37 replies to this topic

[Lee in Texas]

Has anyone done it? I understand it wouldn't be easy, but it sure would be cool to see an amateur telescope with insane aperture. I'm thiking a proof-of-concept prototype would be relatively inexpensive to build. Just use 6 small blanks and try to get everything figured out.

[mark cowan]

It's a wonderful idea.

I had the same one about 10 years ago.

Conservatively speaking, and with a lot of creative scrounging, you can make it work correctly, in prototype form, for around a quarter of a million...

And you'll still have a horrible secondary obstruction.

Best,
Mark

[Lee in Texas]

egad. that bad?

[mark cowan]

Could be even worse. You'd need to maintain sub-wavelength alignment of all the mirrors in real time in order to use it. If your last name is Keck, though, I'd say go for it!

The two 10 meter, 36 segment Keck telescopes cost $170 million. If your prototype used 6 mirrors at about 6" diameter, the area is 1/700 of each Keck. If it scales by area, you could do it for half the estimate. But most of the hardware required to coordinate the mirrors doesn't scale by area. The Keck segments are kept within 4 nanometers alignment, twice a second (according to keckobservatory.org, who oughta know).

OTOH, CCDs and computers keep getting cheaper.

Unfortunately that's not the case with piezo actuators, or anything else that can do that job.

I've kicked it around for a long time now, and it still doesn't look practical. Much much easier and cheaper to just make a big mirror, until you get up to 10 meter class or so. The original Multiple Mirror Telescope was converted to a monolithic 6.5 meter mirror in 1999, 20 years after it was built!

Best,
Mark

[Mike I. Jones]

Could skip trying to coherently phase all the apertures together and go spectroscopic a la an amateur version of the Hobby-Eberly Telescope (HET) at McDonald Observatory. Spectrometers want LIGHT!!! Just aim the mirrors together to feed into a 200-300um multi-mode fiber, run it into a spectrometer and sensitive CCD camera, and do stellar spectroscopy.

Agree with Mark - multi-aperture coherent phasing with atmospheric compensation is not yet in the realm of even well-heeled amateurs (unless Bill Gates gets a hankerin'). Not yet, anyway.
Mike

[mark cowan]

I left the atmospheric compensation out of the original price tag.

Best,
Mark

[mconnelley]

Hello:

I've thought of the same thing, and it would be very difficult. For an amateur sized scope, you probably wouldn't need to update the mirror positions very frequently, but definately every night. If you thought collimating was hard, try co-collimating each of the segments to each other!

An easier way may be to use a monolithic meniscus mirror, like Subaru,
Gemini, and VLT use. For an amateur scope, it would be something like 1/2" for a 20" mirror. For such a scope, astigmatism will be the primary problem, plus any residual spherical aberration. A set of, say, 8 force actuators on the back of the mirror in a ring plus one in the middle of the mirror should allow the user to control astigmatism. This adjustment would have to be done on a star after regular collimation. The advantage is a lighter mirror and faster cool down. The disadvantage is a longer set up time and the need to understand how to figure your mirror on a star, and a more difficult design and construction. However, since the mirror is one piece of glass, you do not have to phase the different segments (i.e. you can't have piston or tilt discontinuities).

Cheers
Mike Connelley

[gnabgib]

The Keck is not a multi mirror telescope! It is a segmented monolith. Each of the "segments" is an off axis parabola and when combined form a complete parabola. The original multi mirror telescope never worked well because of all the aformentioned reasons. The success of the Keck telescopes is because of their approach to "simulate" a solid parabola with individual sections. Advancements in actuator technology with more powerful computers made this possible. I worked on the prototype warping harnesses to make the off axis parabola years ago at the Lawrence Berkeley Laboratory and also worked on the first mirror segment supports and positioner assemblies. Our first positioners were able to repeat position to within one nanometer over a two millimeter range while hooked up to a 400 kilogram weight. I believe that an individual today could build a "mini-keck" but why go to all the trouble to build a 48 inch "multi-mirror" when a 48 inch monolithic mirror is more practical and far less costly. Now if someone wanted to build a 10 meter AMATEUR telescope that is something I could get behind!!
Kevin

[gnabgib]

Another brain tumor!! Didn't someone at Stellafane many moons ago exhibit the beginings of a multi-mirror telescope? As I recall the persons last name was Dilworth. Was it ever completed? Anyone else remember this?
Kevin

[Alan French]

Another brain tumor!! Didn't someone at Stellafane many moons ago exhibit the beginings of a multi-mirror telescope? As I recall the persons last name was Dilworth. Was it ever completed? Anyone else remember this?
Kevin

Yes, I remember it. I don't recall if it was Donald Dilworth of not, however.

Clear skies, Alan

[mark cowan]

The Keck is not a multi mirror telescope! It is a segmented monolith. Each of the "segments" is an off axis parabola and when combined form a complete parabola...

Well, yes, but whether the original poster was asking "can you combine the light from multiple mirrors" or "can you make a mirror from multiple segments" the implementation problems are much the same. I don't even want to mention making multiple off-axis paraboloids, for obvious reasons.

Interesting history though.

10 meter amateur scope, hmmm...

Best,
Mark

[gnabgib]

Mark;
We developed a warping harness to mechanically warp the mirror blank. When this warped blank was ground and polished spherical and then released from the warping harness is was an off axis parabola! The degree of off-axisism (bad english?) was controlled by how much warping was applied to the blank. The methods used are well within the abilities of todays "amateur" telescope makers.
Kevin

[Lee in Texas]

My question was about using six "production" mirror blanks to make one big telescope. Same concept as pre-'99 MMT in Arizona. Before this thread, I didn't know it had already been tried. I read about a proposed MMT a couple of years ago and the idea sounded fascinating.

I don't remember the mfr's name, but I saw a website with 48" hex cell mirror blanks. Imagine 6 of those making up an amateur telescope. If only it would work.

[mark cowan]

Well, there you're talking about standard paraboloids and some kind of method to converge all the beams together at a common focus? See all the previous caveats!

Best,
Mark

[mark cowan]

Kevin,

We developed a warping harness to mechanically warp the mirror blank. When this warped blank was ground and polished spherical and then released from the warping harness is was an off axis parabola! The degree of off-axisism (bad english?) was controlled by how much warping was applied to the blank. The methods used are well within the abilities of todays "amateur" telescope makers.

I like the idea well enough! Was the harness over the full backside, the edge, or a combination?

Best,
Mark

[Lee in Texas]

Well, there you're talking about standard paraboloids and some kind of method to converge all the beams together at a common focus? See all the previous caveats!


Best,
Mark

No. I meant off-axis parabaloids, making one big parabaloid. When I said "production" mirror blank, I meant using a piece of glass that doesn't have to be custom-made.

What a shame that no one could get it to work. Imagine a 12foot amateur telescope.

[gnabgib]

Mark;
The warping harness was attached to the edge of the blanks using invar blocks epoxied to them.
Kevin

[gatorengineer]

Actually, its done on a fairly regular basis these days in terms of CCD imaging. I have seen a number of photographs produced from compilations of photos taken by astronomers a few thousands of miles apart with different scopes. Or by one amateur runnign simultaneous exposures on different scope. This is effectively a multi-mirror amateur set up.

As it stands right now the only advantage of these compilation is a reduction in exposure time. But perhaps someday an amateur version of the "lucky" camera equal to Hubble? Never say never.......... Look where digital astro imaging was 15 years ago.....

[mark cowan]

But without the phase information, which means it's the same as just stacking off a single OTA. Not multi-mirror, even though the original complex wavefront was rectified by different OTAs. Multi-mirror gets you more resolution.

I like the "lucky" approach - how fast and how sensitive do you think the sensor needs to be? As I understand it that's the key there.

Best,
Mark

[Lee in Texas]

Maybe in 15 years, controlling a 12ft MMT will be withing amateur's grasp. But hey, in 15 years, maybe a 12ft mirror would be in the same price range as six 48" mirrors and their controllers. It would be cool to see something like that belong to a club.

[Owen]

http://amazing-space...ele_mmt_big.jpg

This is one possibility.

However, using computer control over TCP-IP for a standard alt-az mount, starting with all scopes at polaris, there is no need for the mirrors to be in the same 'scope, or even the same country.

A chunk of software akin to registax would stack the images from the 'scope ccds.

The main benefit would be a smoothing and averaging of atmospheric turbulance - a helpful by-product.

That would be one hell of an ATM project, and is distinctly achievable. The software could be an open source project (or two - one for the imaging, one for the control) and the 'scope design would need to be standardised.

High quality optics would need to be used to maintain sample consistancy, but the main criteria would be the 'scopes operate at the same magnification, and the CCDs are of similar resolution (to minimise the image interpolation within the slice).

So to progress, we'd need a smallish, easly controllable 'scope design that is robust and reliable, using off the shelf CCDs.

We'd need comms software to control the movement of the scopes, and to stream the images back to a base. We'd then need image interpretation software akin to registax to take each feed as a single slice.

Initially, the imaging could be done in synchronised 'burst' mode (ie each CCD takes 30 frames from a designated start time, sent by the base controller), so netwroking from remote sites would be alot less of an issue. This would also indicate that a webcam type set up would be better than a very high resolution DSLR or astronomical CCD - it will also help with the cost.

It could become a 'community' of 'scopes - the more the merrier

Thats a project definition and possible route to sucess if ever I saw it.

So Who Wants to help build ATLAS (The Amateur Telescope Large Array System)


Owen

[Owen]

A little more thought on this.

The positioning, time and frames could be from a web-based RSS or XML feed - standard mechanisms for this sort of information.

That would leave local control software needed to position the 'scope, fire the CCD, collect the images. The images could be 'hosted' on a website too.

That would certainly cut down a chunk of the work needed.

Oh yeah, we wouldnt be restricted on aperture either - the key is magnification. as long as that is the same, you could use a good 10" upwards in the system, or a 4". All contributions would be welcome


Owen

[userlain]

Owen,

That would be an awesome idea if we had something to research. NEOs perhaps. Any ideas?

Jacob

[userlain]

Amateur Sky Survey with CCD and Atmospheric Normalization

or

Arse-CAN for short.

[Lee in Texas]

Amateur Sky Survey with CCD and Atmospheric Normalization

or

Arse-CAN for short.

LOL.

But seriously- Owen's idea sounds cool. Maybe in the future I can take part in something like that. Could everything be controlled from a central location? (firing of the CCD and go-to instructions?)