To: 
From: MLThiebaux 
Reply-To: MLThiebaux 


>So, what this analysis says to me, is that a 1 wave error on the flat will,
>in essence,  convert a system with a perfect primary to one with a PV errror
>of 1/7 wave, which is equivalent to a 1/3.5 wavefront error.... am I
>understanding correctly?
>
>Puzzled,
>Scott Donaldson
>
>

Yes. The numbers speak and that's what I hear too.  But, hey, this is all
theretical fun.  The whole world is welcome to prove it wrong.

Going back to something you said earlier...

>I am skeptical, unless you are talking about hyperbolic with a sagitta
>around lambda/10.   I need to see the math on this.

Well, I wonder why hyperbolic seems to be a problem.  The hyperboloid of
revolution is the correct surface to use for a Newtonian diagonal (the
traditional plane is just a special case.)  The two are the only reflecting
surfaces that have finite real and virtual focal points.  If deviations
from the plane are accepted, a hyperboloid, figured and collimated
correctly, would work perfectly as well as a plane.

Somebody out there is probably thinking I am pushing hyperbolic diagonals.
Wrong.  If I had the choice I'd reach for the flat 10 times out of 10.  The
flat would be humongously easier to collimate.  On the other hand, not as
much fun.

This excursion into hyperbolic space was supposed to be an exercise in the
imagination to shed some light on why a curved "flat" might
actually work: it might be closer to a good hyperbola than a perfect plane.

Martial Thiebaux
Rawdon Hills, Nova Scotia

--- BBBS/NT v4.00 MP