A Treatise On The Principles And Practice Of Dock Engineering

PORTLAND CEMENT. 121 often develop the highest ultimate values. Uniformity of results is a great desideratum. Oonsiderable divergency in the results is a most unsatisfactory feature, no matter how high the average may stand. It should not fail to be noted that the care taken in the préparation of the specimen briquette, and the method of applying the testing weight, exercise a very considérable influence on the results obtained. In Germany, much importance is attached to a test in which the cement is mixed with standard sand, on the ground that the cementitious power of the cement can only be estimated properly on this basis. Indeed, it has been found that of two samples of cement, one finely and the other coarsely ground, the finer cement was the weaker of the two in the neat condition, but much the stronger in combination with sand. The test has also been introdueed into this country, not with any unanimity of approval. It is difficult to procure a standard sand of rigid uniformity, and the efliciency ot the test suffers in conséquence. The criterion usually adopted is passage through a 400-mesh sieve and retention by a 900-mesh sieve. A briquette made with 3 parts of such sand to 1 of cement should exhibit a tensile strength of, at least, 150 Ibs. per square inch at the end of a week, with a regular inerease, as the period is lengthened, to 250 Ibs. at the end of a month. Compressive tests are also used in Germany, and not without reason, tor concrète is particularly designed to withstand compression, whilst its use in positions of tension is strictly prohibited. The ratio of the compressive strength of Portland cement to its tensile strength may be taken at about 10 to 1. The only objection urged against this course apparently is that ■“ Portland cement will bear a greater (compressive) stress, without fracture, than it can be subjected to in practice.”*—an argument which, like the boomerang, has a curiously reflex action. It may pertinently be asked wherein the distinction lies, that the statement is inapplicable to tensile stress. The author is of opinion that an extensive series of experimental results in compression would be a very valuable addition to our data on Portland cement. Time of Setting. —The time of setting for ordinary cement, under normal ■conditions, will vary between two and five hours. Slowness in setting is, generally speaking, indicative of strength. A quick-setting cement probably contains an excess of clay, but fine grinding has also an appreciable effect in accelerating the setting action, in some instances to such an extent as to justify special retardative measures. The usual way is to thoroughly aerate the cement by spreading it over a floor, under cover, to “ cool,” by which means the aluminate of lime becomes partially hydrated and its activity moderated. Sulphate of lime or gypsum, added to the cement during manufacture, retards the setting action, but any excess over 2 per cent, is harmful. Common soda accelerates hardening, though it weakens the * Shaw on “ Portland Cement,” Min. Proc. L.E.S., vol. xix.