Why no premium Hyperstar?

[ollypenrice]

Every year I expect Starizona and/or Celestron to produce a proper Hyperstar telescope and yet evey year they don't do so. Here is a incredibly fast and affordable optical system which is occasionally seen to produce excellent results. (Remember Greek Anthony's narrowband wonders from a few years ago?)  Clearly few people get the best out of the Hyperstar because the engineering is cobbled together out of a system intended to work at a tolerant F10, not an excrutiatingly intolerant F2. Focusing by moving the primary is never going to be the way forward, for instance, and very high precision collimation features would be a boon.

Surely this would also be a project for the one man band astro-engineers to play with? There's the Ian King-tweaked GSO RC, for intance. The used market could supply plenty of mirrors and correctors for the pre-edge models, at least.

So come on, astronomy industry, re-engineer the Hyperstar! Personally I doubt that I'd ever shoot luminance in a Hyperstar but the colour would be another matter.

One little tweak would be to have a device for rotating the cables a little between shots. Since the speed of the Hyperstar would allow you to take plenty of subs you could stack in Sigma Clip to make all diffraction effects from the cables disappear.

Olly

[pete_l]

I suppose it all comes down to economics.

Most observers are just that: observers. A Hyperstar is totally an imagers tool and would therefore have a much smaller target market. So the sales numbers would be low and would therefore have to include all the development costs - which would make the product much more expensive. After that it would be hard to compete with all the small, cheap reflectors. And then there would probably be patent-wars which is the surest way ever devised to kill off innovation and slow down development.

Maybe a better route would be for someone to pick up Meade's old F/3.3 FR and bring that up to date with a decent image circle. Though I can't see that Meade is in any position to do the work.  With a bit of clever design, you may even be able to get a filter wheel in the optical path - something you can't do with a Hyperstar.

[Naemeth]

I guess it is also very expensive to manufacture a system that works well at F/2. I guess the push-pull system they use is at the right price for Celestron, so they continue to use it.

[ollypenrice]

Not so sure. The RC is effectively imaging-only as well, and the GSO ones are being rehoused in better engineered packages by third parties. I don't see a great difference.

I've just seen that Tak are re-introducing the Epsilon 130 for around £2300. This will be F3.3, 430mm FL and with a 44mm image circle. I bet this drums up a bit of interest. I couldn't live with the diffraction spikes at that kind of focal length (just too many stars in the field) but if they'd make a version with an optical window instead of a spider...

Olly

[Mike Hawtin]

Olly poses a very interesting question here and I think an outfit like Starizona would probably turn a very nice profit if they were to offer a properly engineered upgrade kit for serious imagers.  From a practical point of view there are only two issues to address in order to turn a HyperStar equipped SCT into a very formidable imaging machine.  In fact there is only one real problem and that is collimation but that problem has two main causes.  Firstly the push/pull collimation screws on the HyperStar unit will not collimate out a tilted camera sensor. Secondly  the offset focuser rod in combination with the flexible adhesive compound used to fix the mirror to the tube which slides along the baffle tube is not stable, gravity moves the mirror as the ota's attitude changes and so as a result collimation drifts out during a session.

With my own C6 I have fitted a tilt adjuster between the HyperStar lens and camera to address the first point and for the second I've done away with the stock focus mechanism and fitted a crayford focuser with a central focus rod to move the mirror assembly.  So two simple changes have been all that was necessary to turn a temperamental nightmare into a practical set up.  So consider the costs involved, a tilt adjuster, perhaps with some sort of filter drawer system at one end and a stock focuser with a length of studding and tube clamp arrangement at the other. Sounds like a practical proposition to me, I did mine on the kitchen table with basic hand tools.

Mike

[Gedan]

Interesting ideas here on how to remove diffraction spikes through cabling......looks like a hassle but could be worth it.

http://www.werbeagentur.org/oldwexi/hyperstar/tipstest_hyperstar.html

[JamesF]

I can't help wondering if one wouldn't end up with a more flexible instrument by making, say, an f/4 newt with twice the primary surface area (of the f/2 Hyperstar system) and an optical window rather than a spider.

James

[ollypenrice]

I can't help wondering if one wouldn't end up with a more flexible instrument by making, say, an f/4 newt with twice the primary surface area (of the f/2 Hyperstar system) and an optical window rather than a spider.

James

Then you're chasing the focal length up, though. You also need a high quality parabolic mirror, whereas the Hyperstar's is spherical and, therefore, cheaper to make and pretty insensitive to collimation. The Hyperstar, I'd have thought, would really all be about collimation of the front element and since that's a lens it should be fairly stable in a well made structure. There are only two optical elements in the Hyperstar and one is spherical. (Primary mirror and lens. I won't count the corrector because it's non-collimatable.) In a fast imaging Newt there are three, primary, secondary and corrector. It seems to me that a properly made Hyperstar is potentially simpler than the fast imaging Newts, and faster. Maybe that's optimistic? I don't believe the Hyperstar lens will attain premium apo standards but it has proved to be pretty good.

Gedan, moving the cables would not be a hassle. You make an outer ring round the front of the tube. An arm, probably curved as the Austrian chap suggests in the link, goes to the middle of the scope and meets a rotable ring round the camera. The inner ring, arm and outer ring move together and the cables are attached to the arm. You move it between subs or, ideally, have a sequencer move it for you. I believe the diff effects would vanish entirely.

Olly

[pete_l]

From a practical point of view there are only two issues to address in order to turn a HyperStar equipped SCT into a very formidable imaging machine.  In fact there is only one real problem and that is collimation but that problem has two main causes. 

I wonder if there's a technology cure for this?

Instead of having a traditional focus knob that's mechanically coupled to the mirror, how about a series of servos, of the kind used in adaptive/active optics systems? The amount of focus could be set coarsely by mechanical means (or stepper motors). But after that have the electronics make microscopic (and at short focal ratios, they would be microscopic) adjustments to the main mirror's position to compensate for tilt, flex, sag and thermal expansion.

To have all of this "just happen" the whole system would need a new set of electronic interfaces that none of the existing software packages have facilities for - a sort of auto-focus/collimate. Probably the closest we have at the moment is the software for DDMxxx mounts that does a trial run across the proposed imaging course to allow the mount to compensate for mirror flop, since the mount contains no tracking capabilities.

[Edit] Of course the ultimate would be to have a Hyperstar capable (i.e. slim) camera with a pick-off CCD, so that all of these adjustments, and A.O. too, could be done internally by the OTA in real time.

[ollypenrice]

Good thoughts from Pete. I think that servo collimated and focused fast imaging reflectors would have an awful lot going for them if they could be make to work. Meade added servo collimation to some of their ACFs, didn't they, though this was user controlled? Something like CCD Inspector could probably be made to drive the servos, no?

Olly 

[Mike Hawtin]

Good thoughts from Pete. I think that servo collimated and focused fast imaging reflectors would have an awful lot going for them if they could be make to work. Meade added servo collimation to some of their ACFs, didn't they, though this was user controlled? Something like CCD Inspector could probably be made to drive the servos, no?

Olly 

Don't bet the farm on it Olly, I've had CCDInspector report zero tilt in X and Y with the collimation value less than one pixel and yet the frame visually showed errors.  I only use it now to get somewhere handy and then go by visual assessment.

Mike

[ollypenrice]

Don't bet the farm on it Olly, I've had CCDInspector report zero tilt in X and Y with the collimation value less than one pixel and yet the frame visually showed errors.  I only use it now to get somewhere handy and then go by visual assessment.

Mike

Didn't know that, Mike. Thanks. I've never used it myself. I always go off primary sources, such as drift aligning and visual checks of incoming images.

Olly

[sharkmelley]

I have a Hyperstar tha needs collimating and a potential camera tilt problem.  I wondered if WinRoddier could be used to perform a diagnosis (rather than CCDInspector)?  I notice that in the French magazine "Astronomie" they always use WinRoddier to produce an analysis of the accuracy of mirrors during equipment reviews (why don't UK magazines do similar?).  So I wondered if a similar approach could be used for a "cloudy night" collimation.

Mark

[ollypenrice]

I have a Hyperstar tha needs collimating and a potential camera tilt problem.  I wondered if WinRoddier could be used to perform a diagnosis (rather than CCDInspector)?  I notice that in the French magazine "Astronomie" they always use WinRoddier to produce an analysis of the accuracy of mirrors during equipment reviews (why don't UK magazines do similar?).  So I wondered if a similar approach could be used for a "cloudy night" collimation.

Mark

I hadn't noticed that fact in Astronomie but I do like their bench tests. I asked AN about the possibility of doing this and the problem is simply one of finding someone with the kit to do it.

Olly

[laser_jock99]

I can't help wondering if one wouldn't end up with a more flexible instrument by making, say, an f/4 newt with twice the primary surface area (of the f/2 Hyperstar system) and an optical window rather than a spider.

James

I wonder if I could convert my Cape Newise 8" F6 into a Hyperstar...........

[ollypenrice]

I wonder if I could convert my Cape Newise 8" F6 into a Hyperstar...........

Aha, you have one of these. How does it perform?

That's a tiny central obstruction.

Olly

[perfrej]

As a former owner of and EdgeHD 925 with Hyperstar I can say that I do like the Hyperstar and hated the scope. Flimsy thin "tin" tube, flexure in the main Losmandy bar and a mirror focusing system that just doesn't work.

I did not have any problems with collimation of the HS itself, and the scope held collimation surprisingly well. EdgeHD, interestingly enough, was made with input from Europe where a good mirror lock was conceived and presented to Celestron. They did, however, change it in order to cut costs. Personally, I find mirror locks evil as they prevent fully automated use.

I hate spikes nowadays so the great scopes like the AG-12 is not on my list. Instead, I keep thinking about the spider-less reflector... I don't know how much you would lose in optical quality if a simple clear glass of high quality was used as a secondary mirror holder, like in a Schmidt or Maksutov design.

Back on topic, the Hyperstar does, in my view, have a major drawback: the placement in the central obstruction. It's OK if you have an M26C that doesn't add to it, but the prospect of a filter wheel is pretty far off. And the spikes...

/per

[ollypenrice]

As a former owner of and EdgeHD 925 with Hyperstar I can say that I do like the Hyperstar and hated the scope. Flimsy thin "tin" tube, flexure in the main Losmandy bar and a mirror focusing system that just doesn't work.

I did not have any problems with collimation of the HS itself, and the scope held collimation surprisingly well. EdgeHD, interestingly enough, was made with input from Europe where a good mirror lock was conceived and presented to Celestron. They did, however, change it in order to cut costs. Personally, I find mirror locks evil as they prevent fully automated use.

I hate spikes nowadays so the great scopes like the AG-12 is not on my list. Instead, I keep thinking about the spider-less reflector... I don't know how much you would lose in optical quality if a simple clear glass of high quality was used as a secondary mirror holder, like in a Schmidt or Maksutov design.

Back on topic, the Hyperstar does, in my view, have a major drawback: the placement in the central obstruction. It's OK if you have an M26C that doesn't add to it, but the prospect of a filter wheel is pretty far off. And the spikes...

/per

This all makes sense. Gedan's link, though, does resolve the spikes issue; http://www.werbeagen..._hyperstar.html 

I've been arguing the case for optical windowed fast Newts for a while. The Tak Epsilon with a window would be my ideal starting point. At short FL spikes are out for me too. Far too many of them and far too many small square stars. A spikeless Epsilon would be quite something. F2.8, full frame coverage, small spot sizes... Bit of a beggar to collimate but it can be done.

Olly

[sharkmelley]

The Tak Epsilon is my main workhorse - the spider is big and chunky - it needs to be to maintain accuracy for the f/2.8 optics.  Would a glass disc provide the same level of support?  I'm just asking the question - I'm not an engineer so I have no idea of the answer.

Mark

[ollypenrice]

The Tak Epsilon is my main workhorse - the spider is big and chunky - it needs to be to maintain accuracy for the f/2.8 optics.  Would a glass disc provide the same level of support?  I'm just asking the question - I'm not an engineer so I have no idea of the answer.

Mark

I'm sure it would. It might even be better. It's a common enough system in catadioptrics and the fast Riccardi Honders. It also keeps the mirrors clean.

Olly

[Peter Drew]

We used to make 12" F3.5 Newtonians and others with optical quality windows. The windows were heavy and expensive but they eliminated spikes and were very rigid as far as collimation was concerned and kept the optics clean. Downsides were longer cooling times if not kept in an observatory and the dew magnet problem. I have a "spare" 12" optical window if anyone is interested.    

[Zakalwe]

What about the Riccardi-Honders design?

http://www.pbase.com/boren/officina_stellare_riccardihonders_veloce_rh_200

Very, very fast (f3). No moving mirror malarkey.

Pricey though.

[ollypenrice]

What about the Riccardi-Honders design?

http://www.pbase.com/boren/officina_stellare_riccardihonders_veloce_rh_200

Very, very fast (f3). No moving mirror malarkey.

Pricey though.

And a bit tricky. I tried to help an owner acheive an orthogonal chip and we failed. It really was ultra sensitive and probably best done on a bench, even an improvised one. A formidable device once fettled, though, I bet.

Olly

[Zakalwe]

And a bit tricky. I tried to help an owner acheive an orthogonal chip and we failed. It really was ultra sensitive and probably best done on a bench, even an improvised one. A formidable device once fettled, though, I bet.

Olly

I guess at f3 there isn't much room for error. Similar to a Hyperstar.   Ye cannae beat the laws o' physics laddie ...". 

[ollypenrice]

I guess at f3 there isn't much room for error. Similar to a Hyperstar.   Ye cannae beat the laws o' physics laddie ...". 

Which is why, whenever at all possible, I refract!   

Olly