Re: Re[2]: [ATM] Question about "Null-testing"

["David Harbour" <scarab2@cox.net>]

Gentlemen:

I just wrote you guys a massive treatise on what I remember of Tex
refiguring the McDonald 82" mirrors, and my hand brushed a hot key of some
kind, and it all disappeared. I'll try a shorter version here.

Tex first isolated the errors on the primary with a null test. (left the
mirror in the tuhe, acquired a bright star, nulled it with a straigh edge,
took a photograph of it; bit turned up edge, strange surface texture). Here
is how he isolated the errors to each surface, Richard.

Then he put the secondary back in, to see what its errors looked like
superimposed on  those of the primary; this procedure he repeated with the
Coude mirror. Thus, he isolated out what was wrong, where.

He carried the straight edge on two wires to insure perpendicularity and to
have a small obstruction.

The primary had a seriously turned up edge; he compensated for this and all
errors by figuring the secondaries alone.

It is curious that so little (so few people) believe that much can be done
with the knife edge beyond Foucault testing at center of curvature. He had
parallel light when only the primary was installed; he put the knife edge
near the prime focus. Then he had the Cassegrain focus, to get parallel
light from it for the knife edge. Same with the Coude mirror (I know you
understand this, Wihng Man). This is how he isolated where the errors were
and put all the corrections only on the two secondaries.

Sorry I lost the original I just labored over; my hand slipped and it
disappeared.

"Gross" errors can be measured at focus with a knife edge, Richard.

His computer, like Ritchey's, was between his ears (in the early 60s)

I wished I hadn't a;ccidently erased my first, book length treatise. But
that is what you get from Bill Gates.

Dave

P.S. The 82" was out whack by over one wave on the wave front. He restored
it, without interferometry, or Microsoft (or the apple) to one tenth wave
performance.

Short version, to replace much longer version, now in Billy Gates Limbo.

Dave

P.S.- Why don't one of you, with a large collection of Sky and Tel
magazines, go hunt for this article for us. I got rid of all my old Sky and
Tel magazines, slowly, after they tried to kill me once (wasn't their fault:
I should have had both magnetos on when I took off)-

Dave

----- Original Message ----- 
From: "Richard Schwartz" <richas@earthlink.net>
To: "'David Harbour'" <scarab2@cox.net>; "'ATM List'" <atm@atmlist.net>;
"'Bob May'" <bobmay@nethere.com>; "'Richard'" <cnc@cncservo.co.uk>
Sent: Tuesday, July 05, 2005 10:23 AM
Subject: RE: Re[2]: [ATM] Question about "Null-testing"


> I don't understand how this wonderful test can isolate the errors on a
> surface in an optical system that has several surfaces.   Why does the
knife
> edge have to be suspended on cross wires?    Is there a computer program
to
> reduce the data... and how do you get the data in the first place?
Please
> explain.
>
> I have a copy of Texereau's book right behind me on the bookshelf.  This I
> refuse to open until the government of France is overthrown and they
decide
> to join us in our holy crusade against the lesser nations and races.  (But
I
> would not refuse to go out dancing with the French girl.)
>
> . . . Richard
>
> -----Original Message-----
> From: atm-bounces@atmlist.net [mailto:atm-bounces@atmlist.net] On Behalf
Of
> David Harbour
> Sent: Tuesday, July 05, 2005 5:59 AM
> To: ATM List; Bob May; Richard
> Subject: Re: Re[2]: [ATM] Question about "Null-testing"
>
> Gentlemen:
>
> Null tests not requiring auxiliary optics will not lie to you. At the risk
> of repeating myself, Texereau was one of the last men to demonstrate this
> when he used a knife edge suspended on cross wires to refigure all of the
> mirrors of the McDonald observatory's 82" optics; if memory serves me, one
> mirror was out by one and a quarter waves at the focal plane and the other
> two by one half and or three quarters. Using his knife edge, and a bright
> sstar (no auxiliary optics) he nulled tested the mirrors back to
> beautifullly near perfect curves (somewhere as near as I recollect, about
a
> tenth wave on the wave front). It seems to me that the only way this test
> could go awry is if someone couldn't figure out where the focal plane is,
or
> couldn't interpret the shadows. And that is not a case of the test lying
to
> you; that is the case of you being a dummy.
>
> Why do atms resist this wonderful test, which is so easy to do (tape a
razor
> blade across the opening of the eyepiece draw tube?). Does anyone have a
> theory they would like to elaborate for us why atms seem to avoid this
test,
> so much more sensitive than a Ronchi grating over the eyepiece opening
> instead?
>
> Dave
>
> P.S. The Ross test nulling lens can lie through its teeth, sometimes, if
you
> don't get certain things right. The other only requires a star; all of
them
> produce parallel light, which works so well for null tests made at focus.
>
> ----- Original Message ----- 
> From: "Richard" <cnc@cncservo.co.uk>
> To: "Bob May" <atm@atmlist.net>
> Sent: Monday, July 04, 2005 12:58 PM
> Subject: Re[2]: [ATM] Question about "Null-testing"
>
>
> > Hi Bob,
> >
> > Monday, July 4, 2005, 6:47:31 PM, you wrote:
> >
> > BM> Like any other test, the Ross and other null tests can lie to you if
> you
> > BM> allow them to do so.  The test is still a variation of the Foucualt
> test
> >
> > How so?
> >
> > BM> Using interfrometery (there's a Yahoo group for that which is fairly
> active)
> > BM> that will show errors a lot better than what the geometric tests
like
> > BM> Foucault and so forth.
> >
> > Exactly, but that's the way a Ross null test is done. I assume
> > you know of a method of using a Ross null lens with a Focault
> > tester that I don't. If that is the case, the immediate question
> > it would pose to me is what effect would the off axis effect of
> > Focault have on the results?
> >
> > -- 
> >
> > Best regards,
> >   Richard in the UK
> >
> >
> >
> > _______________________________________________
> > ATM mailing list http://www.atmlist.net/
> >
>
>
> _______________________________________________
> ATM mailing list http://www.atmlist.net/
>
>


_______________________________________________
ATM mailing list http://www.atmlist.net/