Re: [ATM] three flats was (no subject)

[<jsanchez@skipanon.com>]

Assuming that you have some sort of viewer that shows 
fringes and a reference flat of known good quality (ideally 
of 1/10 wave accuracy), knowing if the tested flat is 
concave or convex is no big deal. As someone in the group 
already said, place the flats so as to display a fringe 
bullseye and press the center. If the fringes move away from 
the point of presssure, then the flat is convex. Otherwise 
it is concave.

This works as long as the difference is of at least one 
fringe. If less, then you have to observe the direction in 
which the fringe lines bend (the bow). Since fringe lines 
always move towards the wider air gap, you can determine if 
the flat is concave or convex by observing the direction in 
which the bow lines move.

Sometimes a flat has a sunken or a raised central area 
(doughnut) and shows concavity and convexity at the same 
time. In this case the straight lines show what opticians 
call "the bird." The image of straight fringes resemble the 
wings of a seagull in flight. The solution is the same as 
when polishing a telescope mirror with a central doughnut or 
a raised center and a turned edge. That is, find a polisher 
or a stroke that improves the figure and polish away.

My experience is that flats tend to go convex during 
polishing. If you have a good spherometer you can try to 
leave the flat slightly convex after fine grinding so that 
the figure will be corrected to flat during polishing. How 
much is hard to say because it depends on the size of the 
flat, the thickness, the material, and even the spirits that 
float in your workshop.

Some say that nothing is harder to make in optics that a 
good flat. If you can do that, you can do anything.

Here is a secret: don't even try to make flats out of plate 
glass. Pyrex is good to about 1/10 lambda in flats 6" or 
larger. For anything better than that you need fused quartz 
or Zerodur.

Enjoy the trip. Regards,

Julio 



---- Original message ----
>Date: Mon, 7 Nov 2005 08:32:23 +1000
>From: "Thomas Janstrom" <tomjan@ozemail.com.au>  
>Subject: RE: [ATM] three flats was (no subject)  
>To: <jsanchez@skipanon.com>, <atm@atmlist.net>
>
>Hi Julio,
>
>I already have that program, and yes it works a charm! But 
my real
>problem is judging weather the plates under test are 
concave or
>convex....
>
>Clear Skies, Thomas.
>www.tjanstrom.com
>www.norsewines.com.au 
>
>-----Original Message-----
>From: atm-bounces@atmlist.net [mailto:atm-
bounces@atmlist.net] On Behalf
>Of jsanchez@skipanon.com
>Sent: Monday, 7 November 2005 3:21 AM
>To: atm@atmlist.net
>Subject: [ATM] (no subject)
>
>There is a small, free, program on the Web that allows 
>entering the fringe data for three flats and returns the 
>error of each one of them. It works like a charm and saves 
>the effort of doing the calculations each time. The program 
>can be downloaded from:
>
>http://otterstedt.abflug.de/atm/three_planes.zip
>
>Regards,
>
>Julio Sanchez
>
>_______________________________________________
>ATM mailing list http://www.atmlist.net/
>
>-- 
>No virus found in this incoming message.
>Checked by AVG Free Edition.
>Version: 7.1.362 / Virus Database: 267.12.8/162 - Release 
Date:
>5/11/2005
> 
>
>-- 
>No virus found in this outgoing message.
>Checked by AVG Free Edition.
>Version: 7.1.362 / Virus Database: 267.12.8/162 - Release 
Date:
>5/11/2005
> 
>
_______________________________________________
ATM mailing list http://www.atmlist.net/