From: "James Lerch" 
To: "ATM List" 
Reply-To: "James Lerch" 


Hi Michael,

The simple answer to your questing is YES!  In practice I would HOPE that
professional coating shops like Clausing and Spectrum have setup their
chambers so that the effect is minimized.  (BTW, I'd love a tour of their
chambers! :)

As far as empirical data,  I've read that a nominal single layer metallic
coating is ~100nm thick

In playing with my home made coating system here's some experiences I've had:

#1 A single on-axis evaporator 10" from mirror surface (on a 10"
F/5) should
have caused a coating thickness change of ~19%, with the center being high.
Assuming a reference wavelength of 550nm, and a center coating thickness of
100nm, that would mean the surface figure was altered (19/550 = 1/29 p-v
surface = 1/14.5 p-v wave-front)  Visually and on the Robo-Foucault this
pretty much appeared to be what happened..

#2 I now use a 12" diameter ring of 6 filaments, about 5" above the mirror.
This results in ~1.5% thickness deviation IF I keep the mirror surface
5" away from the plane of the ring.  If I move the mirror closer to
the ring, I add additional material to towards the 70% radius of the
mirror.  Moving the mirror further away adds more material towards the
center of the mirror.   I
have been known to use this trick to 'bully' a figure around a little (a
very little!)

#3 My next chamber will have two rings of filaments:
    A) Ring 1 will have a diameter of 12" and 5 points
    B) Ring 2 will have  a diameter of 24" and 10 points

    The plane of both rings will be ~4 inches away from the mirror surface,
thus affording the option of only needing a 24*24*8" rectangular
chamber that will have a ~3% coating thickness change (if all my math is
correct...) The up side to this arrangement is keeping the chamber small,
shortening the mean free path between filaments and surface, at a cost of
having to load and fire 15 filaments for each coating!


If your really bored, I have written a 'Near Field Evaporative Simulator'
that I use to help me figure all the above data out.  Your welcomed to play
with it by downloading it from here:
http://lerch.no-ip.com/atm/nearfieldsim2.zip (10kb) This version will do up
to 3 Evaporative rings, and models up to a 150mm radius mirror.  I have
another version that will do up to 600mm radius mirrors, but its a CPU/MEM
hog from hell!


Take Care,
James Lerch
http://lerch.no-ip.com/atm (My telescope construction,testing, and coating site)
----- Original Message -----
From: "Michael Burr" 
To: "ATM list" 
Sent: Thursday, March 27, 2003 10:01 AM Subject: ATM coating affects figure?


>
> Greetings:
>
> I've searched the archives and found little information about this, so I
> thought I'd ask the list.
>
> Does anyone know of empirical information regarding the thickness of
> aluminum coatings, and how variances in thickness from the center to the
> edge of a mirror might affect the mirror's figure?
>
> I've seen oblique references to aluminum coating being 1/2-wave thick. If
> that's the case, and there's a 15 percent difference in coating thickness
> from center to edge on a 16" mirror, does that mean the coating
effectively
> undercorrects the figure by 7.5% (i.e., 1/2 times 15%)?
>
> If so, would I be completely nuts to try and "tune" the figure to the
> coater? In other words, I could try to overcorrect the figure by 7.5%. It
> seems to me a better solution would be to use a coater that distributes
> aluminum more uniformly, but it's an interesting theoretical question
> anyway.
>
> Thanks
> MTB
>

--- BBBS/NT v4.00 MP