Fatigue failure of bolted joints can result from either cracking or self-loosening and can be prevented by appropriate design and fabrication methods. Fatigue design rules for bolted joints are available in a number of Standards. In the UK, BS 7608 'Code of practice for fatigue design and assessment of steel structures' covers bolted, welded and riveted joints. In order to comply fully with BS 7608, bolts must be made of a material with a tensile strength equal to or below 785N/mm2 and must conform to the relevant thread and manufacturing British Standards. In BS 7608, welded or bolted connections are separated into different Classes depending on the geometry of the joint and on the loading type. Stress range versus endurance design curves (S-N curves) are given for each fatigue Class.
For bolted joints in shear, where the fatigue loading is transverse to the bolt axis (lap joints), fatigue cracks are likely to initiate from the edge of a bolt-hole due to the stress concentration. However, if the bolt is pre-tensioned close to its proof stress, fatigue cracks are more likely to initiate from fretting between the mating surfaces, away from the hole. For lap joints, the stress range, S, is defined as the nominal stress range in the member on the net cross sectional area. In shear, the best fatigue performance is achieved with reamed bolt holes used in conjunction with high strength friction grip bolts thus allowing for higher clamping loads to be applied, typically around 75% of the proof load.
Bolts under axial tensile fatigue loading are defined as Class X in BS 7608 and are most likely to fail in the first thread engaging the nut, in the radius between the bolt head and the shank or in the thread-to-shank run-out. For bolts under tensile loading, the axial stress range is calculated by the effective tensile stress range divided by the nominal tensile strength of the bolt material. The stress calculation should include effects of axial and bending loads, as well as any prying action. For diameters above 25mm, a thickness correction factor needs also to be applied.
BS 3580 gives useful guidance on the different factors affecting the fatigue resistance of V-shaped screw threads. Fatigue cracking can be prevented by an adequate pretension of the bolt during initial tightening so that the major part of the fluctuating load is transmitted through the mating parts and not through the bolt itself. Where possible, the mating parts should be machined and accurately fitted together prior to tensioning. Under shear, the movements of the bolt head during cyclic loading could also result in self-loosening and loss of the clamping force. An adequate preload during the installation minimizes these movements and increases the fatigue strength of the connection. Preload can be applied by torque tightening methods such as with a calibrated wrench or by the turn-of-the-nut method. A simple vibratory test, the Junker test, also allows for the vibration-resistance of a bolted joint in shear to be experimentally determined.
- Bickford J H: 'An Introduction to the Design and Behaviour of Bolted Joints', 2nd edition. Publ: Marcel Dekker, 1990. ISBN 0-8247-8167-8
- BS 3580: 'Guide to Design Considerations on the Strength of Screw Threads', BSI London, 1964
- BS 7608: 'Code of Practice for Fatigue Design and Assessment of Steel Structures', BSI London, 1993
- Kulak G L, Fisher J W, Struik J H A: 'Guide to Design Criteria for Bolted and Riveted Joints', Second Edition, Publ: John Wiley and Sons, New York, 1987. ISBN 0-471-83791-1