[ATM] Fixing edge zones

[Richard Schwartz]

I think the problem is one of visualizing the error and modifying the  
stroke to correct that error.  If you have a null test for the desired  
shape, the error visualization is not a problem, so there is no reason  
to go back to a sphere.  Getting to the desired surface is no more  
difficult than getting a sphere with a primitive tester like those  
described in the authoritarian elitist literature of the 1930's.

Sent from Richard's iPhone

On Oct 8, 2009, at 19:21, "Mel Bartels" <mbartels at bbastrodesigns.com>  
wrote:

> I consider this good advice because often, in the end, you get to the
> finished figure quicker if you start with a spherical smooth  
> beginning.
> Going back to spherical is always good advice because you can't go  
> wrong
> with this approach.  Continuing to fix a figure with major problems  
> can be a
> black hole that will suck your enthusiasm across the event horizon!
>
> Mel Bartels
>
>
>
> -----Original Message-----
> From: atm-bounces at atmlist.net [mailto:atm-bounces at atmlist.net] On  
> Behalf Of
> Mike Coslo
> Sent: Thursday, October 08, 2009 6:09 PM
> To: atm at atmlist.net
> Subject: Re: [ATM] Fixing edge zones
>
>
>> I'm in the final stages of figuring my first 8" mirror.  I first  
>> over shot
>> the parabola by quite a bit, but with some help from the experts, I  
>> was
> able
>> to get close to a parabola using COC strokes.  Using figureXP, my
> wavefront
>> error is about 1/3 wavelength.  I need to increase the ROC of zones  
>> 3 and
> 4
>> (I'm using a 4 zone mask). My question is what type of stroke do you
>> recommend using to knock these zones down? Any help would be awesome!
>
> This is non-standard advice, but you may want to go back to a sphere,
> and then take it to a parabola. Occasionally a mirror can be  
> overworked.
>
>     - Mike -
> _______________________________________________
> ATM mailing list http://www.atmlist.net/
>
> _______________________________________________
> ATM mailing list http://www.atmlist.net/