An Improved Method for the Design of Electronic Function Generation Networks

A method is described to overcome the errors in diode function generators when operating at voltages considerably lower than the maximum, without a significant increase in the number of sections required. The technique, based on varying section length, produces a maximum error in each section which is nearly a constant percentage of the local value. An application of this to the production of a square root network is analysed in detail and the results are presented in graphical form to allow the designer to select a combination of accuracy and the number of sections for any particular problem. Several units have now been constructed using commercial components and the experimental results are in excellent agreement with the predicted values of error. The method has been found to be a simple and inexpensive way of producing fixed function generation with a considerable improvement in accuracy.