Encyclopædia Britannica, Ninth Edition/Anemometer

ANEMOMETER, or ivind-measurer (from ave/xos, wind, and /xerpoy, a measure), a contrivance for indicating the rate and direction of the wind. Ever since the birth of true experimental science, it has been recognised as a matter of great importance and interest to man that he should know something of the laws according to which the atmospheric currents and changes are produced. Among meteorological phenomena, none deserve more attention than the elements of the wind ; and none have received more. Yet anemometry has been all along the least suc cessful department of meteorology, partly owing, of course, to the nature of the agent to be dealt with, which is the very type of fluctuation, and partly to a mistaken path of observation that was too long followed. The error which, from the days of Hooke, in the middle of the 1 7th century, to within a comparatively recent period, misdirected the efforts of inventors, was the idea that the elements to be determined in the case of the wind are its direction and pressure, whereas practically they are its direction and velocity. If the currents of air were anything like uniform, it would be a comparatively simple matter to deduce the velocity from the pressure ; but their variability is so very great, that the relations between the velocity and the pressure become unworkably complex. We know, from the elementary principles of dynanrc?, that the pressure at any instant will vary as the square of the velocity. Obviously, therefore, the relative variations of the pressure will be twice as great as those of the velocity; and the latter are too great, as we find them, to encourage us to double them artificially. It must also be remembered that, from the inertia of the indicating apparatus, errors will in every case arise ; and these also will be doubled if we take the pressure instead of the velocity variations. From all this it will appear that comparatively little importance is to be attached to the earlier and to all statical modes of anemometry.

The essentials of every anemometer are two : (1), a wind- vane, to show changes of direction; and (2), a luind-gauge, to show changes of velocity, or of force. There is also a distinction which we may have in every kind of meter. They may be either non-recording, as merely exhibiting the variations to the eye; or recording, or marking them per manently on paper.

The wind-vane, or weather-cock, is simply a flat sheet of thin metal, or two sheets in the shape of a thin wedge, at the end of a metal rod, the whole turning freely on a vertical axle. Besides serving to show the direction of the wind itself, the vane is often used to turn the gauge, so as to present the pressure-surface with its face always direct to the wind.

Of all the gauges that have been invented or proposed it

would be impossible to give even the names. The primi tive statical forms of anemometer that is, such as set against the force of the wind a gradually increasing re sistance, with some means of indicating the point of balance have been divided by Dr Robinson (Trans. Eoy. Irish Acad., vol. xxii. p. 150) into three classes : First, vertical windmills, kept facing the wind by a vane, and acting by winding on their axle a string against some form of gradu ated resistance. Of this nature were Dr Hooke s anemo meter (which is generally considered the first practical attempt in wind-measuring), and those of Wolfius and Martin. The second class consists of those in which the wind acts on a flat plate, usually a foot square, pressing it against the elasticity either of a metallic spring or of a mass of confined air, a previous graduation of the elastic force having been made for each instrument. The first gauge of this type was invented by the celebrated Bouguer about the middle of last century ; and since his time there have been many attempts at improvement of this principle, the apparent simplicity of the spring- measure being its recommendation. Modified forms of Bouguer s wind-pressure gauge are in use even at the present day ; and about a quarter of a century ago a self- registering adaptation of it was constructed by Osier, and very generally employed. The third class measures the wind-force by the difference of level it is capable of pro ducing in an inverted syphon, or U tube, containing water or some other liquid. Lincl s anemometer, invented in 1775, is the best known of this type, and is still in common use. It is simply a U tube, with one leg bent at right angles towards the upper end, and attached to a vane, so as to have its mouth constantly facing the wind. It is about half-filled with water, and a scale gradu ated to hundredths of an inch is attached. The difference of level in the legs of the syphon indicates the force of the wind; but it is only by experiment that the pressures corresponding to these differences can be laid down. Thus each instrument requires a carefully-constructed reference table of pressures and velocities corresponding to its read ings. According to Lind, a difference of level of 1 inch indicates a pressure of 5 2 fl> on the square foot, and that corresponds to a velocity of about 32| miles an hour, which is in common language a high wind. A difference

of 3 inches would indicate a storm, and one of G inches a hurricane. Rivalling this in simplicity is a fourth class, that may be added to Dr Robinson's, viz., pendulum anemometers, where the pressure-plate is swung as a pendulum, and indicates the wind-force by its deviation from the vertical. The first of this form was invented by the Marquis Poleni, and gained the prize of the French Academy in 1733. They are still employed as simple rough-and-ready indicators to the eye of the strength of the wind.

Several other anemometers have been contrived, which, from being essentially non-recording, may be mentioned as more curious than valuable. Among these are (1.) A musical anemometer, suggested by Dr Hooke, and constructed by Delamanon, in which the wind was caused to sound automatically, in a set of pipes, a particular note corresponding to its velocity; (2.) Leslie's anemometer, or rather anemometric principle, which deduces the velocity of the wind from its effect in cooling a thermometer, it being sup posed that the rate of cooling is proportional to the velocity of the wind; (3.) Brewster's evaporation anemometer, by which the velocity may be deduced from the amount of water evaporated in a given time from a rough open surface, such as sponge or flannel, the two being in proportion.

Osler's anemometer, the last of the pressure-gauges we shall mention, being one of the most trustworthy and most common registering anemometers in this country, merits a more particular description. It traces with pencils on a sheet of paper (which is moved along, either in a vertical or a horizontal position, at a constant rate by means of clock-work) lines which indicate changes of the wind, both in direction and in pressure. Its general principle will be readily understood without a diagram. Changes of direction are recorded in this way. The axle that carries the vane, and turns with it, has keyed on its lower end a pinion which, working in a horizontal rack, moves it backwards or forwards as the axle turns one way or the other; and a pencil fixed to the rack traces its movements on the register-sheet. Changes of pressure, or wind-force, are measured by means of springs. A square foot of a light metal plate, placed vertically and turning with the vane-rod, receives the impulse of the wind, and is forced against three springs, so arranged that one, two, or all of them may be pressed on according to the violence of the wind. The extent of compression against the springs is transmitted through the hollow vane-axle, by means of chains and pulleys, to a light spring, which presses a pencil gently against the paper. Variations of motion of the upper plate thus correspond to definite lateral deviations of this second spring pencil, and a wavy line is traced on the paper as it moves along. The register-sheet is ruled across its length into twenty-four equal parts, one of which passes the pencil points each hour; and lengthways it is ruled—first, with lines corresponding to the cardinal points, under the direction-pencil; and secondly, on the field of the other pencil, with lines spaced by actual experiment so as to correspond to gradations of 5, 10, 15, 20, &c. lb of wind-pressure on the square foot. Thus the pressure-curve, in the same way as on the indicator of a steam-engine, includes between it and the horizontal line of zero-pressure a space proportional to the amount of work done by the wind on the upper pressure-plate. Of course, so long as the wind blows in one direction, the direction-pencil traces a horizontal line; and if there be no wind at all, both pencils trace the zero horizontal lines.

We have not space to enter particularly into the history of the direct velocity-measuring anemometers. As long ago as 1783, one was contrived on this principle by Mr Edgeworth for the measurement of air-currents; but to Dr Whewell belongs the credit of first appreciating at its true value the velocity as opposed to the pressure measure. In the Cam. Trans., vol. vi., he describes an instrument of his own devising, which was afterwards extensively used. Sir W. Snow Harris recommended it strongly to the British Association in 1844, exhibiting results of his own observations with it during three previous years. Owing, however, to some practical defects, such as the great frictional resistance of toothed wheels and endless screws, which have more effect on the indications of a gentle wind than of a high one, and to the want of an arrangement for indicating the times of the variations, Dr Whewell's instrument has now fallen into disuse. A description of it may be found in Drew's Practical Meteorology. One of the most valuable contributions to this science was made by Dr Robinson, of Armagh, who between the years 1843 and 1846 conducted a number of experiments and calculations on the proper form to be given to the revolving vanes. He found radical defects in the principle of all the vanes previously employed; either their moving power was not sufficiently great to make the frictional correction inconsiderable, or their velocity, in place of being less, was often three or four times greater than that of the wind, requiring greater complicacy of machinery to reduce the speed of the tracing point; or, lastly, their results were not identical, though constructed after the same type. The form adopted by Robinson to remedy these defects is that shown in the figure. Though it had been suggested to him by Edgeworth, it deservedly bears the name of Robinson's Cups (fig. 1), as he was the first to show experimentally and analytically its advantages. The arrangement consisted of four light metal hemispheres C, of as large a diameter as convenient, extended at the ends of light and strong metal radii, so that they shall be as far from

Fig. 1.—Cups and Vane of Anemometer.

the axis of motion as possible, and so reduce the effect of friction to a minimum. They require no vane to keep them facing the wind, that represented by F, F, in the figure having a movement quite independent of the cups, as will be explained presently. He found that the cups move at almost exactly one-third of the rate of the wind a happy simplicity of ratio and this for all instruments made after one type. Experiments were made with sets of two, three, four, and six cups; but the velocity ratio—between the cups and the wind was simplest and most constant with four, the number now universally employed. In the paper already quoted (Trans. Roy. Irish Acad., vol. xxii.), Dr Robinson describes a self -registering apparatus, to which he had fitted the horizontal cups. But we shall describe instead one of the more modern and simple registers, which seems to leave little further to be desired in the way of compactness and accuracy. It is the invention of Mr Beckley, of the Kew Observatory, and is described in full in the Report of the British Association, 1858, p. 306. The outside, or wind-receiving parts are represented in the above figure, and consist of Robinson's cups C, for the wind-velocity, and a double fan F, or windmill governor, for wind-direction. The fans serve as a vane, turning so as to keep their axes constantly at right angles to the wind, and are more steady than the ordinary vane. They are connected with the outer brass tube B, and carry it round with them on any change of wind-direction. The second figure shows the manner in which the motions of the fans and cups are recorded. R is a cylinder, which carries the register-sheet, on which the two pencils P, Q, press lightly. It is turned round at a uniform rate by clock-work in K, making a complete revolution in twenty-four hours, or moving through half an inch per hour. Horizontal lines divide the sheet into hour spaces, marked 1, 2, 3, &c., in the figure. The fan, or direction-axle B (fig. 1), communicates its motions to a mitre wheel M (fig. 2), which, gearing into the bevel wheel T, moves the cylinder D, with its brass pencil P. By a change of the wind from north to south, the pencil would trace a line along the cylinder between the vertical lines N and S. In like manner the velocity of the cups is communicated by a shaft passing through the direction-axle to a mitre wheel W, which moves the bevel wheel N, the cylinder L, and its pencil Q. Vertical lines are drawn on the paper, so that the pencil moves over one space, while the wind travels 10 miles, or the cups one-third of that distance.
Fig. 2.—Recording Apparatus. The velocity range does not extend beyond 50 miles, that being sufficiently high above the average rate of the winds. The sheet requires to be renewed, and the clock-work rewound every twenty-four hours, a matter of only a minute or two. The figure shows a simple brass spiral pencil; but the pencil devised by Beckley for the Kew Observatory consists of a strip of brass fixed spirally on edge, so that for equal increments of rotation of its cylinder its point of contact with the register-sheet shifts along by equal increments horizontally.

Anemometry now forms a most important feature in all meteorological observations, and many important and re markable results have appeared since the invention of self-recording apparatus. See Meteorology.