Mirror Cells

[Astralstroll]

No sooner do I decide I know what I want to build with regard to a mirror cell then I read another article or thread and change my mind. Assuming that I go for a reasonably thick 16". Mirror, I have seen recommendations from 3 point to 18 point, wood or aluminium construction, a variety of ways to attach the mirror to the cell and so on.

Maybe I will build the rest of the scope first and ponder this some more!

Dave.

[swamp thing]

"Reasonably thick" is not a very accurate measurement

What kind of scope are you thinking of building?

For a 16" mirror of 1.5" thickness, I would go with an 18 point cell. For a 2" thick mirror a 9 point cell should be okay.

[Peter Drew]

I have a 17" 1.75" thick plate glass mirror sitting happily on a circular mat of swimming pool cellular polythene cover which is simillar to bubble wrap but having much stronger bubbles. The mat is secured to an adjustable backplate for collimation.

[Astrobits]

Well, Dave, you've seen my 16" in it's mirror cell, held in with three blobs of silicone. I'm not one for making anything more complicated than necessary unless I'm going for decoration not function. The more complicated the construction the more bits to go wrong. If you don't use silicone then the only other option that does not use edge clips is the sling. Personally I don't like edge clips at all. Not only do they introduce a bit of diffraction, they have the potential to damage the edges of the mirror, accelerating the corrosion of the Al coating at best and scratching at worst. There is a way of avoiding the edge clip problem and that is to make a groove around the edge of the glass, 2-3mm deep somewhere around halfway between the two surfaces, into which the clips fit.

Peter, I had a problem years ago with bubble wrap introducing astigmatism on a mirror I was making. I didn't expect bubble wrap to have any such effect on a 40mm thick 500mm dia. bit of glass but it did. If you have any astigmatism showing up in your images check out the pool bubble wrap under your mirror.

Nigel

[Moonshane]

I think it depends to some extent on how you'll be transporting the mirror. my 16" is in a solid tube and it sits on its side when being transporting to I needed mirror clips and a good 18 point cell which I made.

if it's just being transported level and in normal use then silicone is (presumably) OK.

In general if making a cell, I'd say that 6mm ali / 3mm steel for the mirror support triangles, and 12mm ali / 6mm steel for the pivot bars should be fine. you can use plywood too but this needs to be thicker still I think.

[Peter Drew]

Nigel. No support problems with the 17" in over 30 years. I tried to point out the difference between the material I used and ordinary bubble wrap, the bubbles on mine cannot be deflated by finger pressure however hard you try!. I mounted my 30" 1.75" thick mirror on the same material without problems. The mirror does also have a 3" diameter perforation and this sits on a boss with a rubber "O" ring fitted to the mirror backplate. The majority of the weight (80 lbs) is carried by a sling, no front mirror clips.

[Astrobits]

Peter, I don't think that my problem came from deflated bubbles as the astigmatism in the mirror was triangular and bubble wrap has a triangular pattern as I discovered after the event. Anyway it's good that yours has lasted so well.

Moonshane, I can assure you that when the silicone is cured you can hang the mirror cell upside down, with extra weights on the mirror if you want, and it won't come off. I plan to do a little demo sometime to show just how tenacious silicone is. In fact it can be a bit difficult removing the mirror from the cell for re-coating etc. Mine has endured going up and down a rough path to my observing site ( wheelbarrow style ), travelling 400 miles by car to an exhibition, and shipping 9000 (?) miles from South Africa without moving from it's original position on the cell.

Nigel

[Moonshane]

cheers Nigel - very good to know - but was it clear silicone or coloured?? (I refer of course to our earlier discussion!)

maybe I need to change my views a bit. this forum is superb.

[philthy]

Nigel, Is your 16" mirror siliconed into a wooden cell?

I ask because I have seen somewhere that is not recommended for mirrors > 12"

[Astrobits]

It is siliconed onto an MDF disc ( the same size as the mirror ) with a bit of epoxy on the MDF where the silicone is. I use a 3mm (1/8" ) spacer so that I can get a saw in to cut the silicone when the time comes to have the mirror re-coated. The silicone that I have always used is the clear, and I look for the ones that specifically mention greenhouse use, i.e. sticks to glass.

Nigel

[swamp thing]

It is siliconed onto an MDF disc ( the same size as the mirror ) with a bit of epoxy on the MDF where the silicone is. I use a 3mm (1/8" ) spacer so that I can get a saw in to cut the silicone when the time comes to have the mirror re-coated. The silicone that I have always used is the clear, and I look for the ones that specifically mention greenhouse use, i.e. sticks to glass.

Nigel

if I'm reading this right, that's a 3 point cell then ?

[Astrobits]

if I'm reading this right, that's a 3 point cell then ?

Correct. Three blobs about 1 1/2" dia. and about 40-50% out from the centre of the mirror. With 8" and 10" mirrors the blobs should be about 1" dia.

Nigel

[Astralstroll]

If you can make a 3 point cell using silicone blobs, would it be possible to make a 9 point too?

[swamp thing]

So, Skywatcher, Meade, celestron, orion optics, in fact all telescope makers have it wrong then. All we need is a bit of MDF and a tube of silicone?

No ones noticed this before so all that talk of mirrors biscuiting and needing proper support is hot air and nonsense. We can simply stick it to a sheet of MDF with three blobs of silicone.

[azzythehillbilly]

I am even more confused than before I started to read this thread.

Azzytyhe hillbilly

[philthy]

I'm going to convert my GSO 12" Dob to a truss version at some point in the future, and was considering using a wooden disk with silicone for the mirror cell.

Now, if it's good enough for a 16" then there's no question about 12".

[Astrobits]

So, Skywatcher, Meade, celestron, orion optics, in fact all telescope makers have it wrong then. All we need is a bit of MDF and a tube of silicone?

No ones noticed this before so all that talk of mirrors biscuiting and needing proper support is hot air and nonsense. We can simply stick it to a sheet of MDF with three blobs of silicone.

In a production facility there is not the time to carefully leave the mirror sitting around for a day or so while the silicone sets so they resort to CNC machined components that can be made and assembled in a matter of minutes. Using cheap labour a complete mirror cell is probably not much more expensive than three blobs of silicone anyway. Having a layer of epoxy on one side and glass on the other, the silicone does not set so fast as it would in a bathroom or greenhouse situation. I have arbitrarily left mine for 24 hours ( little more than overnight ) but have not checked just how long it really needs. This is something that I will have a look at when I have built my new workshop.

Nigel

[swamp thing]

In a production facility there is not the time to carefully leave the mirror sitting around for a day or so while the silicone sets so they resort to CNC machined components that can be made and assembled in a matter of minutes. Using cheap labour a complete mirror cell is probably not much more expensive than three blobs of silicone anyway. Having a layer of epoxy on one side and glass on the other, the silicone does not set so fast as it would in a bathroom or greenhouse situation. I have arbitrarily left mine for 24 hours ( little more than overnight ) but have not checked just how long it really needs. This is something that I will have a look at when I have built my new workshop.

Nigel

How about the small manufacturers that don't have cheap labour, CNC machines etc why are they still persisting in using "PLOP" to design their mirror cells to make sure they give adequate support when according to you it's simply not needed. All. We need is some MDF and silicone?

[Astrobits]

How about the small manufacturers that don't have cheap labour, CNC machines etc why are they still persisting in using "PLOP" to design their mirror cells to make sure they give adequate support when according to you it's simply not needed. All. We need is some MDF and silicone?

I have not invented three point bonding with silicone but have used it following various articles in Sky and Telescope in past years. From memory ( I can't find those articles now ) the three point support allows some deformation of the mirror but the deformation is spherical and simply changes the focus a little. The optimum position for the three points was at the 40% zone.

The use of silicone to bond the mirror was mentioned in other reports.

After trying it out I have used the technique for many years now without problems. I don't know how large a mirror the method will work with, I will be making an 18" x 1" thick mirror and I will test it out to check it's performance.

The use of MDF is optional, you can use whatever you wish, Aluminium, steel, plywood. It would be best that you check the bonding to the material of your choice before relying on it.

As for as the "small" manufacturers which manufacturers are you referring to? As far as I am aware there is only one manufacturer in the U.K. that makes both the telescope and mirrors ( or at least they used to ). Others just either make the telescope or the optics. Most telescope makers would, I assume, make the telescope before receipt of the mirror so a simple multipoint with mirror clips can be made in advance and the mirror installed as soon as it arrives and the telescope prepared for delivery. If I was making telescopes with mirrors from another source that is the way I would work so as to not have a costly item hanging around my workshop longer than necessary.

Sometimes new methods take some time to become accepted.

How many others here have mentioned sticking down the main mirror with silicone?

If most contributors here decided to start making telescopes which type of mirror cell would be used?

How many have advocated the sling method to avoid mirror clips?

The old ways are still the choice of the majority, including small telescope makers.

Finally, I have found that it works for me and I have tried to give as much info on my technique as possible so that should you wish to try it you can.

Nigel

[Glasspusher]

Interesting thread. I am no expert with the mirror cell design software but a very quick run through PLOP shows a 12 inch F5 mirror (25mm thick) with a 3 point support shows a deformation of 1/11 wave PTV, a 16 inch F5 25mm thick with 3 points shows 1/3.4 waves PTV deformation.

John

[Peter Drew]

I'm a believer in keeping things simple provided that the results are good. For many years at Astrosystems we bonded mirrors from 4.5" to 10" diameter to an aluminium backplate. The trick was to bond a full size disc of good quality thin felt to the backplate using Evostick Timebond impact adhesive, letting it harden and then bonding the mirror to the felt in a separate operation. The felt ensured that no strains were transmitted fro the backplate to the mirror. This technique needed no clips and ensured long term collimation. During occasional demonstrations at astro societies members of the audience were invited to try and pull the mirror from the backing, can't remmber anyone succeeding. When aluminising was necessary, the bonding could be released by introducing some turps subs to the felt and leaving for 24 hours. I was using my 17" Dob last night which has the mirror resting on a disc of heavy duty plastic bubble material, the images were at least as good as the 16" SCT's also on site.

[Astrobits]

Interesting thread. I am no expert with the mirror cell design software but a very quick run through PLOP shows a 12 inch F5 mirror (25mm thick) with a 3 point support shows a deformation of 1/11 wave PTV, a 16 inch F5 25mm thick with 3 points shows 1/3.4 waves PTV deformation.

John

From the original parabola. Re-fit a parabola to the new curve and you get much less "error".

Nigel

[cs1cjc]

My reaction is that the RMS value is more important than the PTV. John suggests Plop, http://www.davidlewistoronto.com/plop/ which is indeed a wonderful tool for modelling induced mirror error from point supports. Other support systems cannot easily be modelled so any induced error is unknown.

For a 300mm diameter, F/5, 25mm thick pyrex mirror, with a 63mm secondary, Plop calculates that 3-point support (each at about 0.43 of the radius) gives an RMS mirror surface error of 1.8E-5mm which looks unacceptable. A 6-point support, the next simplest, 3 circumferential beams each with 2 supports to give 6 equally spaced supports at about 0.6 of the radius, reduces this error by a factor of nearly 10. This is much simpler to construct than an optimised 9-point support which would give only a modest improvement.