SoylentNews is people
SoylentNews
ESO's Very Large Telescope has combined the light from all four of its Unit Telescopes into its ESPRESSO instrument for the first time, effectively creating a 16 meter aperture optical telescope:
The ESPRESSO instrument on ESO's Very Large Telescope in Chile has for the first time been used to combine light from all four of the 8.2-metre Unit Telescopes. Combining light from the Unit Telescopes in this way makes the VLT the largest optical telescope in existence in terms of collecting area.
One of the original design goals of ESO's Very Large Telescope (VLT) was for its four Unit Telescopes (UTs) to work together to create a single giant telescope. With the first light of the ESPRESSO spectrograph using the four-Unit-Telescope mode of the VLT, this milestone has now been reached.
After extensive preparations by the ESPRESSO consortium (led by the Astronomical Observatory of the University of Geneva, with the participation of research centres from Italy, Portugal, Spain and Switzerland) and ESO staff, ESO's Director General Xavier Barcons initiated this historic astronomical observation with the push of a button in the control room.
[...] Light from the four Unit Telescopes is routinely brought together in the VLT Interferometer for the study of extremely fine detail in comparatively bright objects. But interferometry, which combines the beams "coherently", cannot exploit the huge light-gathering potential of the combined telescopes to study faint objects.
Previously: First Light for VLT's ESPRESSO Exoplanet Hunter
(Score: 2) by HiThere on Wednesday February 14 2018, @06:26PM (6 children)
Synchronizing the distance between the telescopes to in integral number of wavelengths (or figuring a correction factor) when they aren't attached to a stable platform is a bit questionable, though. Radio telescopes have done the equivalent for a long time, and part of the reason they could and optical telescopes couldn't is that the wavelengths they were dealing with were enough longer to make it practical That they can do it with light at all is rather amazing. I'd be really surprised if they could do it cheaply with telescopes not attached to a fixed relative position.
OTOH, calculating the current atmospheric distortion isn't something I would have believed possible either, but these days all the big telescopes to it. So maybe. But it's a lot more difficult that you seem to be thinking. I still think two 5-mile mirror telescopes in Neptune's orbit would be better, and probably easier. The orbital diameter would be enough that you could use parallax to measure the distance to a lot of stars that we believe we know the distances to, but where the chains of reasoning have lots of "almost certainly"s in them.
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 0) by Anonymous Coward on Wednesday February 14 2018, @07:10PM
The great advantage that radio astronomers have is that they can do all of their image formation well after the fact. Their wavelengths are so long that they can digitize the radio waves as they arrive and time-tag them with an atomic clock (each antenna has its own clock and digitizer). Because of the time-tag, you can easily combine the signals from other telescopes because their data are time-tagged as well. You can sample many many points of the radio wave, and the errors from the time-tagging and digitization are well below what is needed to phase them all up later in software. In the optical regime you just can't digitize the waveform fast enough, nor time-tag it, so you have to do it all on the fly and only record the phased up image. The big segmented mirror telescopes have to do that for each segment.
As you mention, you're screwed when it comes to free-flying telescopes (or telescopes on floppy booms, for that matter). Positioning something to a gnat's ass is hard enough, but to position and control it to a gnat's ass, you have to sample and know it's position to a gnat's ass of a gnat's ass, and that ain't easy when you're talking nanometers on top of things that are vibrating at a kilohertz and higher.
(Score: 0) by Anonymous Coward on Wednesday February 14 2018, @08:09PM (1 child)
I just read the press release. They are not interferometrically combining the light from the four telescopes, just incoherently. That means they've quadrupled their photon collecting capability, but they haven't increased their angular resolution. You wouldn't want to run a swarm of small telescopes this way (except maybe for some very specific purposes) because you'd only have the angular resolution of the small telescopes.
For instance, take the 2.4 m diameter Hubble primary. That's the same collection area as 2200 50-mm apertures, but why would you want to fly 2200 cubesats that gives you the image resolution that you'd get with a 50-mm telescope? Now, if you could combine those 2200 apertures interferometrically, and fly them in a swarm that had a diameter of 24 meters (or 240 meters), that would be very interesting.
(Score: 2) by bob_super on Wednesday February 14 2018, @08:30PM
As fine controls improve, the swarm of mirror-sats reflecting light onto a sensor-and-analyzer center sat starts to make a lot more sense than the massively costly and risky business of the Webb.
Serviceability and upgrades would be a lot better too: just switch the sat at the focal point for an upgraded one every decade, without having to relaunch the mirror array (or vice versa).
(Score: 2) by bob_super on Wednesday February 14 2018, @08:14PM (1 child)
> Synchronizing the distance between the telescopes to in integral number of wavelengths (or figuring
> a correction factor) when they aren't attached to a stable platform is a bit questionable
It's absolutely amazing the precision that can be achieved:
https://arstechnica.com/science/2018/02/lisa-pathfinder-mission-a-glorious-success/ [arstechnica.com]
Now, that's with a much smaler mass than a telescope would be, but swarms of satellites keeping very precise tabs on each other to run observations are not a new topic.
(Score: 2) by HiThere on Thursday February 15 2018, @06:24AM
Well, my first thought was "I think that's only two independent pieces" and my second was "that sure wasn't any cubesat"...
But it *is* an sort of existence proof that the thing should be doable. I don't, however, think it would be doable with cubesats...at least not any way that they're normally launched. IIRC the LISA pieces needed to be in stable relation to each other to make that work (though I'm not sure quite what that means when things are in independent orbit...possibly it had to do with calculating a correction factor).
But the proposal was for lots of independent launches of the cubesats, which would make positioning them a real challenge, unless it's all done by calculating correction factors. But for that to work relative motion would need to be essentially nil (or at least calculable by a simple function) and accurate to a fraction of a wavelenght. Now the light collecting power of a telescope depends on the area of the mirror, but the resolution depends on the diameter. So for a decent telescope you'd need hundreds of cubesats whose position relative to each of the others was calculated to, I think, 1/8 of the shortest wavelength you want to observe, and for the advantage you'd want them spread in a plane over as wide an area as possible. (You'd need to talk to a telescope builder to get that figure accurate. but it's in the same range as the Lisa positioning.) This hasn't been done on the ground until recently, because accurate positioning was too difficult. I think I read when they started doing it with mirrors on tracks at one of the large telescopes a couple of decades ago...and ran into problems with getting position accurate enough. With radio telescopes they do it from one side of the earth to the other, but those wavelengths are between meters and kilometers, so positioning is a lot easier. (OTOH, since each cubesat would have it's own focal point, I think the plane over which they are spread can be flat rather than the surface of a parabola. Or if you're going to use fancy calculations to form your image, perhaps it could be spread in a 3-d cloud of known positions, and then you could look in any direction just by rotating the cubesats independently...which would mean they'd need to handle their gyroscopes with exceedingly fine precision.)
So, while it's got lots of possibilities as a design, I think it's well beyond the state of the art. (That said, I'm a programmer, not a satellite designer, and even then I'm essentially retired, so perhaps it isn't as far beyond the state of the art as I think.)
Javascript is what you use to allow unknown third parties to run software you have no idea about on your computer.
(Score: 2) by JoeMerchant on Wednesday February 14 2018, @08:47PM
The small-satellite telescopes could be free floating, that seems to be... ambitious, unless you're willing to accept resolution at a fraction of the digital sensor and use stellar navigation to correct for pointing errors.
What seems more practical to me would be for the swarm to dock to itself - changing pointing direction could be a challenge, but if the docking mechanism is robust enough, it should be able to handle hundreds, if not thousands of separate mirrors and sensors.
🌻🌻 [google.com]