Q: How, in the days before digital exchanges, were telephone calls traced? And why did it always take so long to do (at least in films, where the villain always hung up just before the trace was made)?
A: Twenty or more years ago, as a young trainee, I was coached in the covert skills of tracing. The principle was simple but relied on fleetness of foot. The technology of the day was Strowger switching, which consisted of electromechanical devices that connected calls through a two-dimensional 鈥� horizontal and vertical 鈥� matrix of electrical contacts. Each contact was connected to another switch 鈥� and so on until it reached the called telephone. My task was to inspect the first switch, see which contact the wiper blades were resting on, run to the next switch, inspect that one and so on. This process could involve running up and down stairs in the chase to trace the call before the handset was replaced.
A: The familiar TV depiction of 鈥渢racing鈥� calls should more accurately be described as a 鈥渂ack trace鈥�. To explain the process it is necessary to understand the way in which a call was set up in the days before electronic exchanges.
Advertisement
Upon lifting the handset, a signal was sent to a piece of equipment called a uniselector. This rotated over a semicircular set of contacts to find a free switch called a first selector, which was a two-motion device. As its name implies, this selector could move both vertically and horizontally. The dial tone was returned from this selector, and the selector was stepped vertically under control of the dial.
Horizontal, or rotary, motion then took place automatically to find a second selector, where again vertical movement was under control of the dial and rotary action took place automatically to find a final selector. The final selector then accepted the third digit to give vertical motion, and the fourth digit to give rotary motion. Thus, if the number dialled was 4567, the first selector would accept the 4, the second selector would accept the 5 and the final selector would accept the 6 and 7 to call the required number.
A forward trace was a relatively simple task. All that was required was to follow the call from switch to switch using the grading charts, which gave information as to where connections were made. Bearing in mind that a uniselector had 50 outlets, any one of which could be selected to find a free first selector, the first selector had 20 outlets to find a free second selector, the second selector also had 20 outlets to find a free final selector and any one of up to 300 final selectors had access to the called number, it could take up to 15 minutes to do a forward trace.
To trace a call from a called number back to the calling number is a bit like only knowing the meaning of a word and then trying to find the word in a dictionary. A great deal of searching over 3-metre-high racks of equipment in narrow rows, sometimes 45 metres long, was required. You had to use ladders, and in larger exchanges calls were traced on different floors.
This procedure applied only if the call was from and to the same exchange. If the caller made a call to another exchange the process had then to be repeated at the final destination exchange.