From: Mark Holm 
To: Ed Taychert 
CC: atm{at}shore.net
Reply-To: Mark Holm 


Hi Ed,

That is an interesting observation, I hadn't tried that path.  Maybe the
"classic" nine point cell is never optimum (with refocusing
allowed).

A troubling problem with most of Plop's nine point cells is that the
triangles come out very thin.  The altitude of the pivot point in
particular is pretty small.  The unconventional "inside out"
designs have the thinnest triangles, but
even the more conventional looking ones are fairly thin.  Getting decent
relative accuracy on that small dimension needs a level of precision hard
to make without a milling machine.

Mark Holm
mdholm{at}telerama.com

Ed Taychert wrote:

> Hi Mark,
> You say on your web page:
>
>  > All of the older 9-point cell designs arranged the triangles as shown
> in the first image below.
>  > Plop also returns this arrangement for thicker mirrors. [...]
>
> I think that this is more a function of the start-defaults of "automatic
> cell design."
>
> For example:
>
> If you do an automatic design with D=200mm, th=25mm, fl= 1000mm,
> plop will give you a 2.2 nanometer cell of the older type shown in the
> first image.
>
> If you then change the thickness to 15 mm and "run" from the
run menu (w/o
> automatic cell design), the support point will flop over and you'll get
> an example
> of the "newer" second image of a cell.
>
> If you then change the thickness back to 25 mm and run again, (again,
> from the
> run menu) you will now get a 2.0 nanometer cell of the new type ...
>
> ... I don't think that plop really wants to make the older 9 point
> cells, it's just that
> its optimizer doesn't always let it get there.
>
> - Ed.
>
>

--- BBBS/NT v4.01 Flag-4