Re: Mirror test set ups- - OpFr
- --- In firstname.lastname@example.org, "Dale Eason" <atmpob@...> wrote:
>I agree with Dale on this.Seems I was pushing the limits with my test
> Group parden us while we work through some of these issue. At
> present I think Ric's Igrams are not good enough to do the analysis
> that he is trying for. Ric does not have the experience to recognize
> when the analysis goes bad. It is not Ric's fault. This particular
> area requires more understanding of the process and what can go
> wrong. I am trying to add that information to the help file of the
> program as well as some automatic garbage detection.
setup and it's not good enough to test a f/3 at present.
Perhaps Dale could include the FFT image of the f/3 mirror in the Help
section to show what a unacceptable unwrap looks like...nothing like a
picture to set things straight.
- Hi Tom
I'm going the same way with an XY foucault tester so that I can add a
Z axis for a bath interferometer. Also because I was impressed by the
ease of use of the simple caustic test and the full one. I initially
used normal foucault + mask for all of testing. Often finding that one
pupil was much larger than the other after prolonged spells. I used
the caustic test to verify what I had done. It's much much easier.
I'm not too keen on large source slit less testers. The right sized
slit is very useful for showing that a mirror is truly spherical right
up to the edge. No need for a ronchi screen. Also the caustic test
depends on being able to locate a wire on the image of the slit /
centring diffraction patterns.
My main concern with the bath approach is vibrations. I can't even
mount something on a solid wall where I will be mostly working. 3rd
floor bedroom, low solid walls, floorboards and lath and plaster. All
put together in 1911.
The appeal of the ross is really just a quick look and see.
--- In email@example.com, "tomofreno2000" <tomofreno2000@...>
> Hi John,
> I used the Ross Null Test a few years ago during figuring of a 22"
> f/4.2 mirror. The lens radius was 2 3/4" and I used about 65% of the
> radius to reduce contributions to wavefront error from the (Anchor
> Optics) lens. It did give a null for the finished paraboloid (and
> also straight interference fringes with a grating), but I didn't find
> it as useful or sensitive as the Foucault or Ronchi tests. That may
> well be because I had more experience with those. I find it easier to
> see where I am in figuring with the Foucault test by looking at the
> relative radius of curvature of different areas on the mirror. Plus it
> is more intuitive for me - the surface just stands out in relief.
> I am currently in the final polishing (9 hrs), beginning
> figuring/testing, phase of a 10" mirror, and starting to put a Bath
> interferometer together to use Open Fringe for the analysis since from
> what I've seen from others it will give more information/insight into
> subtleties of the surface for final figuring, as well as a more
> quantitative measurement of surface error than judging shadow edge
> positions with the Foucault test. I'll still use the known Foucault
> along with it though as a sanity check. In fact I plan to use the
> Foucault tester for x and z adjustment, and just add a y adjustment
> plate with optics/laser for the Bath that I can set on the Foucault
> tester to easily switch back and forth between the two with only minor
> re-adjustment. I use a simple slitless tester as described at:
> I won't bother with the Ross Null again. I find it much more
> informative to see qualitatively what the surface looks like with the
> Foucault, and just as quick.
> --- In firstname.lastname@example.org, "a.johnw" <a.johnw@> wrote:
> > I haven't made a mirror for some time. Last time I used a std faucoult
> > set up and then after it was more or less finished tried the simple
> > caustic test, the caustic test and the Dall null test.
> > Came to the conclusion that the Dall null test is by far the easiest
> > to use but am wondering about using it for a fairly fast mirror. I
> > know that higher order aberrations aren't "completely" accounted below
> > F4 but am concerned about having to use a semi reflective mirror or
> > beam splitter in the test set up. (Also for the other tests)
> > I've read that the accuracy of the semi reflective mirror isn't too
> > critical. I've seen comments like find a microscope cover glass that
> > flat to a part of a wave which makes me wonder. Those are very thin
> > not big enough either. Semi silvered mirrors are generally a lot
> > thicker. Beam splitters are tooooo expensive so I'm not to keen on
> > that option.
> > John