The square tube ABCD, Plate XXIX. Fig. 1, 2, 3, 4, 5, 6, is composed of four pieces jointed into each other. The first part, A, serves to carry a pulley, E, over which a thread passes, attached at one extremity to the end of the rod F, and at the other, to a scale, G, in which are placed different weights successively, in or der to mark the degrees of velocity upon the dial-plate X. The part A may therefore be removed when the graduation is completed. The second piece B, which receives part of the rod F, carries upon its two sides two upright plates, H, I, Fig. 4, 5, which support a toothed wheel, K, that works in the rack L, fixed upon the rod F, made of very dry fir. On the upright plate H, the dial-plate X, eight or nine inches in diameter, is fasten ed by two screws, M, N; and through the centre of this dial-plate passes one of the extremities of the axis of the wheel, which carries the index that points out the different velocities of the wind. The third piece, C, which contains the spring, has a square hole at 0, con taining two friction rollers of copper, a, b, for relieving the motion of the rod F. The fourth piece D, has like wise two friction rollers, c, d. The extremity F, of the rod, is fixed to the plane surface Q11, which is exposed to the impulse of the wind. This plane consists of a square piece of thin sheet copper, or white iron, one of whose sides is six inches, and is covered with fine wax ed silk, which is sewed to the plane by means of small holes, that permit the needle to pass through it. The hollow handle VX, Fig. 3, has a small cone of copper, V, and a circle of copper, soldered at X, and pierced with a hole, so as to admit an iron bar, on which the cone of copper and the whole instrument is made to turn, by means of a vane fixed at the extremity A. In the formation of this instrument, particular attention must be paid to the construction of the toothed wheel K. Before this wheel is made, a spiral spring must be chosen, which can be compressed with a force of 18 grains. Suppose the greatest velocity of the wind which it ig required to measure, be 88 feet, per second, and that the plane surface is six inches square ; then, as it appears from the table at the end of this article, that when the wind has a velocity of 88 feet per second, it exerts a force of 17 pound 111 ounces avoirdupois, upon a square foot of surface, it is evident, that it will i exert a force of 171b. 1 l oz. or 41b. 7 oz. nearly upon a surface six inches square, because it has only an area of one-fourth of a square foot. When the spring is at rest, mark the point on the rod F, where it enters the tube, then, having put 41b. 7oz. into the scale 0, mark the other point on the rod F, where it now enters the tube : The distance between these two points, which we may suppose to be six inches, is the length of the scale for measuring all the velocities of the wind below 80 feet But as 22 : 7, the ratio between the circumference and the diameter of a circle, so is six inches, the length of the scale, to one inch and nine-tenths nearly, which is the diameter of the wheel K. In order to find the inter mediate divisions of the scale, we have only to place successively at G, the fourth part of the weights in Col. 2 or 3 of the table already mentioned, and mark at the places on the dial-plate to which the index successively points, the velocities in Col. 4, 5, 6 or 7, which corres pond to the weights whose fourth part was taken. Thus, if we wish to mark the velocity 4.40 feet per second on the scale, we take 0.011, the fourth part of 0.044 and place it at G, when the index will turn to the point in the dial-plate which answers to a velocity of 4.40 feet, or 4 feet 4 inches and 8 tenths per second. M. Demenge has pointed out a method of finding the intermediate divisions, after several of the leading ones have been ascertained ; but it is unnecessary to explain it here, as it is certainly more safe and accurate to find each point of the scale by actual experiment. In order to simplify this instrument, M. Demenge proposes to substitute the pulley Z, Fig. 6 in place of the toothed wheel and rack. This pulley has exactly the same diameter as the toothed wheel K, and is moved by the which, after coiling round the pulley, crosses below it, and has its extremities fixed at Y and U. It is obvious, that as the rod F cannot move without making the circumference of the pulley move through the same space, the same angular motion will be given to the index as when it is placed upon the toothed wheel—Figure 1 is a perspec tive view of the machine, the dial.platc and index being only dotted in order to spew the toothed wheel. Figure 2 is a section of it, by a horizontal plane ; and Figure 3 is a vertical section of the instrument. The whole length of the instrument is about two feet three inches.
M. Demenge has likewise contrived another anemo meter, which consists of six sails moving horizontally, and fixed on a vertical axis. These sails are enclosed in a circular cage, with twelve fixed screens or shutters placed at an angle of nearly 30° with the radius of the circular cage, so as to give the wind free access to the advancing sails, without offering any resistance to those which are returning. The vertical axis carries a crown wheel, which drives another wheel, fixed on a horizontal axis, at the extremity of which is fixed an index, which points out the velocity of the wind upon a dial plate. This velocity is measured by means of a spring, having one of its ends fixed on the horizontal axis, and the other in a nut, and of sufficient strength to prevent the axis from making more than one revolution, when the sails arc acted upon by the strongest wind. By means of this
spring and a series of weights, proportioned to the force of the spring and the size of the instrument, the degrees upon the dial-plate must be ascertained. This is done by suspending each weight successively to a cord coiled round the horizontal axis, and marking the point on the dial-plate to which the index successively points. The places thus ascertained will indicate the force of the wind, corresponding with the magnitude of the respective weights.
An ingenious anemometer, different from any of the preceding, was proposed by the Marquis Poleni, in the Dissertation which gained the prize of 1733. A plane surface was suspended from a centre, and exposed to the action of the wind, which impelled it from the vertical position, according as the velocity of the wind was greater or less. The force of impulsion was therefore determined by the angle which the plane surface formed with the vertical line. Sec Polcni, De la meilleure ma niere de mesurer men le chemin d'un vaisseau.
About 11 years afterwards, Mr Pickering described, in the Philosophical Transactions, an anemometer ex actly the same as that of the Marquis Poleni, of which, it is probable, he was wholly ignorant. It is represented in Plate XXIX, Figure 7, where A is the vane for turn ing the apparatus to the wind, B a graduated quadrant, with an iron rim, notched or cut into teeth, so that each tooth answers to a degree on the limb. Through the centre of the quadrant passes an iron pin, which serves as the centre of motion, for the moveable radius D, which carries the sail C. This sail is a foot square, and is formed of painted canvas, stretched across a wooden frame. It is shewn separately, with its two arms or radii, in Figure 8. On the upper side of the frame at b, is a small spring, which catches at the notch or degree to which the wind may raise the sail from its vertical posi tion, and prevents the sail from resuming that position when the wind has subsided. Near the bottom of the stand is fixed a circle of wood, on which are described the 32 points of the compass, which are shewn by means of the index WS, which moves along with the vane and quadrant. When this machine is so placed that the cast and west points, upon the wooden circle, coincide with the east and west points of the horizon, and are exposed to the action of the wind, the vane A will turn round the quadrant till the sail is perpendicular to the direction of the wind, and the sail will be elevated from its vertical position into an inclined position, where it will be retain ed by the spring falling into one of the notches. In this situation the index will point out the quarter from which the wind blows, and the notch or degree, at which the sail rests, will be the angle to which it has been elevated by the wind.
It is obvious, however, that the angle of elevation is not a measure either of the force or velocity of the wind. This does not seem to have been considered by Mr Pick ering, who observes, that by this instrument " the rela tive force of the wind, and its comparative power, at any two times of examination, may be accurately taken." Let us, therefore, investigate this point, and find by what means the instrument may be made to ascertain, not only the relative, but the absolute velocity of the wind. Let AB, Plate XXIX, Fig. 9, be a section of the sail in its natural or vertical position, and let it assume successively the positions AC, AD. Then if AB or AC be taken to represent the resistance opposed to the wind by the sail, when in its vertical position A7nB, it is obvi ous, that when the sail has been raised into the position AC, the surface has been contracted, and intercepts a clAumn of air, whose diameter is only Am ; and when it has attained the position AD, its surface has been still farther contracted, and it intercepts a column of air whose diameter is only An ; But An, Am arc the co sines of the angles DAB, CAB ; therefore the resistance arising from the oblique position of the sail may be re presented by (4., where is the angle of eleva tion, and F, the resistance of the sail, in its position AB. Now since, when the sail is at rest in the position AC, there must be an equilibrium between the force of the wind and the resistance which is opposed to it, this re sistance will be a measure of the force of the wind. Let the resistance of the sail be now represented by Cp, we may decompose it into the two forces Co, Cq, of which Co is the only part that opposes the wind, Cq tending merely to produce tt pressure on the centre of motion. But Co being balanced by the force of the wind, is a mea sure of that force ; and Cp being equal to F xCos. p we have Co : F xCos.9=Sin. C/i o : Sin. Cop=Sin.CAB : Sin. ACm. But CAB is the angle of elevation (?), and Sin. ACm=Cos.9. Hence Co : F xCos.9=Sin.9. : Cos.9 and F xCos.9xSin.9 Co= consequently dividing by Cos. 9 Cos. co we have 9, that is, the force of the wind is proportional to the sine of the angle of elevation. Had the body, or sail, to be acted upon by the wind been a spherical ball, as in the quadrant for measuring the ve locity of water, from which the idea of Poleni's instru ment seems to have been taken, it would have always presented the same surface to the wind, and therefore its resistance F, would have suffered no change by be ing elevated from the vertical position. We should then Sin.9 have had Co : F=Sin. 9 : Cos. 9 and Co=F X 9. But Sin. co AB Hence the force of the wind Co varies as the tangent of the angle of elevation.