Magès Note no. 560.108: reasoning on a continuously variable transmission (1956)

In this document

The starting principle: two engine curves

The note opens with a handwritten graph contrasting the general engine power curve, marked A, with a curve B, the "geometric locus" of the points of best efficiency at each speed p. 159.

Magès illustrates the problem of a conventional gearbox: at 100 km/h in 3rd gear at 4000 rpm, if only 20 HP are required, this power is "necessarily taken from a zone of high specific consumption" — point C on the graph p. 159. With a continuous and automatic variator, the same need could be met around 1800 rpm, at point D, "to obtain minimum consumption (and an appreciable gain in comfort)" p. 159.

The general rule and its implementation methods

He draws a "GENERAL RULE" from this: only use the engine at the speed providing the required power on the minimum specific consumption curve B p. 159.

To apply this, Magès envisages that the throttle opening be regulated by a cam controlled by the rotation speed, then notes that the engine torque follows a line roughly between 1000 and 4000 rpm on the 2 CV engine, which could "correct the throttle opening" p. 159.

Extensions of the reasoning

On page 160, he continues: without a conventional gearbox, curves A and B could become "very similar if not identical" p. 160. He imagines an engine operating "always at full throttle", except in the transition zone near idle p. 160, with lean mixtures and automatic heating of the intake air increasing as the speed decreases, a "factor of efficiency" p. 160.

Conclusion: the throttle would disappear as a regulating organ, replaced by an organ automatically modifying the gear ratio according to the effort requested by the driver, so that the engine "automatically takes an equilibrium speed depending on the power requested of it" p. 160.

Context

In 1956, Citroën was developing the DS, launched the previous year, and Paul Magès had already imposed the hydropneumatic suspension on it. The search for an automatic continuous transmission is part of a broader reflection of the time on comfort and efficiency, extending the spirit of hydraulic innovation that characterizes Citroën's design offices. The hydrostatic transmission, a subject widely documented in the following decades of the Magès collection, finds here one of its first theoretical formulations.

Archivist's perspective

This document seems to constitute one of the very first written formalizations, by Magès himself, of the founding reasoning of the continuously variable transmission — nearly twenty years before the successful developments on the hydrostatic transmission found in later pieces in the collection, such as the note on the adaptation of the engine to an automatic variator (1973) or the technical presentation to the B.E.A. from the years 1973-1974.

The thirteen-point approach, supported by hand-drawn curves, probably illustrates a working method characteristic of the engineer: starting from a simple physical observation (the power curve) to deduce, step by step, the mechanical consequences and the components to be designed p. 159-160.