Re: [ATM] Plop Flop

["Jan van Gastel" <jhm.vangastel@wanadoo.nl>]

A PLOP-optimized 18 pts cell will give a 1/646 lambda RMS error. My 18
points cell (20 incher, 2 inch thickness mirror), a Kriege design, because I
didn't know PLOP welol enough by then, gives, according to PLOP, a 1/554
lambda RMS error. Even if I should have misplaced one or more of the points,
even a whole ring of points (even both rings) as much as 5 millimeters, the
error is still far within the 1/128 lamda, which is adviced as minimum
tolerance. So, if working carefully, tolerances are met very easily,with non
professional tools.

Jan
http://home.wanadoo.nl/jhm.vangastel/Astronomy/


----- Original Message ----- 
From: "Mark Holm" <mdholm@telerama.com>
To: <jaykirk2@compusmart.ab.ca>; <atm@atmlist.net>
Sent: Friday, July 02, 2004 3:25 AM
Subject: Re: [ATM] Plop Flop


> Hi Jay,
>
> I stayed up too late last night watching the fun at JPL as Cassini -
> Huygens did it's orbit insertion (gota love the way that Doppler curve
> tracked the prediction even when the s/n dropped to where the data was
> all fluffed up), so I am not up for running out all the numbers tonight,
> But in the next couple of days, I will do it for you.
>
> The first thing though is to convince you to give up the 27 point
> design.  You don't need it.  According to Plop, a vanilla 18 point
> design will more than suffice.  (see the 18 point designs at
> http://pong.telerama.com/~mdholm/atm/cells/index.html  I plan on giving
> you a slightly less optimized one than those, because the dimensions
> will be easier for fabrication.  The decrement in performance will be
> much too small to worry about.)
>
> I am hoping you will take my advice on this point.  In the last year,
> two stubborn (perhaps just a bit ignorant too?) atm's have stuck to more
> complex designs after I used Plop to show that simpler ones would do
> just fine for them.  I don't know what the lure of complex cell designs
> is.  Seems to be hard to break.
>
> Let me say one more thing on that point.  Kriege and Berry is a great
> book.  The authors wrote the best they knew, but, they were working with
> just plain bad methods when they sized and laid out their cells.  So was
> every other atm at the time.  Plop, with a (very) little help from me,
> is going to give you a much more scientific result.
>
> Plop says the 18-point cell residual deformation will be approximately
> 1/500 wave RMS.   If you think that isn't good enough, well (it is hard
> to say it politely) you are mistaken.
>
> I hope that a couple of the other Plop experienced atm's on the list
> will run your case at 18 points and back me up.  (Jeff Anderson-Lee and
> Tom Krajci are you listening?)  The backup isn't for my peace of mind.
> I am dead certain of the result.  Just want to make sure I convince you.
>
> Jay's mirror dimensions are:
>
> >508mm
> >diameter, 54mm thickness, 2895.6mm fl (yes, f5.7), and a 79mm
obstruction.
> >
>
> It would be slightly helpful, Jay, if you told us the glass type.  The
> numbers will be a little different for different glasses.  Actually,
> this design is so far from being critical that the glass type won't
> matter a lot, but might as well do it right while we are at it.
>
> I can and will translate the results into the same format used in Kriege
> and Berry..
>
> May be Saturday before I can get back to it.  Remind me if you don't
> hear from me in a couple of days.  Sometimes I forget things.
>
> Mark Holm
> mdholm@telerama.com
>
> _______________________________________________
> ATM mailing list http://www.atmlist.net/
>

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