After I mentioned Tomosy's Camera Maintenance and Repair books in a
recent post Tim Hughes e-mailed me privately asking about the content of
these books. I mentioned to him that they cover building simple test
instruments. One of the items in book #1 (of 2) is how to build a
simple, horizontal autocollimator using one side of a pair of 7x50
binoculars, a first surface mirror (possibly from a broken Polaroid
reflex) and a 2-way mirror (possibly from a broken rangefinder camera).
The purpose of this simple collimator is to test the infinity focus of
SLR's by running the collimator beam into the camera and lens and then
back out again via the first surface mirror (which you cut to fit and
place at the film plane).
Re: Gary Reese's use of a vertical autocollimator for testing filters,
you could easily turn this simple device vertically but building a
rotating table for the target that didn't introduce any wobble would be
a real challenge I think.
If Gary is discarding filters because they don't pass this test then
presumably the filters are either not square in the mounting ring or are
not plane-parallel or both. A rotating table would have to be more
mechanically precise than the errors to be detected. Sounds like a
tough job even measuring the most cheaply produced filter.
Any mechanical/optical engineers around who can say how this is done?
Or perhaps I don't really understand the problem and the tolerances
aren't as critical as I might think?
Chuck Norcutt
Woburn, Massachusetts, USA
-------------------------------------------------------------------
Gary Reese, describing a vertical autocollimator, said:
It projects a purple colored multipointed "star" image through a lens
(or even a filter if you prefer). You look into tube of the the
collimator like a telescope (with a side mounted objective) and you see
the star pattern reflected off a shiny surface below the optic (polished
metal in my case). In a perfectly centered lens without aberrations and
with minimal flare, the star image comes right back through the optic
and seems perfectly superimposed with the projected image. It is
centered, sharp and with indistinct ghosts. Upon rotating the lens, the
star remains stationary. But in a lens with faults, all hell breaks
loose in the image. It might get fussy, or uncentered, or have strong
ghost images, or rotate all over the viewfinder, or become overlapping
multiple images of the same intensity - all combinations of the above.
To see what these look like, go to the C.R.I.S. site on the WWW. I
recall they are about US$3,000 and up.
Gary Reese
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|