A great part of a Clerk of Works' time is necessarily taken up, particularly during the earlier stages of a building contract, in testing the materials brought on to the site, so as to enable him to exercise sound judgment in their acceptance or rejection. As already stated, diary entries are made upon all these matters, while special entries in a test-book are kept with regard to such materials as cement, the testing of which occupies some time, and is of an elaborate character.

This is one of the most important materials to which tests have to be applied, and with regard to it, as with all others, the Clerk of Works must remember that he is only justified in applying such tests as are mentioned in the specification, or such as will enable him to give a true opinion as to whether the material reaches such a standard as the specification contemplates. If a tensile test is specified, that tensile test should be applied; while if the cement is " to be sound and properly cooled," the tests for these are implied, even although they are not specifically mentioned.

Cement is a difficult material to test, and its proper testing requires apparatus which is not in everybody's possession; consequently, if the more severe tests of tension, boiling, and specific gravity are required, say, in accordance with the specification of the Engineering Standards Committee, and upon small works, it is better to select samples and send them up to some recognised testing association than for the Clerk of Works to test himself. This remark applies also to all other materials which require highly scientific tests to be applied to them. It is most unwise for a Clerk of Works to undertake such if he has had no really scientific training and experience in conducting delicate tests. They lie beyond his province, and he should not hesitate to state this if the question is raised, undertaking only such tests as are comparatively easy to perform on the spot. There is no difficulty, for instance, in applying the test for fineness. What is needed is a measuring-glass divided into one hundred parts, and a small sieve properly graded to the correct number of divisions per square inch, and with wire of standard thickness, such as can be obtained from most mathematical instrument makers. The cement is poured lightly into the glass, filling it up to the point marked 100, and is then thrown on to the sieve and shaken by hand until all passes through that can be got to do so. The remainder is poured back into the measuring-glass; the height to which it rises denotes the percentage of residue. It is perhaps more accurate to weigh the cement before it is sifted, and after, with extremely delicate scales, working out a simple proportion sum afterwards to ascertain the percentage left behind. But this is a little more delicate and difficult, if it be more accurate, and the measured percentage is very generally accepted in place of it.

To test the weight of cement is also quite easy it under the specification it is permissible; but it must always be remembered that it is highly unreliable. A bushel measure must be obtained, and it must be seen that this is stamped by the inspector of weights and measures, it being insufficient to allow such a measure to be made up by the Contractor's workmen, with the possibility that it may be considerably too large. This is filled lightly through a hopper with the cement until it is overfull or heaped, and the surplus is struck off with a straight-edge. The remainder is weighed on an ordinary weighing-machine, together with the measure. The measure is weighed separately, and this weight deducted from the total, giving the net weight of the cement, which is compared with that permitted under the specification. It is usually stipulated that cement shall weigh at least no lbs. per striked bushel, but, as said already, this is an unreliable and old-fashioned test, which is generally superseded by that of specific gravity; which, as it needs special appliances and is difficult to accomplish, should not be attempted by a Clerk of Works

It is also frequently specified that the cement shall be properly cooled. This is usually interpreted to mean that it shall not sting the bare hand when plunged into it, a rough test quite easy to apply. It is also well recognised that if the cement strikes cold to the touch it has probably been overcooled, and has become inert. If anything more than this rough test of coolness is required it is best to refer this also to a specialist, though the application of the test is by no means so difficult as is that for specific gravity. A small pat of cement should be made up with 10 per cent. of its weight of water, and a delicate thermometer bulb plunged into it, say that of a clinical thermometer. The temperature reached shortly after the bulb is plunged in should be noted, as also should the rise of temperature during setting, which should not exceed 6o; while inert cement will show a rise of less than 2°.

Duties of Clerks of Works - Testing Materials 133

If the rise be excessive it shows that the cement needs spreading out on a boarded floor under cover, and turning over with a spade from time to time, until such time as a pat treated in this way shows a rise of temperature which is not greater than 6°. Until then the cement is unfit for use, containing an excess of unslaked lime. Of course, if the cement be inert it is valueless, and must be rejected.

The tensile test usually specified is that a cement briquette, after lying one day in air and seven days in water, should not break under a tensile stress of less than 400 lbs. per square inch. This test is sometimes applied by a Clerk of Works on large works if he is provided with a proper testing machine; otherwise he ought not to attempt it. The cement is as usual mixed with 10 per cent. of its weight of water, and lightly trowelled into a mould having much the shape of an hour-glass, 1 inch by inch section at the smallest point. It is left in the mould, lying on a piece of glass or slate, for 24 hours in order to set, and is then taken out of the mould and put in water, kept as nearly as possible at a temperature of 60° for seven days. It is then removed to the testing machine, and a gradually increasing load applied until it snaps, the load that causes fracture being noted. There are several different makes of testing machines; but in all the same result is achieved.