Modern Gasworks Practice

THE STORAGE OF GAS 675 tliere is no such treatment at our disposal, with the result that practical experience and the lesson of precedents chiefly influence the designer. Herein lies one of the paramormt advantages of the steel tank, our knowledge of which permits of a reason-ably accurate computation of the stresses to which it will be subjected. Occasionally one may come across the most elaborate calculations for arriving at the factor of safety obtaining in a masonry tank worked out to two decimal places. The inconsistency of such procedure should be self-evident when it is remembered that the theories so far laid down for the treatment of these tanks are merely the result of surmise, and are built up from a series of pure structural considerations, which probably have no relation to actual conditions. Some forty years ago a series of formulæ were deduced wherewith it was proposed to calculate the stability of masonry tanks. Tliese formulæ are still occasion-ally used at the present day, and may be foimd quoted in textbooks. A cursory examination of the formulæ, liowever, would appear to show that they contain such elementar/ errors as that of assuming the horizontal pressure of earth. constant for all depths and independent of the distance below the surface. Any one acquainted with the theories of Rankine and others will perceive the absurdity of such an assump-tion. It must be realized that the brick tank of a gasholder is not subjected to a simple bending moment as in the case of an ordinary retaining wall, but stresses are set up on the lines of those occurring in the plates of a cylindrical boiler ; and, instead of the wall tending to tilt over from the base, the whole undergoes, in almost all cases, __ slight circumferential Stretching, whilst there is also a stram in the vertical direction. As in the case of the steel tank, which receives support from the bottom curb and floor space, so is the brick wall restrained to a certain extent by its adhesion to the foundations. H. W. Alrich, in a communication to the American Gas Institute, has shown that the restraint exercised by the foundation will be effective, approximately, a distance up from the bottom of the tank equal to l to | of the total depth of the tank. Accordingly, if in the diagram (Fig. 418) the dotted line represents the tank as constructed, the position and shape it will take up wh.en filled with water will be of the nature of that shown by the füll line. As before explained, the theoretical thickness at the top is nil, but the thickness allowed in practice is such, as to prevent a certain amount of Stretching in the upper portion ; thus there is in reality a restraining influence at both. top and bottom. Restraint due to excess thickness Line of Water Pressure _______________ ,B //////// Restraint due ' ' ' ' to foundation Fig. 418.—Showing effect or Water Pressvre on Masonry Tank. The author has made a particular study of the various mathematical treatments which have been suggested for the computation of the stresses induced in masonry tanks, but finds that on examination they are more likely to be misleading rather