A method of scaling and checking computer circuits
L. Jubin Lane · Transactions of the American Institute of Electrical Engineers Part II Applications and Industry · 1958
IN RECENT YEARS the analog computer has become available to a large number of design and development engineers. It is not difficult for an engineer to become proficient in the use of the analog computer, but one problem which confronts all who use it is that of scaling: How many volts represent one inch? Three popular scaling methods are: (1) the “actual equation-machine equation method,”1-2 (2) the “per-unit method,” and (3) the “direct analogy method.” The first method is better suited to the mathematician than to the engineer because it deals with equations of the system being studied. The system equations are rewritten in terms of computer voltages and are then called “machine equations.” This tends to cause the engineer to think in terms of the computer rather than to retain a physical picture of the actual equipment he is developing. The other two methods are better in this respect since the thinking tends to be more in terms of actual physical variables. However, when using either of these two methods, many engineers feel that the tie between actual and computer variables is not firm or concrete and that the probability of making a mistake is high enough to introduce uncertainty, especially when integrators or differentiators are involved.