BALANCE, THE TORSION. The first successful attempt to make an even balance or other weighing machine with beams oscillating on pivots, which should dispense with knife edges, and thereby avoid their well-known defects of liability to damage by wear, rust, and overloading, was made by Frederick A. Roeder and Alfred Springer, in Cincinnati, Ohio. in 1682. They used as a pivot a steel wire stretched tightly be tween abutments. The balance-beam being- firmly attached to the wire, its oscillation caused the wire to twist slightly, hence the name " torsion balance." The simplest form of torsion balance is a very light beam supported at its middle point. which is also its center of gravity, by a stretched wire, the wire being firmly fastened to the beam. A weight placed at one end of the beam will exactly balance a weight at the other end. The sensitiveness of such a balance de pends upon having the torsional resistance of the wire almost infinitely small. This requires a very thin wire, and as thin wires, when stretched horizon tally, are n01. strong, the balance can be used only for very small weights. Such a balance was Ritchie's. mentioned in the Encyclopedia Britannica, and it was a total failure for large weights. If the wire is made large enough to have an appreciable strength, its torsional resistance prevents the balance being sensitive. To get rid of the effect of the torsional resistance in diminishing the sensitiveness of the balance was one of the chief ends of Messrs. Roeder and Springer's efforts, They accomplished it in a number of different ways. but the simplest, and the one which is adopted in practice, is the placing of the center of gravity of the beam above its point of support. In knife-edge balances such a placing of the center of gravity would make the beam top-heavy, or in unstable equilibrium ; the center of gravity would always tend to reach its lowest point, and tip the beam. In the torsion balance, however, this top-heaviness acts in the opposite direction
to the torsional resist :Mee of the wire, and may be made to entirely neutralize it. We thus have the torsional resistance exerted to keep the beam while the high of gravity tends to tip it out of the horizontal. The adjustment of the pi:within of the center of gravity so as to neutralize the torsional resistance is most easily made by having a poise immediately above the (peter of the torsional wire, and making it adjustable vertically by means of a screw and nut. When the torsional resistance is entirely neu tralized. the balance becomes infinitely sensitive, and any smaller degree of sensitiveness that be desired may be obtained by simply lowering the poise. The torsion balance is made in many forms, but in general the wires are shaped like a thin llat band in section ; instead of being round. the two ends of a strip are brazed together so as to make a ring, and this is tightly stretched over a frame or truss of steel or other metal. of the shape shown in Figs. 1 and 2. In an even-balance scale three of these frames are used, and two beams. an upper mind a lower. The end wires are -25 in. wide by •010 in. thick. The practical sensitive ness of this scale, when vibrating at the rate of 10 oscillations per minute, is about 2 grains.
Fig. 2 shows a druggist's prescription balance sensitive to say grain in actual use. It has a capacity of 6 oz. in each pan. The wires are about -04 in. by '004 in. Their torsional resist ance is overcome by the small renrel weight seen in the cut attached to studs on the lower beam. See Trans. A. S. Xining E.. vol. xii, p. 560; Trans. A. S. if. E.. vol. vi, p. 651.