Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Concrete Mirror?


jamespels

Recommended Posts

OK, before you laugh too loud, let me explain...

I have been trying to think of ways to make a very large mirror (meter scale) without the huge expense or weight of glass. It has been an interesting thought experiment that went through optical grade plastics (like the polycarbonate used in spectacle lenses), through GFRP / CFRP skinned foam (still thinking about that for a radio dish but cannot think how to get much better than mm accuracy - great for Ha but useless for optical!) and suddenly hit on concrete in a slightly silly conversation with a colleague.

Originally suggested as a joke, I am not so sure it is out of the question:

1. concrete is a ceramic, like glass

2. concrete can be moulded to avoid the huge thickness at the edge

3. it can be "roughed" with structural concrete (possible lightened further with hollow cores), then fine finished with a self-levelling compound or similar. Spun, the self-leveller will even form a reasonable parabola...

4. it can be ground in a very similar way to glass, without the nasty glass power being produced

5. it has very good thermal stability, much like glass.

6. Being thinner at the edges than an equivalent glass mirror, it will have much less thermal mass and take much less time to return to shape after a temperature change. The shape could even be calculated to maintain a parabola as it warms / cools.

7. mounting plates can be formed as part of the concrete structure so the mirror can be attached directly to the cell - not sure if this is an advantage or not but it does remove the need to mirror clips.

8. Material cost for meter scale should be c £1,000, not £10,000!

Issues that need to be resolved:

1. Getting a hard surface should be possible with the right compound but it will need to be sealed very well. 

2. Finding the right shape to keep it light but strong; ideally it will be shaped so that it maintains a near parabola as it expands and contracts

3. Finding the right stuff to grind it with - gut feeling is that the same material used for glass should be about right.

4. I know nothing about grinding mirrors, other than what I have learned from excellent presentation from John Nicholl two years ago

5. How to get it silvered? 

QUESTIONS:

1. Considering the above, is there a good reason why this simply will not work?

2. Has anybody tried this before and, if so, what did you learn?

3. Any ideas for addressing the issues listed?

Very interested in trying this as an experiment with other people in the Stoke / North Staffs / East Cheshire area - let me know if you are up for it. If I go ahead with this, the first step will obviously be to try it out with a 6" or 8" mirror. No real advantage at this scale but there is no way I am trying a meter-scale mirror without knowing what I am doing!

Thanks,

James.

Link to comment
Share on other sites

  • Replies 28
  • Created
  • Last Reply

What an interesting idea, I wouldn't have thought concrete alone would be able to achieve a smooth enough finish but then there are all sorts of coatings that could be applied to get around that.

Large observatories sometimes have a machine for aluminising large mirrors on site rather than transporting them, it may be a place to look for this.

TSED70Q, iOptron Smart EQ pro, ASI-120MM, Finepix S5 pro.

Link to comment
Share on other sites

1) Is glass a ceramic?

2) So can glass be moulded, most glass is moulded.

3) There are already hollow or sandwich mirrors for faster cooling (Hubble Optics)

4) Doubt 4 is correct, concrete is a mix of cement and aggregates and is not as homogeneous as glass.

5) No reason to question 5, except is it better or worse?

6) Why is it thinner at the edges, nothing so far would imply that. Visit a Sky Tower and they put 2" thick glass in the floor the stand on and look through, they say this is stronger then the 6" of concret the platform is made of so a concrete mirror may need to be significantly thicker. Also the absorbtion/loss of heat may be lower meaning increaded time to equilibrium.

7) Not necessary aa good idea, if the figure goes wrong you could throw mounting and "mirror" away.

8) House building concrete might be cheaper but you are likely not actually talking about that.

Strange thing is concrete seems never to actually set off fully, there is an arguement that some roman concrete is still hardening after 2000 years.

Issues to be resolved:

1) Suspect you are looking at casting, grinding, sealing, resurfacing then a final grind of the final surface - 2 grindings, the initial concrete base and 2 additional layers of material.

Link to comment
Share on other sites

I don't think concrete is going to be the best idea. Though it has a ton of compressive strength its lateral stength is its weakness, thats why they use rebar in it. Especially when it gets thin. So thinning at the edges would only increase the likely hood of it chipping/flaking. It is also a very porus material. So it will absorb and hold moisture, even after it is set and cured. You could seal it but Im not sure how effective that is or isnt.  Also grinding to the tolerances needed would be very difficult. You have several different material types within concrete itself. Mainly the aggregate and the motor but there are several different types within each of those. So you would come across materials that would take more to grind than others. i.e trying to grind away a rock vs the motor. Now you could get concrete with less and smaller aggregate BUT that then makes the concrete weaker. Not that its weak just that its weaker than using large aggregate. Theres also the expansion and contraction issues that arrise since the concrete is so poarus. Concrete expands and contracts a lot. not to mention how heavy it would be. I dont think it would be lighter than glass would it? I definately agree cheaper though. Thats my personal take on it anyways.
 

BUT you are not far off. There have been mirrors made from granite. I'm not sure how large but I know I've seen them. Definately look into that if you havent already. Obvisouly just like the concrete it would require some type of reflective cloating to be put on it.
 

Link to comment
Share on other sites

Many thanks for the feedback - this is really helpful.

To clarify a few things:

1. Concrete strength and thickness: I agree that the concrete could chip easily but the reason that the edges can be thinner than glass is that glass mirrors are typically ground from solid disks and need to maintain the flat back to fit the cell nicely. With concrete, you have alternative mounting options (though comment about this above noted) and you can make the edges as thick or thin as you like. There is no reason not to reinforce the concrete - not with rebar (!!!) but lightweight fibre options are around.

2. Variable hardness: I would plan to make the main structure from fairly typical concrete to a rough shape (possibly conical or stepped inside shape but this is conjecture so far), then spin it with a layer of hard self-leveling compound. These have fine grain with a high cement content so tend to be very hard but rely on the substrate for strength - hence the two layer finish. This surface layer will be the most critical and I would minimise work on it by spinning at the right speed to firm a paraboloid of roughly the right shape.

3. Porosity: Something I had not thought of bit concrete is used for factory and kitchen floors so it must be possible to seal it. Will need to consider this further.

4. Is glass a ceramic? Turns out the answer is yes, but that ceramic can mean practically anything other than metals and alloys! That said, this seems to come down a lot to how you define ceramic - I was using it in the sense that quite a few properties are very similar but will concede the point if pushed!

5. Granite has been used before: I did not know that so thank you! From a builder friend, granite and concrete work tops are finished and polished in a similar way so that is at least not negative!

6. Would concrete be lighter than glass? If the mirror was cast as a solid disk and ground in the same way, then no. However, I would plan to shape the structural concrete backing carefully to remove weight - this will almost certainly mean reinforcement is needed (see above).

Thank you all for the feedback so far...

James.

Link to comment
Share on other sites

Greetings

how about hybid concrete? use a high % of silica sand and fire the whole mould like engineering brick in a kiln this would give most of the required properties... would it not??

Andy ( just another daft solution from the dark side )

Link to comment
Share on other sites

Have you thought about segmented mirrors? I think this would be easier to accomplish with granite than it would with glass. Though it would still be a tough build. Though I think I would find the challenge very fun.

Link to comment
Share on other sites

Greetings

how about hybid concrete? use a high % of silica sand and fire the whole mould like engineering brick in a kiln this would give most of the required properties... would it not??

Andy ( just another daft solution from the dark side )

Have you thought about segmented mirrors? I think this would be easier to accomplish with granite than it would with glass. Though it would still be a tough build. Though I think I would find the challenge very fun.

Interesting this, glaze on pots is that like glass, can segmented lumps of clay or porcelain be glazed then polished back to shape then silvered,  cannot see large lumps fitting into a kiln.  Great thread - thanks  :grin:  :grin:  

Link to comment
Share on other sites

With fibre based mirrors the pattern causes issues, known as 'print-though'. There's research into using carbon fibre as a mirror base then putting thicker layer of epoxy etc on top that's polished to the final shape. One different form of this is using carbon dust and gluing together.. however that's not giving you a robust end product.

I was looking at if a carbon fibre mirror could be used for a large mirror for a dob/nasmyth mount (http://en.wikipedia.org/wiki/Nasmyth_telescope).. 

Link to comment
Share on other sites

I have more or less dismissed CFRP / GFRP (carbon / glass reinforced plastic). The main issue I can see is that it is difficult to get the surface stiff enough to sand / grind to the right shape. It is right up there for a large radio dish (I was thinking of a 3m dish capable of mapping space at the hydrogen line - this would need a surface accuracy of around a millimeter to get the required sensitivity, which I did think was achievable. Had to abandon that idea when my wife found out :) and when I realised just how radio-noisy it is round here). One meter is a lot more "practical"!!!

NickK - the Nasmyth design is exactly what I was thinking of if it ever gets that far, with the focuser at or near the hinge line. At F4, a 1m mirror would need a massive structure for a dob layout that would be a nightmare to make stiff enough and the EP would be two stories up!!! 

Mick J - not sure if it was intentional or not, but glazing & baking it is another option. I am on the edge of the "Potteries" so I must be able to find somebody who could help with the glaze and a kiln to bake it in...

Link to comment
Share on other sites

Have you considered polymer concrete?

Its not really concrete its just one of its names, epoxy granite is a more accurate name.

Its actually epoxy resin filled with lots of different filler materials to give it the right characteristics. There are some very expensive machine tools using it for its high dimensional stability, and vibration damping capabilities.

I could imagine using it to cast a rough parabola and then having a thin layer of pure epoxy on the surface to make polishing easier, if you picked the right composite ingredients it could be pretty thin and strong and wouldn't have the water problems of concrete (its essentially a plastic).

Concrete also continues to age and move for many years, especially large pieces.

its an interesting idea with the concrete though, it certainly got the neurons firing :)

Link to comment
Share on other sites

I can't see how spray silvering a first surface mirror would work for an astronomical quality reflection. You could paint the surface of a second surface mirror with a broom, it would be the interface between the silver and the finish on the glass surface which would form the reflective quality. For a front surface mirror the coating has to conform exactly to the optical surface and not deviate in thickness. As far as I know this can be achieved only by traditional silvering or vacuum aluminising.  :smiley:

Link to comment
Share on other sites

At a guess, I would have to presume that these Angel gild people have a product that will form an even surface. This may not be exactly the same surface as the glass (surely it will pool to a degree?) but a sub- millimeter level over a 1.8m mirror probably will not be that noticeable - would be very interesting go see a Strehl (sp?) rating for it, though. 

Got to say, while it is impressive, that size is just to big for me. 1m at f4 works because it will cope with "normal" EPs and a Naysmith / Coude layout gives a manageable structure.

Doesn't solve the problem of silvering a concrete mirror, though - I really wouldn't want to be recoating every 6 months!

Link to comment
Share on other sites

A meter scope is not unusable, an American trucker made his own 1.8m f6 recently, works fine, just not very portable. http://m.slashdot.org/story/193923

If you mount it correct then you can work with quite thin large glass as long as it doesn't have internal stresses. Some guys in France have made some meter class mirrors.

The porosity and anisotropic on hardness will kill you, you would be likely to get scratches all over the place that you'd never shift.

I have heard about the possibility if molded mirrors which might enable low cost large mirrors to be made. Thermal expansion is likely to be an issue, though if it isn't too massive it might thermalise quite quickly.

I have come across concrete and telescopes... A bloke I knew made a concrete mirror blank to help test the structure as it has a similar density to glass.

Cheers

Peter

Link to comment
Share on other sites

Your mm scale is out by at least 5000 times. Your mirror surface would have to be accurate by at least a quarter of the intended shortest wavelength. 

At red that means <= 0.18um approx

At blu that means <= 0.1um approx

Link to comment
Share on other sites

Your mm scale is out by at least 5000 times. Your mirror surface would have to be accurate by at least a quarter of the intended shortest wavelength. 

At red that means <= 0.18um approx

At blu that means <= 0.1um approx

Fair point - I was thinking that if the surface was smooth it would be OK but of course not - some will be in phase and some out otherwise. How would this look in the EP? I guess it would form an image but I cannot imagine what it would be like - colour rings? some light & some dark patches?

Link to comment
Share on other sites

You would have an unwanted random interference pattern (light/dark) if it was slightly out but mostly I guess the image would be simply fuzzy due to the light not being focused properly. Depends how far out it is as to what effect you'd see really.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.