8 
Fig.7 shows diagramatically the simulator and the two servo systems 
for the measurement and transfer of the computed tilts to the 
negative stage. For simplicity, only the y components are shown in 
this diagram. 
In the simulator a tubular space rod is held cardanically in the 
points A and B. This is to be tilted through the required angles 
OC (oeroendicular to plane of drawing) andOC^. For this, the 
must be introduced as in the equations 
given above. 
The value \J ' f can be transferred from the spindle for setting 
the enlargement directly to the Z spindle of the simulator, while 
the dimensions are reduced to one-third simply by selecting the 
appropriate spindle pitch. The point B, about which the space rod 
tilts, is thus located at the distance l/3 * U * f from the null 
position of the steering point A. 
When the table is horizontal the space rod remains (independent of 
the enlargement) in its null position, i.e. parallel to the Z spindle 
of the simulator and the negative stage is therefore also not sub 
jected to tilting forces. Now, in order that when the table is in 
clined through /3 x and /3^ the spacerod be tilted through the re 
quired angles CX and (X respectively, i.e. in order that the 
x y 
equations 
tan oC, 
x 
y 
\J 
\J 
are satisfied, the steering point A must be moved, in a plane 
perpendicular to the null position of the Z spindle, in the x and 
y directions by the amounts l/3 * f 
r 
respectively. These movements are executed by the simulator cross 
slide. 
As can be seen in Fig.7, the value for example of the secondary 
y component, 2.4 * f r * tan /3^, is taken off from the y spindle of 
the table’s cross slide, reduced to the required scale 
1/3 ’ f r * tan ¡3 by choice of the appropriate pitches, and trans 
ferred to the simulator's cross slide. The process is the same 
with the x component which is not shown in the figure. The space