Re: ATM lies, damn lies, and mirror tests

[Bratislav Curcic <epabcc@epa.ericsson.se>]

Steve, 

> I have an 8" f/5 in my basement, so I took my laser collimator down to
> it and did a test.
> small and easily distinguishable 1/16" red dot.

Yeah, that's about right. At about f/600 (8" f/5 has 40" f.l.; 1/16"
at 40" f.l makes about f/640) Airy disc should be around 800 microns,
or 0.8 mm. Not that this is around 120 times LARGER than a diffraction
spot for your 8" f/5.

> Continuing with this logic, the
> small 1/16" red dot must be the Airy disk (or it and a few (many?)
> surrounding brighter rings), but so what? It's still easily
> distinguishable

Agreed. Even naked eye.

<<<< digression >>>>

> My, you presume a lot. I have, in fact, ground, polished and figured 3
> mirrors - a 4.25" f/4.5, and two 8" f/5s.

Oops ! I thought that Foucault veteran would immediately scream foul 
when transverse aberrations started to be expressed in 1/16th of an 
inch at focus.

Apology in order. Accepted ?

<<<< end digression >>>>

> I don't follow at all why you think the distance from the laser
> collimator to the mirror matters one bit.

Well, I took 2*F as I thought you'd have your laser pointer OUTSIDE
scope. In that case 2*F is MINIMAL distance laser beam will travel 
before coming to focus. Agree ?

> Since the laser beams are
> straight lines, it doesn't matter whether they originate from two ft or
> ten ft away.

Yep. Now think about this : let's place our laser pointer right at
optical axis. It hits very center of the mirror that we can safely
assume to be flat. Note your focussed 1/16" image on the screen or
better, take a CCD snapshot. Now, I park my car in your driveway
causing shifting of your house so the front of your laser moves 550nm
(ONE WAVELENGTH OF LIGHT!) relative to the back. This will displace our
spot at the focus 0.55 * 10 ft / 2 inch = 33 microns. Three CCD pixels !

I again remind you that displacement of just COUPLE of microns, depending 
on place of origin will deem your mirror 'junk'.

So this is the list of 'fatal questions' :

1) focused image of laser pointer is TWO ORDERS of magnitude larger
   than theoretical allowable diffraction disc of your mirror

2) slight angular movement of laser pointer will diplace your 'dot' at least
   an order of magnitude more than 1/8 wave error on the mirror would;

3) best resolution of commercial CCDs is only 9 microns; still larger
   than diffraction spot of 1/2 wave junk 8" f/5 can do (don't forget
   that 800 microns spot size !!!!)

4) you are still testing about 1/250th square inch of your mirror, twenty
   thousands more to go

But don't let me stop you ! They always say "learn from other peoples'
mistakes" but I seem to remember well only my own ones !!! 

Cheers,
Bratislav