7 
This concept cf the scanning system was developed by Hobrough and 
was optimally adapted in its optical parts to conform with the 
electronic conditions. To avoid a loss in resolving power caused 
by the finite size of the scanning spot, a 2:1 reduction is imposed 
between the cathode ray tube and the dispositive. The imaging 
objective has a very large relative aperture in order to utilize 
as much as possible of the limited luminescence of the phosphor. 
Both these factors call for a relatively large field angle of the 
objective. Since space for the condenser lenses within the cross 
slide system is limited, conditions are encountered which are op 
tically relatively difficult to meet. 
To reduce inertia and reflections causing light losses to a minimum, 
no special picture carrier or correction plates are used in the 
new instrument. Happing in the B8 Stereomat must therefore be done 
from normal, distortion-free diapositives printed emulsion-to- 
emulsion on 9 l/2 x 9 l/2 inch glass plates only l/8 inch thick. 
The plates are positioned in the instrument emulsion up, eliminating 
any unnecessary reflecting surfaces in the path of rays. At the same 
time, a non-reversed model is produced. If pictures with considerable 
lens distortion are to be restituted, this distortion must be 
corrected during printing in a projection printer. Plate unflatness 
is without effect since the optical axis is always normal to the 
plane of the diapositive; thin plates may therefore be used. Kelsh 
diapositives, normally printer mirror-reversed on l/4 inch plates, 
cannot be used without special arrangements in the B8 Stereomat and 
are not recommended for plotting. 
The principle of the automatic function of the instrument is ex 
plained in Figure 4. If, as shown, the space rod intersection is 
not at the correct height in the stereo model, then different image 
details lie on the optical axes of the left and right scanning 
systems. Since the flying spots move exactly similarly and synchronously 
with respect to the left and right optical axes, they cross 
corresponding light-dark borders in the two pictures at different 
times. The resulting alternate currents in the two photocells 
therefore have a phase difference, the magnitude of which is de 
pendent upon the height error of the space rod intersection. If the 
intersection has the correct height position, the phase difference 
is zero.