In the early 1950s two U.K. aerospace subsidiaries – Avro Aircraft, which designed planes, and Orenda Engines – conducted some groundbreaking IT work in Malton, Ont. just outside Toronto.
Avro was behind many early aerospace developments, including an experimental “flying saucer.” But in Canada it is perhaps best known for the ill-fated supersonic plane known as the Arrow.
The Arrow was designed to fight Soviet planes. That meant knowledge of their speed, manoeuvrability and armament was required during the building phase. Knowing which missiles to use was was nother issue, the main contenders being the Hughes Falcon and the Douglas Sparrow. The characteristics of the Arrow, its missiles, target and “uncertainty” factors, together determine what is known in fighter plane circles as the “probability of kill,” or PK.
To evaluate which missile to recommend, Avro bought a computer, the CRC102A, a predecessor to the IBM 650, from a Minneapolis firm. It was used for a variety of aerodynamic analyses, and in the process demonstrated to Arrow team members the usefulness of computers, still a very new concept.
Cooperating with the Institute of Aerophysics, use of the CRC102A led to additional knowledge about supersonic flight. Computer programs in machine language were fed to the computer, along with data, by paper tape telex at a speed of a few characters per second. The memory of the computer was tiny and slow, storage being made up of a cylindrical drum. This required knowledge of the drum’s rotational speed and instruction time in order to use the machine effectively.
The Arrow project also saw Avro’s first use of a non-aerodynamic computer program, and one of the first of its kind in the world: a simulator for aircraft/missile engagements. Its job was to determine the PK for a variety of Arrow-borne missiles with different targets. Early computers had no graphics capability, and trajectory results had to be printed and plotted by hand. It saved many days of manual work (since earlier simulations were done with, believe it or not, toy airplanes) although each simulation still took several hours to run.
Peter Bandler, who helped install the FERUT computer at the University of Toronto, (subject of a later article), joined the Aerodynamics Department, becoming an expert in “flutter analysis.” He used the computer to considerably expand the knowledge of such phenomena, (flutter is where an aircraft wing, subjected to certain loads and wind conditions, will start to resonate, causing wing failure). In later years, when the new distinctive Toronto City Hall was designed in the shape of two airfoils, he and a colleague did a study to show that the City Hall, under certain wind velocities and direction, might resonate and collapse.
Whether any notice was taken is not known, but the City Hall still stands.
Later, Avro negotiated with IBM for its first larger-scale scientific computer (IBM 701, later called IBM 704). Following its introduction Avro tested out a new language called Fortran, developed initially by Northrop Aircraft. This enabled the writing of programs in a scientifically oriented notation rather than machine language, speeding up application development time considerably.
Also, at the time the North American defence system used a system called Sage, which linked early warning systems to a computer. If it and the relatively unreliable early warning systems happened to be working (by no means a certainty) it could handle about 100 objects in the air at any one time, hardly enough to stop hostile aircraft getting through.
Arrow was designed to intercept these incoming aircraft, so its electronics had to be capable of integrating with Sage, adding further skills to the rapidly expanding information technology of the time.
However, much of the Arrow’s technological prowess would go unproven. One of the most beautiful and versatile aircraft ever built was destroyed, along with the world’s most powerful jet engine, the Iroquois, by short-sighted politicians.