THE purpose of the mechanism described is to produce an instant and automatically variable torque that is always equal to the resistance encountered while the primary driving engine maintains its speed and torque nearly constant.
The principle of operation is illustrated by reference to the action of a stick weighted with a ball at one end and suspended freely by a string at the other so that it can swing as a pendulum. Interference with the swing of the pendulum at any point on the stick results in a transference of part of the inertia force of the ball to other parts of the stick and also alters the amplitude and rate of vibration of all parts. At the same time, pressure is set up at the point of interference. The pressure varies in magnitude with the inertia and is proportional to the change of speed in a unit of time. In a motor-car the engine, which develops a steady pull or torque at a constant speed, is used to drive against a resistance that requires either a variable torque or variable speed. The mechanical means employed to provide this variable torque can be reduced to the principle of the lever, in which varying the point of support changes the pressure required at one end to balance a constant weight at the other end. In practice in a motor-vehicle, this is accomplished by a change of leverage, or ratio, by shifting gears.
Time is another element that enters into the problem. It is possible to transfer more or less force to a given point by altering the frequency of oscillation. Moreover, whereas with an ordinary lever a steady pressure at one end will maintain a steady pressure at the other end, this condition cannot be reproduced with the pendulum unless the system is permitted to accelerate indefinitely. The same result can be secured, however, by alternating the pressure on opposite sides, and the pendulum will maintain a mean position. Similarly, rotary motion cannot be transmitted from one shaft to another except by periodic alternating impulses, which is exemplified by the periodic impulses transmitted by the teeth of gears and also by the alternating electric current.
These facts are made use of in the Constantinesco mechanism, in which an inertia system is interposed between the prime-mover, which imparts an alternating motion, and a system of “mechanic valves,” or roller-type irreversible clutches, which convert the alternating motion into rotary motion. Close similarity exists between this mechanism and electric transmission, as it splits the motion derived from an alternating crank into two components, one applied to a mechanical inertia element and the other into a resistance. These forces are analyzed with formulas.
Application of the torque-converter to a 5-hp. car and its automatic action under driving conditions are described. It functioned perfectly when the author drove the car, which was so simple to control that he believes engineers should give much attention to this transmission.
In discussion of the paper comment is made that power imparted to the engine flywheel prior to starting the car is not immediately available to overcome the resistance, and that this would be disadvantageous when a quick start was required. Efficiency of the transmission is said by the author to have been proved by tests to be approximately 80 per cent for 1 hp. transmitted and that at full engine-power the efficiency should be very high. In the small car virtually no vibrations of any kind were noticeable, and the author asserted that the inertia forces of the torque-converter do not impose undue pressures on the engine bearings. A description is given by one speaker of a somewhat similar torque-converter patented in this Country which is constantly in balance and which, judged from the mathematical angle, is the most perfect power-transmission mechanism conceivable.