Mirror cell questions

[kalasinman]

I know little about mirror cells, but I'm trying to learn. Other than the internet, I'm quite isolated from others who share this pursuit. Any help understanding some basic points would be appreciated.

What are the primary purposes for mirror cells?

 1. To hold the mirror in all telescope orientations?

 2. To facilitate collimation?

 3. To allow the mirror to be removed for service?

  Other?

Secondary purposes. Design features which may be specific to a certain design approach.

1. Would  a flat plate of some material satisfy 1-3 above and any other primary purpose?

2. If so, as the number and pattern of support points produces a pattern of distortion of the wave front, then why not attach the mirror to a  

    flat plate with a continuous application of silicone, or some other glue?

    If not, why not?

Thanks in advance for any contributions.--Jack

[Earl]

its best not to mount it on a flat surface as this can lead to distortions as it could well not be flat as could the under side of the mirror.

Most mirrors are only held at minimal points of contact or around the edge iirc.

[ronin]

It is for all 3 reasons.

Oddly seems the best is 6 points of contact, often better then 9 even if less points.

There was a site called PLOP, stands for PLate OPtimiser.

The site I had has gone but try this:

http://www.davidlewistoronto.com/plop/

Put "PLOP mirror cell" into google and a number of references come up.

[kalasinman]

@Ronin, are these the sole primary reasons?

I have PLOP. Putting the mirror on hard points seems to cause distortions.

@Earl, I don't follow  that a flat plate could cause distortions=bad, while charts of various mirror cell support schemes (which support on points) clearly make distortions=good/better. I'm not disagreeing, it just doesn't make sense to me.--Jack

[Earl]

Cells can be designed to the advantage of the system and even manipluate it and  avoid having to get perfectly flat surfaces which will not be effect by thermal movment and will not have gaps.

A cell has specific customisable control where flat surfaces do not.

[Zakalwe]

1. Would  a flat plate of some material satisfy 1-3 above and any other primary purpose?

2. If so, as the number and pattern of support points produces a pattern of distortion of the wave front, then why not attach the mirror to a  

    flat plate with a continuous application of silicone, or some other glue?

    If not, why not?

I don't know that much behind the theory of mirror supports, so this is a bit of a stab in the dark.

The problem with using a flat plate is that it would have to be very flat. As in optically flat. Plus it would have to be either very thick, or be extensively ribbed underneath to ensure that it doesn't go out of flatness. Plates like this are available, but are expensive. A good machine shop will have surface plates that are honed to a high level, often made of high quality steel or granite (they are so flat that gauge blocks will "wring" to the surface, forming a bond that is incredibly strong). The other option would be to use a mirror blank. The problems with this approach should be obvious, namely weight, cost and the need to have the underside of the mirror also flattened.

The other problem to consider is cooldown times as having a large mass directly against the mirror would slow it's cooldown times enormously.

You'd also have to consider the effect of temperature change and the different rate of contraction. if the mirror was bonded to a support plate then any difference in the contraction rates would distort the mirror. You could make the support plate out of another mirror blank to ensure that there was no difference in rates. Then you'd have more than doubled the cost (as you'd have to grind the underside of the mirror flat).

[kalasinman]

CTE is indeed a worthy consideration Zak. Thanks--Jack

[Peter Drew]

A lot of this boils down to the diameter and thickness of the mirror. Small mirrors, 4.5"-8" of reasonable thickness can successfully be mounted on a flat plate provided that it is fitted with non rigid support points, symmetrical blobs of silicone or bonded to a layer of felt. It's large thin or very thick mirrors that cause mounting problems and require much more sophisticated cell assemblies. Having said that, my 17" F5  1.75" thick mirror sits happily on a flat plate surmounted by a disc of heavy duty bubblewrap type material with no ill effect. 

[callump]

My 20" has bubble wrap as a supporting layer, and I think John Dobson used carpet for some of his scopes.

Callum