THE TESTING OF MATERIALS USED IN CONSTRUCTION. By W. C. POPPLEWELL, M. Sc., Assoc. M. Inst. C. E.

(From the "Mechanical Engineer" [England], Feb. 19. 1898)

In all engineering works, whether they be of masonry, concrete, or metal, the question of strength is one of the most important which has to be considered; and where any material is to be so placed in a structure as to have to withstand considerable stress, some previous knowledge of the strength properties of this material must be within the reach of the engineer who contemplates its use.

In every engineering structure there are two distinct points with regard to its strength which must be aimed at. These are that, in the first place, the structure must be of ample strength to withstand any and all loads to which it may be subjected without permanent injury to its parts; and, secondly, that there should be no more material employed in any one part than is sufficient to ensure the fulfilment of the first condition. The chief reason for this latter condition is sufficiently obvious-it is simply a question of cost. But, apart from pounds, shillings, and pence, it is obvious that it would be an unwarrantable waste of valuable material to introduce any such beyond that which is necessary to ensure safety and efficiency, and, further, it is in many instances advisable to reduce the weight of a structure as far as possible, so as to thereby lessen the stresses due to the weight of the structure itself.

In the case of many engineering structures and works it is usual to apply some kind of test load after the completion of the structure, so as to provide a final check on the work before submitting it to its ordinary load. Thus, bridges are often tested in this manner, by putting upon them loads greatly in excess of those to which they may be expected to be subjected in daily use. Again, boilers are treated in a somewhat similar manner, by applying internal hydraulic pressure double, or more than double, the intended working pressure of the steam. There are, however, many structures which cannot be put to these final tests, owing to the nature of their load and the position of the structures themselves. Examples of such are to be found in roofs and bridges subject to wind pressures. But, even if it were possible to apply these after tests in all cases, it would be preposterous to construct a great and costly work entirely by "rule of thumb," and trust to luck that it would be strong enough to withstand the loads which might be applied to it.

Some previous knowledge, both of the loads to which each. part of a structure may be expected to have to withstand and of the strength of the materials forming these parts, is absolutely necessary. In former times, before such knowledge existed of

the strength of the various materials of construction as is now the case, an engineer had to rely to a great extent on his own or other people's experience, and on his own judgment and instinct for the dimensions of the parts of his structures. Such a plan is of necessity still followed out to a great extent, especially in the case of small and unimportant parts, and also very often in timber structures. It is not an uncommon thing in drawing offices to see a draughtsman vary a dimension several times until he thinks "it looks right." This is perfectly legitimate and allowable in many cases, and is, indeed, necessary in such cases, for instance, as the frames of machines and other structures subjected to a great variety of indeterminate stresses, or to very slight stresses, where stiffness alone is necessary. But in large and important structures and pieces of machinery a certain definite plan must be, and nearly always is, followed.

In the first place, the engineer must know what loads his projected structure will have to sustain as a whole, what part of these are live loads, what part dead loads, and if any shocks are to be expected, and of what magnitude.

Secondly, knowing what the loads upon his structure are to be, he must next be able to calculate what will be the effect of these loads in producing stresses in the several parts of the structure, what are the magnitudes of these stresses, and what their nature; that is, whether the stresses are those of tension, compression, bending, or torsion.

Thirdly and lastly, he must have an intimate knowledge of the properties of the materials he intends to make use of, so as to be properly guided in determining the shape and dimensions of all the parts.

Of these three divisions the first entirely depends upon the knowledge which the engineer may have as to the conditions pertaining to the particular case in question; the second requires calculations depending upon the principles of mechanics, and may be performed by simple arithmetical or algebraic methods, or the problem may be attacked graphically. When these two first conditions have been determined, the last step in the design of the structure may be taken, that is, knowing the stress which may be expected to act on each part, both as to its nature and magnitude, and also knowing what materials are to be used, the engineer can at once proceed to ascertain the form and dimensions of these parts, provided he is fully acquainted, or has some means of becoming acquainted, with the strength properties of the materials. It is towards the fulfilment of this last mentioned condition that the "testing of materials" is undertaken.

For such a case as we have just considered, it is usual to test several small samples of the material to be employed, as in the case of a strip of steel cut from a boiler plate, on the assumption. that the whole of the material is uniform and similar in its properties to the specimens tested, or, as in the case of a chain or rope, to subject an actual piece of the work itself to the test.

Such tests have been called "commercial tests," and may be either specially made in connection with the structure in question, or have been previously made on some samples of similar material. Besides tests of materials carried out for purely commercial and constructive purposes, there is a great deal of work being done at the present time, and has been done in the past, of a more refined and scientific character, with the intention, in most instances, of elucidating the more hidden and complex phenomena displayed by materials under various stresses and under different conditions of stress. Such research has been going on for many years, and is going on most actively at the present time. The knowledge obtained in this way is to a great extent permanent, and though, as time goes on more facts are constantly being unearthed with regard to the strength properties of materials, still these new facts are not, as a rule, such as to render unimportant the knowledge already gained, although it is certainly so in some cases. The value of this scientific or research testing may not be at once apparent, but every test and experiment made in this way is going to help to build up a more complete knowledge of the mechanical properties of materials.

There is another branch of testing that comprises within its limits both those already mentioned, namely, the testing carried on for purely educational purposes. Most of the colleges, or departments of colleges, devoted to the scientific education of engineering students possess some kind of a testing laboratory. Here, in these laboratories, the students are taught by personal instruction and actual experience to make tests of various materials, and in this way they are not only enabled to learn the details of the appliances used and the methods and systems employed, but their faculties of observation are called into play, and the properties of the materials they are dealing with are brought home to them in a manner not possible by mere description and the study of books. The number and completeness of the laboratories at technical schools and colleges is increasing year by year, and this is as it should be, because no branch of the scientific education of a young engineer is of greater help to him in after life than the time spent in laboratory work.

In the following chapters it will be the aim of the author not. only to describe the various testing appliances and methods as used in purely commercial work, but, at the same time, to make these descriptions applicable to the work of an engineering stu

dent.

It has been stated that testing can, according to the apparatus and methods employed, be divided into commercial and scientific testing. These two are not necessarily quite distinct and apart, in many cases they overlap, but, generally speaking, the methods of commercial testing are more crude and the measuring appliances used not of so refined and delicate a character as many of those used in purely research testing, which partakes more of the nature of physical laboratory work. In commercial testing cer

tain standards are usually fixed by purchasers of materials and by certain competent authorities, such as the Board of Trade and Lloyd's, and in the tests to which they are subjected the specimens are expected to exhibit such properties as are required to comply with the standards and regulations laid down. In making a commercial test, therefore, it is necessary to know what properties must be especially observed, so that it may be determined whether the tests do satisfy the requirements, and then to apply such, and only such, tests as may be needful for this purpose.

In scientific testing the case is different. The observations in a test are generally greater in number, more accurately made with. apparatus of greater precision, enabling the observer to see more deeply into the phenomena exhibited during the tests.

All testing requires a considerable amount of skill, experience, and sound judgment in its execution, and to these should be added some knowledge of mechanics, so far as the "strength of materials" is concerned. System and order should be rigidly adhered to, both in the carrying out of tests and in the manipulation and presentation of the results. Every detail should be most carefully watched and attended to, as one mistake may render useless a test or even a whole series of tests. Nothing is of more importance than a correct idea of perspective so far as accuracy is concerned, and a clear knowledge of the necessary limits of accuracy to be aimed at, and which are possible in the various kinds of work undertaken, should be most carefully cultivated Useless attempts at extreme accuracy, where extreme accuracy is neither necessary nor possible of attainment, are always absurd, and in some cases actually mischievous.

The variety of material which is tested, or capable of being tested, is very great.

At the present day most engineering structures, whose design is governed by considerations of strength, are constructed of either iron or steel. These metals have asserted their preeminence in such work by reason of their combining the advant ages of cheapness, strength and durability to an extent not found. to exist in the case of any other material. The kinds of structures built of iron and steel are very numerous. In addition to the bulk of machinery used for purposes of manufacture, the engines which give the motive power to this machinery, and the shafting and gearing which serve to transmit the power from the engines to the machinery, in addition to their use in formation of structures such as these, iron and steel are greatly used for what, in one sense, are more important works. By these are meant structures whose collapse or failure would endanger human life. Of course, this possibility exists in the case of most machinery of any size, but it is especially evident in such structures as bridges, boilers, railway appliances and steamships. In most of these strength is of the first importance, although considerations of form do in most cases affect the design. In addition to iron and steel there are other metals in use for engineering purposes,

although in these, other qualities rather than strength, render their use desirable. Such are copper, brass, gun metal, tin, zinc, lead, aluminum, and various anti-friction alloys. Of these there are several where considerations of strength do enter largely into their use. For instance, copper is used for boiler fire-boxes, stays, steam pipes, and for overhead electrical conductors, where it has to withstand great tensile loads; gun metal also is often used in machine and engine parts where it has to undergo considerable tension or compression. The increasing use of aluminum and its alloys makes it necessary for its strength properties to be known; and lastly a knowledge of the compressive strength of anti-friction alloys in bearings is often needed.

Timber is used to a great extent in constructive work, especially for temporary work. It is, however, too uncertain a material to allow structural parts to be designed with the same certainty that exists in the case of the metals, without the use of a large margin of safety. Still, tests of timber, although approximate in character, are often necessary and useful, and give reliable information when properly carried out and judiciously applied.

Last among the materials of construction which are tested are those substances which are employed in works of masonry and those of a like nature, chief among them being the different kinds of natural stone, bricks and terra-cotta, cements and limes, and these combined with other substances to form the various kinds of concrete which are used. So much depends on the strength and reliability of these, which are very largely used in the case of buildings, retaining walls, bridges, foundations, and harbor works, and so great is the variation in quality of apparently similar substances, that they present a large field for testing operations.

It will be clear from what has been said that the variety of substances which are subjected, or may be subjected, to tests is very great, and not only do these many substances used in constuctive. work present many differences in their qualities and behavior under test, but there are many ways in which any given substance may be tested. Take for example mild steel. It may be used for boiler plates, when its behavior under a tensile test becomes important; if for the rivets of the same boiler, its shearing strength should be known, it may be used in the manufacture of pillars or struts, when its compressive strength is required to be known, or, it may be that the steel is required to construct a propeller shaft of, when its properties under a test in torsion are necessary.

And so it is all through. As a rule, it is not sufficient to know the general strength properties of a substance, but the properties which exhibit themselves under special circumstances, and when made of special forms, must also be known, and the tests applied should be always judiciously selected and carried out, so that the actual conditions of use shall be as nearly as possible satisfied.