INTRODUCTION 
The Photogrammetric Unit of the Institute of Advanced Architectural 
Studies is engaged in a programme of recording Ancient Monuments in 
England in the care of the Department of the Environment. The principal 
product of the Unit is 1:50 scale line drawings showing all architectural 
detail and stonework of the facades of those buildings. An increasing 
requirement has been to provide horizontal and vertical cross-sections 
through parts of the facades. In some cases, what has amounted to a 
complete and highly accurate plan of the building has been prepared. 
This has been possible due to the system of theodolite control which is 
used to establish a coordinate framework for the whole building (Fig 1). 
This cross-section work has so far been carried out on one of the Unit's 
Thompson-Watts plotters which is fitted with rather a complicated 
mechanical interchange. In order to provide a more flexible system for 
the plotting of horizontal and vertical cross-sections,and to eventually 
lead to the possibility of complete digital manipulation of data, it was 
decided that a 'digital' approach to the modification of the second 
Thompson-Watts plotter would be made. Obtaining and storing the 
coordinates of points in numerical form is of course very commonplace, and 
relatively straightforward. However, the coordinates in themselves would 
be of limited value, since the final product required is a graphical 
output. Purchasing an automatic drafting table would clearly be a very 
expensive solution, so attention was given to the possibility of electri- 
cally driving the existing coordinatograph table of the Thompson-Watts 
plotter,.and it was this solution which was finally implemented. For the 
graphical plotting of the facade drawings, the table is connected and used 
normally. For the drawing of the digitally derived cross-sections, the 
table is disconnected from the plotter, and driven electrically through 
stepper motors. 
Concerning the approach to digitising the plotter, the various methods 
have been thoroughly reviewed by Petrie (1980). The method adopted is 
Petrie's 'software' approach, ie linking the plotter to a microcomputer, 
into which various software options can be fed, both to control the basic 
data acquisition, and also to provide other facilities such as orientation 
procedures. The microcomputer chosen was a North Star 'Horizon' with 64K 
store and twin floppy 5i" disks. 
120