The integrator, another and still more ingenious instrument, by M. J. Amsler-Laffon, was invented theoretically shortly after the planimeter just described (in the year 1855), but was first constructed for practical use in the year 1867, the first instrument made being exhibited in the Paris International Exhibition in the year named. It was not introduced into England till the year 1878, and although adapted for other uses than those involved in scientific calculations connected with shipbuilding it was in this connection that attention was first seriously directed towards it. In 1880 the late Mr C. W. Merrifield described the instrument, and traced the mathematical principles upon which it is based, before the Institution of Naval Architects, and in 1882, before the same body, Mr J. H. Biles, naval architect for the firm of Messrs J. & G. Thomson, called attention to the usefulness of the instrument in stability investigations, showing by specimen calculations and other particulars its great adaptability to this class of work, even in the hands of youthful and untrained operators. A still more recent and exhaustive paper devoted to the claims of the integrator upon naval architects was read before the same Institution by Dr A. Amsler, the son of the inventor, at its last meeting. This paper was chiefly concerned with demonstrating the advantages of the integrator in respect of time saved, as well as in respect of its great accuracy.
FIG. 27.
AMSLER’S MECHANICAL INTEGRATOR.
The object of the integrator is to find at one operation the area, the statical moment, and the moment of inertia of any closed curve or figure by simply tracing out the curve with a pointer, the results being read off directly from the instrument, as in the case of the planimeter, and with a correspondingly small amount of after calculation. As shown by Fig. 25, the essential parts of the integrator are a rail L, having groove with which to guide the wheels p and q of a carriage provided with rollers D1 D2 D3 moving on the surface of the drawing. The contour of the figure to be dealt with is traced—in the direction of the movement of the hands of a watch—by the pointer F, this pointer being attached to an arm moving on the vertical centre of the instrument while the whole mechanism runs to and fro on the rail L. Under these conditions the rollers D1 D2 D3 execute movements partly rolling, partly sliding, and by readings taken from the divisions engraved upon their circumferences at the beginning and the end of the whole movement, together with simple arithmetical processes, the nature of which may be inferred from the explanations given of the planimeter readings, the three quantities sought are arrived at.
In a valuable appendix to the paper read by Dr Amsler, before the Institution of Naval Architects, specimen sheets are given of several calculations, of a vessel of about 4000 tons, the forms in which the figures are entered being so arranged as to avoid all unnecessary trouble in measuring and calculating, and to contain at the same time a check on the results. The accuracy and the speed of working depend, of course, to a considerable extent on the person using the integrator, but as showing what can be obtained with the instrument after some practice, the specimens given in the paper referred to are certainly remarkable. For the calculations of the data necessary for the construction of the curves of displacement and vertical position of centre of buoyancy, the complete integrator and arithmetical work took only two hours; for the data requisite for the curve of displacement per inch immersion, and transverse metacentre one hour was taken; and for the complete calculation, affording data to construct a stability curve, the time taken was only eight hours. A similar calculation done in the ordinary arithmetical method, and giving results far less reliable, would have taken as many days. All the work, it should be added, was done without the aid of an assistant. Amongst other calculations besides displacement and stability in connection with which the integrator is greatly advantageous, are those concerned with the strength of vessels and with the longitudinal strains to which they are subject at sea through unequal distributions of weight and buoyancy, already fully referred to in the chapter on scientific progress.
BENNETT & THOMSON, PRINTERS.
PORTRAIT
AND
BIOGRAPHICAL NOTE.