Subscribe to New Scientist magazine
Astronomy
ARTICLE

Giant telescope project begins with a spin

Movie Camera
  • 17:17 07 April 2008
  • NewScientist.com news service
  • Elise Kleeman
Printable versionEmail to a friendRSS FeedSyndicate
 
Expert technicians make giant telescope's mirror
Over the course of two days, 24 tonnes of glass was loaded into a furnace mold (Image: Ray Bertram/Steward Observatory)
Over the course of two days, 24 tonnes of glass was loaded into a furnace mold (Image: Ray Bertram/Steward Observatory)
 

Imagine having to melt 24 tonnes of glass to make an 8.4-metre-wide telescope mirror. Now imagine if that were the easy part of the project.

The mirror-making for the ambitious Large Synoptic Survey Telescope (LSST) project is now underway, but the greatest challenges still lie ahead.

By 2015, the new $400 million observatory should begin recording the entire night sky every three days from its perch in the Chilean Andes. It will provide a vast treasure trove of data invaluable for tracking near-Earth asteroids, watching supernovae explosions and mulling the mysteries of dark matter and dark energy.

Before then, though, LSST project managers must perfect how to gather the faint starlight, capture it for posterity and share the images with the world.

Spin Cycle

At the telescope's heart is its primary mirror. Its unusual design will incorporate a tertiary mirror ground into the middle, as if someone had taken a giant-sized ice cream scoop from the glass surface. To accommodate this extra depth, the mirror was built especially thick – almost a third of a metre at its centre, and about a metre at its edge.

Late in March, workers at the University of Arizona's Steward Observatory Mirror Laboratory fired up the furnaces around the mirror mould, heating the glass until it melted to the consistency of heavy molasses.

Then, they set the mould spinning, giving the mirror surface a gentle curve and eliminating the need for nearly 14 tonnes of extra glass that would otherwise have been ground away.

With the melting and rough shaping now complete, the mirror will rest in a computer-controlled cool-down for the next three months to avoid generating any stresses within the glass.

Trickier than Hubble

So far, the mirror appears to be in good shape, says LSST Project Manager Don Sweeney. "We'll only know for sure after it's down to room temperature," he told New Scientist. "But everything looks very, very good; it's very unlikely that there will be a problem."

For mirror lab director Roger Angel, though, the hard work is just beginning. Once the mirror cools, it will have to be ground and polished to gather light perfectly.

"There's quite a lot of testing there to make sure we get the curves right," he told New Scientist. "The mirrors that we make now are way more difficult in testing than the Hubble mirror."

Wide-eyed

With its vast primary mirror, 3.4-metre secondary mirror and sizable tertiary mirror, the LSST will be able to see a patch of sky about 40 times the size of the Full Moon (other large telescopes can only see a fraction of the Moon at one time).

Capturing and storing such large images is not easy, though. Developing sensors for the camera, which has a 60-centimetre-wide focal plane covered in 10-micron pixels, is one of the biggest challenges left to be solved before the telescope is up and running.

Another is developing a way to process and publish the 30 terabytes of data captured each night. "Scientifically, it is the world's biggest database," Sweeney says. "You have to process all of that data as fast as it comes up, because if you get behind, you'll never catch up."

 
Comment subject
Comment
No HTML except lower case italic tags or lower case bold tags, please:
<i> or <b>
Your name
Your email
 

We need your email in case we need to contact you about the comment. We will not use it for any other purpose.

 
 
There are 8 comments on 2 pages
1  | 
2
 | Next
 | Most Recent
 | See all

The Mirror

By Kevin Morton

Tue Apr 08 06:35:26 BST 2008

How is such a large mirror supported so that it wont

distort the image.

REPORT | REPLY

You'll Never Catch Up?

By Milton

Tue Apr 08 12:03:27 BST 2008

The final comment caught my attention, as it parallels a little-publicised difficulty faced by government-run signals intelligence ("eavesdropping") systems: how to cope with data received much more quickly than it can be processed.

Is Sweeney concerned because they won't be able to *keep* such large amounts of data for long? I.e. Is it a simple mass stoarge issue? Or is he assuming that no plausible processing speed improvements will allow a backlog to be processed in the future? I ask because (a) it's a really interesting problem, and (b) governments are storing the data, apparently gambling that they will be able catch up later ... In fact, that's what makes it interesting. Just how quickly does the teraflop-per-chip advance have to happen to make that even remotely possible?

REPORT | REPLY

You'll Never Catch Up?

By Seth

Tue Apr 08 13:56:33 BST 2008

I smell another BOINC project.

REPORT | REPLY

You'll Never Catch Up?

By Jardam Of Cnt

Wed Apr 09 12:03:17 BST 2008

Unfortunately not. You would flood the internet with data.

REPORT | REPLY

You'll Never Catch Up?

By Richard

Tue Apr 08 23:38:26 BST 2008

Let's see, we'll never catch up? Well this year and early next we'll have 8 core x86-64bit processors which most of the worlds supercomputers are built using.

If we expect a doubling of cores every 24months, then, by the time that this data is needed to be processed, we will be looking at processors hitting the 64 core mark; and that's a conservative estimate.

In addition architectural changes will bring further performance improvements probably twice within that time frame. End result is that you will probably be looking at 50-100 cores that do up to twice as much work as today's desktop chips.

That's a good deal more performance, to ensure that catching up will never be a problem.

REPORT | REPLY

Telescope On The Moon

By actionforspace

Tue Apr 08 23:18:12 BST 2008

Perhaps a telescope on the moon could be built using glass made from lunar dirt.

REPORT | REPLY

There are 8 comments on 2 pages
1  | 
2
 | Next
 | Most Recent
 | See all

All comments should respect the New Scientist House Rules. If you think a particular comment breaks these rules then please use the "Report" link in that comment to report it to us.

If you are having a technical problem posting a comment, please contact technical support.

Printable versionEmail to a friendRSS FeedSyndicate
Cover of latest issue of New Scientist magazine
  • For exclusive news and expert analysis every week subscribe to New Scientist Print Edition
  • For what's in New Scientist magazine this week see contents
  • Search all stories
  • Contact us about this story
  • Sign up for our free newsletter
 
PASSWORD LOGIN
username:
password:
 help
SUBSCRIPTIONS
Subscribe to New Scientist magazine