SOME INVESTIGATIONS ON FILM FLATNESS IN AIR SURVEY CAMERAS 
by 
J 0 Mo Clark and G 0 R 0 Cooper 
INTRODUCTION 
It is of considerable importance in survey cameras that during photography 
the emulsion surface of the film should be flat and should lie in a plane which 
is as nearly as possible the focal plane of the lens. Two types of device are 
in general use to achieve this aim; the vacuum back, in which the back of the 
film is held by suction against a flat plate, and the register glass, a glass 
sheet between film and lens, against which the emulsion surface of the film is 
held by a pressure pad. A valuable addition to the register glass camera is a 
calibrated reseau on the surface of the register glass which lies in the focal 
plane, in contact with the film. Such a reseau not only offers the possibility 
of compensating for film distortions which arise during processing and drying, 
but also permits variations in the general location of the film to be accounted 
for, since the image of the reseau provides a system of co-ordinates whose 
relation to the camera constants is known. Reseau glasses may, of course, also 
be fitted in vacuum-back cameras, and some interest in doing this is now evident 
among camera manufacturers. 
Except at the reseau marks themselves, however, errors may arise from lack 
of film flatness. Fig.1 shows how a ripple in the film surface occurring 
within a distance of the order of the reseau dimensions will give rise to an 
error in the location of the image point. The effect depends on the inclination 
of the image-forming rays to the optical axis of the lens, and might be expected 
to be particularly important in super-wide angle photography. Since the dis 
tortions of modern survey lenses are in the region of only a few microns, lack of 
flatness of the film of a few microns at field angles of say, 20° and upwards 
may make a significant contribution to the total error budget. For this reason 
it was decided to make an investigation of film flatness in some typical survey 
camera assemblies. 
EXFERIj'iENTAb INVESTIG-ATICNS 
Ideally, one would like to have a contour map of the emulsion surface 
relative to the focal plane, with the contours at intervals of somewhat less 
than a micron. This requirement immediately suggests the possibility of using 
interferometric methods, and experiments were initiated to find out whether this 
approach was practicable. The first aim was to obtain, using a register glass 
camera, effects analogous to the well-known phenomenon of Newton's rings by 
reflection. After some attempts, success was achieved by using a film with 
high-gloss emulsion surface, originally made in small quantity for another 
purpose. 
The arrangement of the apparatus used during the main course of the investi 
gation is shown in Fig.2. The lens of the camera was removed and a collimated 
light beam was shone through the register glass onto the emulsion surface. It