P3D Re: Holmes Viewer Lenses

[John W Roberts <roberts@xxxxxxxxxxxxxxxxx>]

>Date: Sat, 3 Jul 1999 21:46:00 -0600
>From: Dave Williams <davidrw@xxxxxxxxxx>
>Subject: P3D Holmes Viewer Lenses

>I have a couple of questions about lenses for Holmes type viewers.  What
>is the purpose of the prism configuration?  

It's not really "prisms" - that's a shorthand notation that roughly describes
the function of the lens sections used. The key to understanding Holmes
viewers is that the infinity points in the left and right views of a classic
Holmes print are a little over three inches apart - greater than the
interocular distance (spacing between the eyes) of most humans. This wide
spacing allows larger prints to be used, giving better image quality, but
a result of the wide spacing is that to free-view the stereo image, most
people would have to diverge their eyes (the right eye points a little to
the right and the left eye points a little to the left). It's very hard
for most people to diverge their eyes - if they can do it at all, it often
results in eyestrain. But if you view the prints through magnifying lenses,
with the optical centers of the two lenses spaced further apart than the
infinity points of the two views, then the lenses bring the images together,
so they can be viewed without having to diverge the eyes.

Of course if the diameter of the two lenses is small, and if your eyes are
much closer together than the spacing between the optical centers of the two
lenses, then you won't be able to look through both lenses at the same time.
So fairly large diameter lenses are usually used. But with large diameter
lenses, there's a lot of glass on the "outside" half of each lens (the side
away from the point midway between the two lenses) that you'll never look
through. That's a lot of money to spend on glass that you'll never look
through, and it makes the viewer bulkier, so those parts of the lenses are
usually chopped off. So instead of round lenses, a commercial Holmes viewer
usually has rectangular segments of round lenses. The fact that the lens
segments are not from the centers of the lenses, but from one side, means
that one side of the lens will be thicker than the other, giving the lens
the appearance of a prism, and causing it to act like one. (However, the
"prism" geometry varies from top to bottom.)

>Can prints be viewed in a
>viewer without prism lenses?  

Yes, but you have to diverge your eyes (or use smaller prints than the
standard). Most people will be unable to use such a viewer.
Or you could make small, nonstandard prints, but that would lead to
complications if you want to pass them around to people with regular
Holmes viewers.

>I'd like to try my hand at building a
>print viewer, both for the fun of it, and because the beautiful ones I
>see on the net (such as Alan Lewis viewers) are beyond my price range.

Sounds like an interesting and worthwhile project. I hope you'll keep us
posted.

>Can non prism achromats be used, and if so, what is the focal length of
>the average Holmes type lens? 

I would recommend you try round lenses with sufficient diameter so they can
be offset far enough to the sides to give you the "prism" effect you need.
(It might be worthwhile to try it with cheap lenses before springing for
achromats, to make sure you have the geometry about right.) I believe some
Holmes viewers are made with a single lens chopped apart to make the left
and right eyepieces - that's an easy way to make sure the focal lengths
are matched, but I have no idea how to cleave a lens.

A plastic-lens Holmes viewer I built from a kit has focal length of about
7.5 inches. Some glass lens Holmes viewers I've measured have focal length
around 6.5 - 6.75 inches. 6.5 - 6.75 inches would probably be pretty good, 
but that's not a scientific analysis.

>Hope everyone has a ball at PSA, wish I could be there!

I would probably be at NSA, except that I have some critical meetings for
work that I have to attend.

John R


------------------------------