FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1926
1926 - 0714.PDF
SUPPLEMENT TO FLIGHT SO THE AIRCRAFT ENGINEER SEPTEMBER 30, 1926 been pretty well borne out in practice, though of course, a certain number of exceptions occur which indicate that sometimes the maximum stress is also slightly affected by the grain. In Duralumin the ductility and the toughness are lower in a specimen tested in tension when the stress is applied in a direction at right angles to the governing direction of the grain than it is in a corresponding specimen tested so that the stress applied is parallel to the grain. As with steel, the degree of difference in properties in the two directions is not absolutely constant and depends, amongst other things, upon the amount by which tho metal has been reduced in forging. The light alloys, in addition to showing the effect of a cross grain in theif ductility and toughness, also show the influence of grain in their tensile properties. The extent to which the grain affects these properties is very particularly influenced by the amount of mechanical work that has been done upon the metal. In a specimen that has only been lightly -worked the difference in the results obtained from longitudinal and transverse test specimens is not very great. When more work has been done the disparity in properties in the two directions increases. If much further work is done the disparity both in strength and in ductility begins to lessen, and in fact, finally disappears in moderately rolled metal. Obviously it is somewhat difficult, therefore, to estimate exactly what will be the difference in properties between longitudinal and transverse specimens, without complete information as to the size of both the original casting and the finished parts, as well as of the conditions under which they have been produced. The safest thing to do appears to be to consider the worst case, namely, the intermediate condition between slight forging and heavy working. A general survey of a considerable quantity of tests indicates that the difference in tensile strength between representative specimens, longitudinal and transverse, will be of the order of 3 to 4 tons per square inch. This means that a tensile specimen cut longitudinally from a fully heat treated and aged forging would have a maximum stress of 25 tons, whereas a transverse test piece cut correspondinglv would have a maximum stress of only about 21 • 5 to 22 tons per square inch. This represents the greatest extent of the disparity in proper- ties in the two directions. These variations in the strength and ductility of a forging should, of course, be taken into account by the designer. Probably it is only in large forgings that any serious notice need be taken of the mechanical effects of the grain, and it seems reasonable to assume that the effect can be taken into full account by assuming a uniform strength of about 22 tons per sq. in. throughout the forging, except where the influence of mass is exceedingly prominent. As has been stated above, the difference in strength of Duralumin in the two directions parallel and perpendicular to the grain becomes quite negligible in rolled metal. In such material there is no need at all to make an allowance for any difference in properties parallel to, and at right angles to, the grain. This applies to the ductility as well as to the strength properties of the metal. This seems to be a convenient stage at which to make some reference to a property of Duralumin which is frequently misunderstood in its scope and effects. It is frequently stated that Duralumin suffers from " laminations," and very frequently these " laminations " are popularly ascribed to the effects of blow-holes. The most usual way of demon- strating the existence of these alleged defects is to fracture a specimen of the metal by bending. Applied to thick material this form of fracture sometimes reveals smooth facets lying amongst the torn and distorted ends of the fibre •which constitute the great majority of the fracture. These smooth facets have, in the past, been referred to as laminations. This is clearly a misnomer, and it is usual to recognise this fact now by referring to the facets as " discontinuities." It can be taken as axiomatic that these discontinuities are a regular feature of wrought light alloys, and that they have their origin in certain phenomena that are inseparably connected with the behaviour of these materials during casting. They are not blow-holes, neither are they the offspring of blow-holes. On the other hand, they are not localities marked by non-metallic inclusions, neither are they places from which non-metallic inclusions have been removed during the process of fracture. Undoubtedly these discontinuities, which are invariably found in some measure in wrought light alloys, are the product of the arrangement ol the macro-crystals of the metal in the cast state. These arrangements, of course, can be controlled to a certain extent by the foundryman, and it is possible by inferior skill in this operation to produce such an arrangement of the crystals as leads to the presence of discontinuities of an unusual size, or to the occurrence of an unusually large proportion of them in the metal. What Appears, however, to be entirely certain is that up to the present no method of casting has been developed which results in Duralumin or the other wrought light alloys being entirely free from these discontinuities. It is a very striking thing that the discontinuities only become evident when the material is fractured in bending— that is, in such a way that the maximum opportunity is provided for the separation of contiguous crystal faces. In other words, the display of the smooth facets is a function inseparably connected with severe cold distortion. If a sample of metal, containing discontinuities to a normal extent, is sectioned and examined microscopically, no evidence whatever can be found of those places which, when the metal is fractured, give rise to the facets termed discontinuities. This, of course, is entirely in harmony with what has been said respecting their method of origin, and with the fact that the discontinuities are not associated with blow-holes or non-metallic inclusions. Evidently then the discontinuity which shows itself after fracture is not an internal lesion in the material existing apart from the fracture, any more than the boundary line between two crystals of ferrite in mild steel is an internal defect. (The evidence appears to tend in the direction of showing that there is a good deal of similarity between the two examples mentioned.) A further point of interest in this connection is that it is only possible, in the ordinary way, to detect discontinuities when the material is in a comparatively thick condition. It is, of course, in this condition in which the material can most readily be fractured by tearing, but apart from this fact there is ample evidence that the discontinuities do disappear from observation when Duralumin is rolled down to sheets of ordinary thickness. This could not happen if the discon- tinuities were serious internal lesions. It is quite certain that such defects, if they existed, would not weld up again during cold working, and the fact that they cannot be distinguished in severely worked material seems to point again to the fact that discontinuities are merely the attributes of specific arrangement of crystalline material, and that they are not internal defects. Another feature of great importance is the mode of distri- bution of the discontinuities -. When these phenomena were first commented upon and were miscalled laminations it was frequently imagined that they ran for considerable distances through the metal. In other words, it was thought that if two fractures taken near together from a piece of metal, each showed a " lamination," there was evidence of the same " lamination " in each fracture, and that the " lamination," therefore, extended for a considerable distance through the material. This idea has definitely been disproved, and it is quite certain that individual discontinuities are quite small affairs. Their distribution is akin to that of an attenuated shoal of small flat fish, which is quite different from that of a squashed whale. In Chestertonian phrase the conspicuous feature of the distribution of discontinuities is its discon- tinuity. In view of the above explanation of the nature of these characteristic formations in Duralumin it becomes possible to assess with some reasonable prospect of success the influence of the discontinuities upon the mechanical pro- perties of the material. It is obvious from the outset that since discontinuities which are revealed on fracture are not pre-existent internal cracks, they cannot be regarded as potential sources of weakness in material that is exposed to normal stresses. In other words, they are not portions of the material at which there will be local concentrations of stress, neither are they localities infested with sharp corners, 6366
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
