Draft #2 - Lens Matching

[T3D John Ohrt <johrt@xxxxxxx>]

Please comment!!!!! You won't hurt my feelings!

Draft #2:  Matching lens magnification and achieving a precise focus

Introduction

Originally a "matching lens" discussion expanded into a more global
inquiry to include "precise focusing".


What Is A System Specification

A system specification is concern with the cumulative effect throughout
the system on a particular area of concern.  For  example, if your
system comprises a camera and a viewer,  the specification for the
maximum difference of magnification includes the effects of both the
camera and viewer.

Do note that "cheating" is allowed.  In this example, an enlarger can be
adjusted during the processing phase to correct for known camera
magnification differences in some situations, such as holmes cards.

The main point to remember is that a system specification is a budget.
If you blow the entire budget on the camera, your work may not be well
received by those with less than perfect viewers!


System Specifications

The specs quoted are largely from MIL-HDBK 141,  Optical System Design.
When possible, specifications were cross checked against other sources.
They were in agreement.

These specifications are for systems that may have to be used for long
periods of time by a wide variety of people.


Unequal Magnitification:   0.5% max,  some can tolerate 2%

Pupillary adjustment:      50 to 76 mm

Rotation or skew:          minimize  (tiring)

Unequal Illumination:      10% max,  avoid rapid level changes

Vertical imbalance:        0.5 prism diopter

Horizontal imbalance:      0.25 prism diopter  ( 0.33 tolerable at low
                                                 magnifications)

Lens Matching

As you may have noticed, the system tolerance for magnification variance
is 0.5%.  Depending on your system, you have two different sets of
matched lens may be involved and the tolerances of both sets combined
must be less than 0.5%.  In such cases, your work may be viewed by many
different viewers.  You should allow at least half the budget (ie. at
least 0.25%) for the viewers.  One recent specification of a desirable
stereo camera suggested 0.2% max as appropriate for the camera, leaving
0.3% for the viewer.

Note that the difference in magnification is the issue.  That is, you
don't have to measure the magnification for each lens in the set.  All
you need to know is the variance, lens to lens.

The nominal focal plane is the nominal location of the film emulsion.
There is a later discussion on why the emulsion layer may not lie at the
nominal location.

One approach to measuring the variance in magnification is to use each
lens to observe the same target for the same location.  Each lens must
be precisely focus at the nominal focal plane.  A stereo camera must be
moved so that each lens is precisely at the location of the first lens
when it was measured.  The distance between the reference points on the
target is then measured at the nominal focal plane.  The percent
difference in length measured at the focal plane is the same as the
percent difference in magnification.  Exactly how you make the
measurement depends on the format size and your measurement method.  The
accuracy must be finer than 0.1% to achieve valid results.


Precision Focusing

There are many applications that do not require precision focusing, but
there are some that do, and others that could greatly benefit from the
technique.  The preceeding test is one that benefits from precision
focusing.

Many successful photographs are taken assuming that the viewfinder
focus screen, rangefinder, or lens barrel markings are sufficiently
accurate.  However there are photo subjects where more precise focussing
is required. The problem is that all the preceding approaches assume
that the focus at the viewer is identical to the focus at the nominal
focal plane.  It is most unusual for that to be within acceptable
tolerance for all subjects, and even if it were, you still can't
perceive the necessary level of detail to accurately focus the camera.

There are two "levels" of precision beyond normal practice..  The basic
distinction is whether you have taken into account the difference
between the location of the photoactive layer and the nominal focal
plane.  There are also two basic techniques to precisely focus the
camara, and either can be used for each level.


Why The Film May Not Be At The Nominal Focal Plane

Cameras that use rolls of film are susceptible than those that use sheet
film.  The following diagram illustrates the problem:

|__________________________|     <- film pressure plate
|| ---------------------- ||     <- film
||  ||                ||  ||
||  ||                ||  ||
||  ||                ||  ||

                       ^   ^
                       |   |
                       |   -------- Pressure plate rail (1 of 2)
                       ------------ Film rail (1 of 2)

As shown, the film is restrained by the film pressure plate and the film
rails.  The pressure plate is spring loaded and presses firmly against
the pressure plate rails.  The resulting gap between the pressure plate
and the film rails is wide enough to handle the thickest and least
flexible films.  So the film can position itself anywhere between the
pressure plate and the film rails and may even not lie in a plane.  On
top of this, is that the ideal image plane lies within the film, but
where varies from film type to film type.


The Ground Glass Surface Approach To Precise Focusing

The first focussing approach is to place the ground glass side of a
piece of glass ground on one side at the nominal focal plane and tape it
in place.  In the case of a film camera, that means place the ground
side against the film rails.  Then place a 15x loupe against the glass,
focus the loupe onto the ground surface, and then focus the camera while
looking through the loupe.

If you wish to compensate for precise film location, the ground glass is
shimmed until the magnification is identical to that of a photo in sharp
focus.  All you do is take a number of photos at various focal settings
and select the one with the sharpest focus.  Remember, you have to do
this for each film type you use, so you need a separate glass for each
film.

If you are the do it yourself type, a flat, ground surface is the
easiest optical surface to make and plate glass works fine and is easy
to work.  You also get two flat, ground surfaces, one on each glass
plate.  You stop when the glass has been ground flat with 400 grit
compound.  Any vendor of amateur telescope making supplies has what you
need.

The second focussing approach is the knife edge approach, which is even
more accurate than the ground glass approach.  There is an excellent
description of this technique and the Ronchii technique of achieving
precise focus at the following URL:

http://www.aa6g.org/Astronomy/articles.html  (select Ronchi)


While the source and availablity of a Ronchii screen may concern you,
they can be easily made with a laser printer and transparency media.
Instructions can be found at the following URL:

http://www.multiboard.com/~joneil/ronchi.html

If you wish to allow for the precise film location, you just shim the
ronchii screen as you would shim the ground glass.

John Ohrt,  Regina, SK, Canada


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