This paper is concerned with the problem of fatigue design for random loads. A theory is outlined which allows the fatigue life to be predicted from the constant amplitude stress range–life curve and the standard deviation and ruling frequency of the stress signal. Laboratory tests are described which verify the theory for welded steel constructions in a railway environment. The results are also analysed by a more traditional method. Areas where the theory might usefully be extended are identified.
Get full access to this article
View all access options for this article.
References
1.
RobsonJ. D.An introduction to random vibration, 1963 (Edinburgh University Press).
2.
BS 5400, Part 10: 1980Steel, concrete and composite bridges. Code of practice for fatigue (British Standards Institution, London).
3.
TunnaJ. M.Theoretical and experimental aspects of two cycle counting methods for fatigue life prediction. Fatigue of Engineering Materials and Structures (To be published).
4.
BilyM.BukoveczkyJ.Digital simulation of environmental processes with respect to fatigue. J. Sound and Vibration, 1976, 49(4), 551–568.
5.
WatsonP.DabellB. J.Cycle counting fatigue damage. Statist. Aspect of Fatigue Testing Symp., Warwick University, February 1975.
6.
SocieD.Fatigue life extension techniques. Technical Report 145, 1982. Department of Mechanical and Industrial Engineering, University of Illinois.
7.
GurneyT. R.Fatigue tests under variable amplitude loading. Welding Institute Research Report 220, 1983.
8.
PookL. P.The effect of mean-stress on fatigue crack growth in cruciform welded joints under non-stationary narrow band random loading. NEL Report No. 690, Dec. 1983.
9.
TunnaJ. M.ArmstrongR. A.The application of the standard deviation/life approach to fatigue life prediction of aluminium alloy welds subjected to variable amplitude loading. British Rail Research Technical Memorandum TMMF 87, Jan. 1984.
10.
WirschingP. H.LightM. C.Fatigue under wide band random stresses. J. Struct. Div. ASCE, July 1980, 106 (ST7).
