Interesting, with the bright lights out, illumination from the safety lights provided an unusual glow to the interior of the polishing lab. Shown at left is a 4-frame panorama of the two 8.4m mirrors awaiting work. The GMT3 substrate is on the Large Polishing Machine at left, awaiting generating of the back plate, and the GMT2 is parked under the test tower at right awaiting faceplate generating. Both are awaiting refurbishment of the Large Optical Generator which is wrapping up. As I went past the spinning oven for the last time, I took a shot of the digital thermometer attached to the oven, to compare to the shot I took earlier. Shown at right, in the 58 minutes elapsed time between frames, the temperature had climbed 19C, so was likely programmed to climb 20C per hour.

As I mentioned above, there is a time-lapse automatically generated as images are taken through the 7 cameras watching the interior of the oven. The cameras are normal small digital cameras, so to keep from melting, they peer through sapphire windows and are actively cooled with fans. Similarly, since the interior is dark except for the glow of glass, the cameras shoot mostly in the blue part of the spectrum, so need flash units to light up the interior. Since this time-lapse clip continues to grow, I saved it early Saturday morning at 300MB, and uploaded it to Youtube to make it easier to view. There is enough resolution to watch it full screen. Note that the clip is copyright Richard F. Caris Mirror Lab.




As you can see, there are many things going on. Of course, from room temperature, the mold that the glass is cast into is assembled from an alumina silicate ceramic. The glass is graded before loading - rejecting some for high stress and impurities, and the best quality set aside to go in last to form the upper faceplate. At the start of the video the glass blocks can be seen placed atop the mold. As the temperature climbs above 800C, the glass starts to soften and at it's highest temperature of 1165C is a liquid consistent with that of honey. At some point it becomes fluid enough to flow down the 1cm-sized gaps to form the ribs and backplate at the base of the mold. The excess that doesn't flow into the mold will become the faceplate, which is about 2 inches thick (note the marks at the edge of the mold, indicating the faceplate thickness). It is a fascinating time-lapse, and I love seeing these every time!

As I write this, the oven has cooled to under 750C. The displays state the oven is still spinning at 4.8 rpm, but as this is under the "freezing point" of glass, I expect the full-speed rotation to end any time now. At left is shown the temperature-profile-to-date, so it doesn't spend much time at full heat - only a few hours, before cooling rapidly (don't blame me for hard-to-read labels - they are assigned randomly). Cooling will slow down soon though, as annealing starts. Through the 400C-600C range, they must cool it very slowly (.1C/hour) to control stress and crystal formation. But all-told, the entire heat cycle lasts about 3 months, so I expect we'll get to inspect the new substrate before the Christmas break! Stay tuned!