Section 2 Fatigue – General considerations

2.1.1 In assessing fatigue performance, the effect of cyclic loading should be taken into consideration. The types of cyclic loading experienced in service, will, in general, depend upon the size, type and speed of the craft. Cyclic wave induced loads created by the passage of waves along the craft sides and the associated local structural response create the highest risk for fatigue damage and cracking for large craft whereas, the major consideration for small high speed craft will relate to impact considerations, hard spots and discontinuities in the structure, etc.

2.1.2 Loading associated with bottom structures, in terms of hull girder response and local pressure variation, are heavily influenced by the length of waves and their direction in relation to the craft length and draught. Similarly, the deck structure is exposed to hull girder loads in response to relatively longer waves and cargo local loads arising from craft responses. Alternatively, these can be associated with structural discontinuities and slamming in shorter waves.

2.1.3 Deck and bottom longitudinal structure require attention to detail design and structural continuity to reduce the effects of stress concentration. Depending on the type of craft, the side shell structure may be exposed to dynamic loading from the internal pressure head from storage of consumables or cargo, in association with wave induced pressure variations, resulting in high cycle local bending stresses applied to the longitudinals and connection details at the transverse bulkhead stiffeners. This may transfer moments resulting in further increased stresses in the side longitudinals.

2.1.4 Every stress concentration and welded joint is a potential source of fatigue cracking and the design, taking note of symmetry, should reflect this. To ensure the structural integrity of the stiffening members particular attention should be given to the detail design. This document provides initial design guidance on fatigue and includes recommendations for the improvement of welded joint fatigue strength, or the bonded joint fatigue strength in the case of composite structures.

2.1.5 The fatigue strength of a structural detail is dependent on the following factors:

1. The direction of the fluctuating stress relative to the detail.

2. The location of initiating crack in the detail.

3. The geometrical arrangements and relative proportion of the detail.

It may also depend on:

1. The material (unless welded).

2. The method of fabrication.

3. The degree of inspection.

2.2.1 Fatigue damage starts prior to the initiation of a crack. With repeated loading, localised regions of slip (plastic deformation) develop. These deformations are accentuated by repeated loading, until a discernable crack finally appears.

2.2.2 The initial cracks form along slipped planes. The crack is crystallographically oriented along the slip plane for a short distance. This is sometimes referred to as Stage I crack growth. Eventually the crack propagation direction becomes macroscopically normal to the maximum tensile stress. This is referred to as Stage II crack propagation, and it comprises most of the crack propagation life.

2.2.3 The relative cycles for crack initiation and propagation depend on the applied stress. As the stress increases, the crack initiation phase decreases. At very low stresses (high cycle fatigue), therefore, most of the fatigue life is utilised to initiate a crack. At very high stresses (low cycle fatigue), cracks form very early. The separation of high and low cycle fatigue is not clear-cut. Generally, the low cycle region is that which results from stresses that are often high enough to develop significant plastic strains. It is usually assumed that the separation zone for low and high cycles is of the order of 10⁴ – 10⁵ cycles to failure.

2.2.4 There are visual differences between high cycle (low stress) and low cycle (high stress) fatigue. In the latter, deformation resembles that seen with unidirectional loading. Strain hardening can occur and the slip bands are coarse. In high cycle fatigue, the slip bands are usually very fine.

2.3.1 For a particular craft component the allowable stresses should be in accordance with the requirements of LR's Rules and Regulations for the Classification of Special Service Craft, July 2022 (hereinafter referred to as the Rules for Special Service Craft).

2.4.1 Applied design loads must take into consideration the appropriate environmental and dynamic conditions. Such loads are indicated in Pt 5 Design and Load Criteria of the Rules for Special Service Craft.

2.4.2 Realistic assessment of the fatigue loading is crucial to the estimation of fatigue life. Little or no published data for loading exists for the types of craft covered by the Rules for Special Service Craft. LR is conducting research in this field, the results of which will form the basis of Rules for Special Service Craft’ requirements in the future.

2.5.1 The design, fabrication and construction of all structural details should be based on procedures and processes to minimise stress concentrations.

2.5.2 Fatigue strength is seriously reduced by the introduction of a stress raiser such as a notch, of the method of termination of stiffeners and brackets, etc. or a hole. Since actual hull structure elements invariably contain stress raisers like fillet welds, end brackets, cut-outs, etc. it is not surprising to find that fatigue cracks in structural parts usually start at such geometrical irregularities. One of the most effective ways of minimising fatigue failure is by the reduction of avoidable stress raisers through careful design and the prevention of accidental stress raisers by careful processing and fabrication. While this section is concerned with stress concentrations resulting from geometrical discontinuities, stress concentration can also arise from surface roughness and metallurgical stress raisers such as porosity, inclusions, local overheating in grinding and decarburisation, etc. as appropriate to the construction material.

2.5.3 The effect of stress raisers on fatigue under uniaxial loading is that;

1. there is an increase or concentration of stress at the root of the notch,

2. a stress gradient is set up from the root of the notch,

3. a triaxial state of stress is produced.

2.5.4 The ratio of the maximum stress to the nominal stress is the Stress Concentration Factor.

2.5.5 Values of the stress concentration factor will vary depending upon:

1. the severity of the notch,

2. the type of notch,

3. the material,

4. the type of loading, and

5. the stress level.

2.6.1 Abrupt changes in stiffness of the structure should be avoided as they can induce local stress concentrations and reductions of fatigue life.

2.7.1 If possible, precautions should be taken in the design against the possibility of excessive structural vibration being induced, for example, by machinery. This would entail investigation of the natural frequencies of the panel members and of the sources of excitation.

2.8.1 The potential modes of fatigue failure are dependent upon the direction of the applied stress relative to the position of the weld and the position of stress concentrations due to structural discontinuities.

2.8.2 For longitudinal butt welds in plates, dressed flush, and lying parallel to the direction of applied stress, the initiation of potential fatigue failures is expected to be found at weld defect locations. In the ‘as-welded’ condition, fatigue cracks may be initiated at the weld start-stop positions or, weld surface ripples.

2.8.3 For transverse butt welds in plates, essentially perpendicular to the direction of applied stress, the fatigue strength depends largely upon the shape of the weld profile. Fatigue cracks normally initiate at the weld toe.

2.8.4 Cruciform fillet weld joints associated with the four way connection of plate or stiffeners, may be separated into two distinct types depending on whether or not the fillet weld transmits direct load i.e. non-load carrying or load carrying cruciform joints. In the case of the non-load carrying cruciform joint, the fatigue crack will initiate at the weld toe and propagate through the thickness of the load bearing plate in a plane perpendicular to the direction of the applied stress.

2.8.5 In load carrying cruciform joints, in addition to the weld toe, acute stress concentration occurs at the root of the fillet weld and generally fatigue cracks are initiated at the root of the weld and propagate through the weld throat. The fatigue life of such connections can be improved either by increasing the throat size of the fillet weld or by requiring improved weld penetration. In high stress regions however, such measures may not be adequate and there is then a need to specify a full penetration weld in order to achieve the necessary fatigue life for the joint.

2.8.6 Tee joints, since they represent a semi-cruciform joint in a three way connection of plates or stiffeners, would be expected to demonstrate similar fatigue characteristics to the load bearing cruciform joint. However, if bending stresses are induced in the base plate material of the tee, which are of a similar or greater magnitude than the direct stress in the tee, then a fatigue crack may initiate in the base plate at the toe of the fillet weld and propagate through the base plate.

2.8.7 Where tee or cruciform connections employ full penetration welds, and the plate material is subject to significant strains in a direction perpendicular to the rolled surfaces, it is recommended that consideration be given to the use of special plate material with specific through thickness properties, as detailed in Ch 3, 8 Plates with specified through thickness properties of the Rules for Manufacture, Testing and Certification of Materials (hereinafter referred to as the Rules for Materials).

2.8.8 For welded stiffeners and girders, fatigue cracks can be expected to be initiated at weld toes and may be associated with local stress concentrations at the weld ends of connecting end brackets or stiffeners.

2.8.9 The most common sites for potential fatigue cracks therefore are:

1. Toes and roots of fusion welds.

2. Machined corners.

3. Drilled holes, cut-outs or other openings.

2.8.10 The main conditions affecting fatigue performance are:

1. High ratios of dynamic to static loads.

2. Loading frequency.

3. Material selection.

4. Welding.

5. Complexity of joint detail.

6. Environment.

2.8.11 For craft operating for long periods in low air temperatures, or high temperatures for composites, the material of exposed structures will need to be specially considered.

2.9.1 Some commonly used weld details have low fatigue strength. This applies not only to joints between members, but also to any attachment to a loaded member, whether or not the resulting connections are considered to be structural.

2.9.2 The heat-affected zone (HAZ) is of great importance to the fatigue strength of welds because this is usually the region where a fatigue crack will develop. Moreover, when the reinforcement of a butt weld is not removed, or when fillet welds are used, a resulting sudden change of section occurs, and stress concentrations occur at the weld toe.

2.9.3 For the specification of welding and structural details, see Pt 6, Ch 2 Construction Procedures and Pt 7, Ch 2 Construction Procedures of the Rules for Special Service Craft for steel and aluminium alloy craft respectively.

2.10.1 A material's fatigue characteristics are fatigue strength and fatigue limit.

2.10.2 The fatigue strength is the stress value beyond which the material will fail at a specified number of stress cycles.

2.10.3 The fatigue limit is the fatigue strength corresponding to an infinite number of stress cycles.

2.10.4 The S-N curve represents the dependence of the life of the ‘specimen’ in a number of cycles, N, to the maximum applied stress, S. N is usually taken (unless specified otherwise) as the number of stress cycles to cause a complete fracture in the ‘specimen’.

2.10.5 Usually no distinction is made between the number of cycles to initiate a crack and the number of cycles to propagate the crack completely through the specimen, although it can be appreciated that the number of cycles for crack propagation will vary with the dimensions of the specimen. Fatigue tests for high cycle fatigue are usually carried out for 10⁵ – 10⁷ cycles and sometimes to 5 x 10⁸ cycles for non-ferrous metals. For a few important engineering materials such as steel and titanium, the S-N curve becomes horizontal at a certain limiting stress. Below this limiting stress, which is called the fatigue limit, or endurance limit, the material can presumably endure an infinite number of cycles without failure.

2.11.1 Complex joints frequently lead to high stress concentrations due to local variations on stiffness and load path. Whilst these may have little effect on the ultimate static capacity of the joint they can have a severe effect on fatigue resistance.

2.11.2 If fatigue control is the dominant criteria, the member cross-sectional shape should be selected to ensure smoothness and simplicity of joint design, so that stresses can be calculated and adequate standards of fabrication and inspection assured.

2.11.3 The best fatigue behaviour will be obtained by ensuring that the structure is detailed and constructed so that stress concentrations are kept to a minimum and that, where possible, the elements may deform without introducing secondary deformations and stresses due to local restraints.

2.11.4 Stresses may be reduced by increasing the thickness of the parent metal and this would theoretically increase fatigue life due to a reduction of the nominal stresses. However, it should be borne in mind that fatigue strength decreases, in general, with increasing thickness.

2.12.1 Since fatigue failure is dependent on the condition of the surface, anything that changes the fatigue strength of the surface material will greatly alter the fatigue properties.

2.12.2 As an example most mechanically finished metallic parts have a shallow surface layer in residual compression. Aside from the effect on surface roughness, the final surface finishing process will be beneficial to fatigue when it increases the depth and intensity of the compressively stressed layer and detrimental when it decreases or removes this desirable layer. Thus sandblasting, glass bead peening, burnishing, and other similar operations generally improve fatigue properties.

2.13.1 Residual stresses arise when plastic deformation is not uniform throughout the entire cross section of the detail being deformed. They therefore comprise a system of internal stresses in the material balanced within the material itself and can exist in the absence of any external loading. Thus if there is an area of tensile residual stress in any cross section at one part of a material there must be a residual compressive stress at some other point. There would in addition be a variation of stress through the thickness of the material, particularly for thicker sections.

2.13.2 In a welded joint residual stresses are induced as a consequence of local heating and cooling cycles associated with the welding procedure and in particular the shrinkage of the weld metal. The actual situation in a welded joint is complicated by practical factors such as the type and size of joint, the welding process used and the weld procedure. In a butt weld for example, high residual tensile stress will exist in the direction of the weld and at right angles to it. In the case of multi-pass or high energy welding these residual stresses may reach the level of the yield strength of the material. As such tensile residual stresses can occur in locations where fatigue cracks are likely to initiate, it will be appreciated that they can lead to a proportional reduction in the fatigue strength of a joint when it is subjected to additional dynamic tensile loads.

2.14.1 The formation of a favourable compressive residual-stress pattern at the surface is probably the most effective method of increasing fatigue performance. As indicated in Ch 1, 2.13 Residual stress, residual stresses are locked-in stresses which are present in a part which is not subjected to an external force. Only macro-stresses, which act over regions which are large compared with the grain size, are considered here.

2.14.2 In general, for a situation where part of the cross section is deformed plastically while the rest undergoes elastic deformation, the region which was plastically deformed in tension will have a compressive residual stress after unloading, while the region which was deformed plastically in compression will have a tensile residual stress when the external force is removed. The maximum value of residual stress which can be produced is equal to the elastic limit of the metal.

2.14.3 The high compressive residual stresses at the surface must be balanced by tensile residual stresses over the interior of the cross section.

2.14.4 The improvement in fatigue performance, which results from the introduction of surface compressive residual stress, will be greater when the loading is one in which a stress gradient exists than the case when no stress gradient is present.

2.14.5 It is important to recognise that improvements in fatigue properties do not automatically result from the use of shot peening or surface rolling. It is possible to damage the surface by excessive peening or rolling.

2.14.6 In order for the desirable effect of surface cold working to be maintained, the cold-working process must be accomplished in the final heat-treated condition and subsequent thermal treatment eliminated when feasible and closely controlled when essential. Exposure of cold-worked surfaces to elevated temperature initially results in stress relief of the plastically deformed zone and ultimately in recovery or perhaps re-crystallisation of the work-hardened area, with complete loss of the desirable residual stress gradient.

2.15.1 There are some processes that are capable of developing high localised surface temperatures which are often difficult to detect and occasionally are responsible for a failure in service. Grinding can be one of these processes.

2.15.2 The rapid quenching of the material immediately below the grinding wheel by the large mass of cold metal can produce cracks or ‘check’. High strength steels (for which grinding is most often used) are particularly sensitive to grinding techniques.