Allowing for Residual Stresses in Design

Engineering Integrity Society


Wednesday, 3 May 2006

MIRA, Nuneaton

ABSTRACTS

'Sources of residual stress'
Frank Sherratt - Engineering Consultant

Residual stresses are created by differential cooling, as in the welding process, chemical expansion, as in nitriding, or by elastic-plastic deformation. After briefly mentioning the first two, the presentation will show how elastic-plastic deformation always creates bodies with residual stress. Inducing this deformation by treatments like cold-rolling and shot peening will be described, and ways of estimating the magnitude and depth of the stresses caused by these processes will be covered.

‘Stress intensity factor weight functions for dealing with residual stresses in LEFM’
Feargal Brennan, UCL

The advantages of expressing stress intensity factors in terms of geometry weight functions are widely appreciated. These represent the influence of a component and crack geometry on the stress intensity factor, independent of applied loading. Weight function solutions have been available for a number of geometric shapes since the mid 1970s. However, the real potential of the weight function approach has not been realised to date largely because of the cumbersome mathematics traditionally associated with the weight function approach.

This presentation explains the basis of the SIF Weight Function, presenting a simple and robust method for determination of a weight function from reference SIF solutions and the subsequent use for calculation of SIFs through residual stress fields. The use of so called "simplifying approximations" in weight function mathematics is also considered in the context of loss of accuracy and versatility.

'Inducing “designer” residual compressive and tensile stress'
Peter O'Hara - Vice President, Metal Improvement Company

Controlled Shot Peening has been used by many industries to prevent premature failure and by some as a design tool to enable higher load carrying capacity. However, the process has not been utilized as effectively as it could because of the concern at reproducibility on a repetitive basis. The introduction of Laser Peening answers those concerns. In the past two years Laser Peening has been in production in the UK and USA and to date over 14,000 components have been processed and the technique of Laser Peening established as a reliable, production proven process offering quality control of the end product, the residual compressive stress, to a level unknown previously.

Laser Peening offers all markets, substantial benefits both technically and commercially. Presently we are working on residual stress solutions to “engineer” a specific residual stress profile to suit desired performance. Laser Peening has the capability of tailoring the stress profile with compressive residual stresses up to 10 mm deep, with virtually no roughening and little cold work. Designers are asking for a residual stress profile through the product and we are able to adjust depth, magnitude and slope as required.

Laser Peening was invented in the 1970’s, and it uses short bursts of intense laser light to create pressure pulses on the metal surface, which generate shock waves that travel into the metal and compress it. MIC is using a Lawrence Livermore National Laboratory developed neodymium-doped glass laser that produces one billion watts of peak power, about the output of a large commercial power plant, in 20-billionth of a second bursts. With 125 watts of average power, the laser can fire five pulses per second. Controls in terms of power density, pulse width and coverage enable the subsurface stress profile to be adjusted significantly.

'The Measurement of Residual stress using MAPS'
John McCarthy, ESR Technology

It is accepted that the underlying stress state in a material before the application of load affects the integrity of the material. However most integrity analyses are performed with no knowledge of the underlying stress state because hitherto there has been no means to measure it without recourse to expensive laboratory equipment or a destructive test. ESR Technology's MAPS system offers the ability to measure the residual stress in steel or iron quickly, non-destructively, accurately and in an industrial environment. This paper will discuss the MAPS system and provide some examples of its use.

'The unmeasured strains: the effect of residual strain on product life and manufacture'
Peter Blackmore, Jaguar-LandRover Chassis Engineering, Norman Thornton, nCode International and Angelo Fanourakis, GKN AutoStructures

Residual strains can be produced whenever a material or component undergoes non-uniform plastic deformation. Mostly, the presence of residual strains is ignored in both CAE calculations and physical tests on components but they can have significant effects on their performance and will contribute, perhaps very significantly, towards the variability seen in component durability. The origin of residual strains will be described and case studies presented that demonstrate their effects on fatigue performance. An overview of some of the main methods used to measure residual strains will also be given.

‘Including forming and assembly stresses in fatigue analysis from FEA’
Ian Mercer & John Draper, Safe Technology

Some manufacturing processes can result in residual stresses and strains in a finished component, which remain in the component throughout the service life. Many manufacturing processes can be simulated with a high degree of confidence as can the in-service loads. The transfer of the residual condition to initialise the in-service loading analysis is not performed routinely. This presentation investigates the effects of the formed state of a deep drawn sheet metal oil pan component on the engine vibration performance. A method for simulating and transferring the effects of the manufacturing process to an engine vibration assessment is discussed. Results from fatigue analyses demonstrate the significant effect of incorporating the manufacturing process.

'A surprising result of shot-peening a notched aluminium specimen'
Frank Sherratt - Engineering Consultant


Tests on rotating-bend specimens of shot peened 7075-ST showed that in order to achieve the highest fatigue strength, based on nominal stress, a notch had to be introduced. A study of crack propagation modes is proposed as an explanation.

‘Predicting Fatigue Crack Growth From Cold Worked Holes In Aerospace Alloys’
David Smith and Martyn Pavier, University of Bristol

Our presentation will be in two parts. The first part on predicting fatigue crack growth from a cold expanded hole will be presented by David Smith. The second part on including crack closure effects in predictions of fatigue crack growth will be presented by Martyn Pavier.

Fastener holes in aircraft components are routinely cold expanded to improve fatigue life, although the precise level of improvement is difficult to quantify. The first part of this presentation will describe an attempt to predict the growth rate of a fatigue corner crack emanating from the bore of a cold expanded hole. The prediction was made using a three-dimensional finite element model of a cold expanded hole combined with a Paris law fit to fatigue data derived from tests for corner cracks emanating from a plain hole. Residual stresses in the finite element model were calculated from a simulation of the cold expansion process. Finite element predictions of fatigue crack growth rate will be compared with experimental measurements giving adequate agreement.

Plasticity induced crack closure is an important mechanism that needs to be included in accurate predictions of fatigue crack growth rate in aluminium alloys. The second part of the presentation will address predictions of fatigue crack growth rates for corner cracks emanating from a hole under constant amplitude loading and a combination of constant amplitude and a single overload.

‘Crack shape control through ‘designer’ residual stresses’
Feargal Brennan, UCL

This presentation reports an experimental study of crack shape evolution in mild steel specimens under cyclic loading. It is widely known that the introduction of compressive residual stresses by cold working the surface can be highly beneficial in improving the fatigue performance of structural components. Although it is recognised that relaxation of surface compressive residual stress can reduce the potential benefits, the effects of residual stress on crack shape evolution are often overlooked. Previous studies have shown that the intensity of the surface compressive residual stress has a pronounced effect on fatigue crack initiation. A recently developed technique which applies differing intensities of compressive residual stress at specific regions in a structure, is shown to influence fatigue crack propagation life considerably.

The results of five fatigue tests under bending load in different thickness plates are presented. Crack growth retardation is apparent in all tests and this is attributed to the constrained crack shape. Rerolling after crack initiation and number of rolling passes are also considered. The fatigue crack propagation path is shown to have been altered significantly by the residual stress and has promoted localised failures in the fatigue specimens. The results suggest the method might be efficiently utilised to implement the leak before break design philosophy but not under bending.

‘The Treatment of Residual Stresses in BS7910’
J K Sharples, Serco Assurance, Birchwood Park, Risley, Warrington, WA3 6GA, UK

BS7910 is a Standard for Engineering Critical Assessments (sometimes referred to as Fitness for Service Assessments or Fitness for Purpose Assessments) to be performed by structural integrity engineers or other relevant competent persons. Such assessments are required in order to assess the significance of defects (flaws or cracks) that may have been detected in an operating structural component by Non Destructive Testing Techniques.

Many structural components (e.g. pressure vessels and piping) contain weldments, and for such components, weld residual stresses are required to be taken into account in Engineering Critical Assessments since they can have a significant influence on structural integrity.

The presentation briefly deals with BS7910 concepts in terms of the Failure Assessment Diagram that is used to evaluate limiting defect sizes in structures and components. The treatment of residual stresses in the code is then dealt with in some detail both in terms of how residual stresses (and secondary stresses in general) are incorporated into the methodology and the guidance provided on the magnitude and distribution of such stresses. The presentation concludes by providing an overview of ongoing developments that will result in improvements being made to how residual stresses are treated in future versions of BS7910.

‘Defect assessment and repair case study’
P E J Flewitt - British Nuclear Group and University of Bristol

A brief description of the importance of residual stresses in engineering components and structures will be presented together with their accommodation in failure avoidance procedures such as R6. An example of the measurement, using x-ray and neutron diffraction, of near and through section stresses developed in a ferritic steel plate subject to special cooling conditions will be described. The experimentally measured stresses will be discussed by comparison to those obtained using finite element modelling. A case history of a repair weld undertaken on a major electrical power generation boiler will be described with emphasis on the contribution of the residual stresses. Finally a methodology for underwriting repair welds made in high integrity structures and components will be described.

 

Home: http://www.e-i-s.org.uk