FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1928
1928 - 0902.PDF
SUPPLEMENT TO FLIGHT 72 SEPTEMBER 27, 1928 THE AIRCRAFT ENGINEER but is available in practically any standard section, although rather expensive. The composition is not available for publication, but the physical properties on round tubes are :— Yield. Ult. Elong. per cent. As drawn 509 52-4 26 Softened 18-0 37-2 70 Welded with own material ... 18-8 34-1 11 This material hardens when cold worked.but can be squared without difficulty. On section tubes the figures are slightly inferior to those on round tubes. I have not yet had suffi- cient opportunity to test the strip material for rolling opera- tions, but I understand that considerable success has been achieved with softened strip hardened and tempered. There is doubtless a big future for stainless materials on aircraft, especially when prices become more competitive. It is admitted that there are certain difficulties in •working to be overcome, but to obtain a solution of the corrosion problem it is worth while. On some experimental work I have used stainless plate in preference to high-tensile sheet, and found that, despite the high initial cost, a saving has resulted, owing to the fact that heat treatment and test pieces were avoided. Solid and tubular rivets are available and work without difficulty. It is a matter for individual taste as to the finish employed on plate work. Personally, I think that if the fitting is sandblasted and then varnished, the result is pleasing and useful. DESIGN STRESSES By CECIL D. HOLLAND, A.M.I.Ae.E. The weight of any aeroplane structure depends upon several factors, the following two being of prime importance :— (a) The load factor, (b) The design stresses of the materials of construction. A load factor may be denned as :—The maximum possible load that may come on the structure, either by an air or ground manoeuvre, divided by the normal load (i.e., load in horizontal flight, or resting on the ground), plus a small amount for safety and errors of workmanship. The design stress may be defined as :—The maximum stress the material would be subjected to if the structure were loaded to the amount "" The load factor times the normal load." The lightest structure weight is possible when the load factor is minimum and the design stress is the maximum. The objects of this article are :—(a) To propose a policy for the determination of design stresses, (b) Comment on the needs of the design stress policy from a load factor policy. Up to the present, the two policies have had a very hap- hazard co-operation, such as (a) high factors and high stresses ; and (b) low factors and low stresses. In the past. similar degrees of indifferent satisfaction have been obtained due to the two errors, in each case, tending to neutralise each other. In present-day structures, when using the latest materials, this satisfaction becomes even more remote. It has been suggested, and even practised, to vary the load factor for different materials on the score of the so-called reliability of the material, e.g., reduction of the load factor for steel struts, as against wooden ones. The real point appears to be this. What is actually meant by the term " reliability ? " The dictionary meaning is "' that may be relied upon " ; relied upon to what ? This calls for a further definition. E.g., when one refers to the '" reliability " of a certain aero engine, one infers the degree of freedom from breakdown in the denned times between overhauls. Thus, various types of engines may all have equal degrees of "' reliability " when the defined times between over- hauls are adjusted to suit each particular type of engine. Now, consider the case of materials. Take the results of a large number of tests on any one type of material; one can determine the maximum, mean and minimum values. If the teats have shown great variation in the results, the reliability relative to the maximum value would be very poor, since the results of any future tests are unlikely to reach or pass this value. The reliability relative to the mean value would be much better, because the results of about 50 per cent, of any future tests would reach or pass this value, and so the reliability is likelv to be very good when the minimum value is used. Moreover, 100 per cent, reliability can be approached if a sufficiently low value be chosen for the reference point, for one does not infer infallibility for any future test. It is generally agreed that the values given in the specifica- tions of materials are sufficiently low. so that any future tests can be relied upon to surpass these specification minimum values. It may be argued that it is the treatment the material receives at the aircraft constructors' works which introduces unreliability. This should not be so. as the effect of all treatments should be known, and are controlled by the double inspection : firms and A.I.D. The effect then, of working to a specification is to reduce all materials to one common degree of reliability, and thus the first comment on the load factor policy will be " That the load factor shall be constant, irrespective of the materials of construction." If this is not so. then the previously given definition for the load factor cannot stand. For the purpose of this article it will be assumed that the amount allowed for in the load factor for safety and errors of workmanship is small compared with the remainder, and that the ultimate strength values of all materials is the minimum required by the specifications. At first sight it would therefore seem natural to use the specification's ultimate figures as suitable values for the design stresses. This has not proved satisfactory in practice due to the "" permanent set " occurring in some parts of the structure, showing that the normal manoeuvres produce stresses which exceed the yield points of the materials used ; moreover the tubular and built-up metal struts and spars invariably fail by local buckling before the ultimate stress is reached. This state of affairs produced advocates for the use of the yield point stress as a suitable design stress. This policy has many disadvantages such as :— (a) Increased structure weights. (b) Excessive weight of non-structural members. (c) Tendency to cause the disappearance from use of the light alloys because of their relatively low yield points. (d) The inability exactly to determine the yield point of many of the materials, vide B.E.S.A. Pub. No. 56. Having no clear-cut policy, the outcome of the two before mentioned view points was a half-hearted compromise, with nothing like a complete set of design stresses, merely a few tensile figures and an odd compression or shear stress, vide A.P. 970. Disadvantage (d) in the yield point policy has been over- come to some extent by the introduction of the term " proof stress."' The definition for the proof stress adopted by the B.E.S.A. for all non-ferrous materials for which a proof stress is likely to be applied is as follows :—" When the proof stress is applied to the specimen for a period of 15 seconds and removed, the specimen shall not have received a permanent set greater than 0-15 per cent, of the gauge length.'" No proof stress is included in B.E.S.A. aircraft steel specifications, but a proof stress is given in D.T.D. specifications for strip steel. Experience indicates that the design stress should be a value between the yield point (or proof stress) and the ultimate failing stress, the problem being to determine this value. Xow the manoeuvres of an aeroplane may be divided into two groups, normal and abnormal. For the purpose of this article it is not necessary exactly to define these groups beyond the statements that the abnormal manoeuvres occur very seldom, and that the worst abnormal manoeuvre is used to determine the load factor (definition given above) and that under normal manoeuvres the structure shall not suffer permanent set; in other words, the stress produced under normal manoeuvres shall not exceed the yield (or proof) stress. If it is possible to determine the ratio of the loads induced 834d
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
