A study on effectiveness limitations of high-frequency mechanical impact

Research output: ThesisDoctoral ThesisMonograph


  • Eeva Mikkola

Research units


High-frequency mechanical impact (HFMI) has recently emerged as an efficient and user-friendly method for improving the fatigue strength of welded steel structures. Further benefit from the treatment can be obtained by using high-strength steels, as the level of improvement depends on steel strength. As a result, HFMI offers potential for high-strength lightweight design of welded structures. This is of interest in industries like shipbuilding and bridge construction, where welding is a widely used joining method. However, for standardizing HFMI as a weld toe improvement method, the limitations of the technology need to be understood. HFMI-treatment introduces compressive residual stresses at the weld toe, improves the local geometry and strain hardens the treated surface region. However, the effectiveness of the treatment is considered to rely on the existence of compressive residual stresses. This means that high mean stresses and high peak loads during variable amplitude loading, which might relax these stresses, can limit the benefit from HFMI. Therefore, this work focuses on the effectiveness limitations of HFMI under different loading histories. The aim is to understand the influence of different loading histories on fatigue improvement and to evaluate proposed fatigue assessment guidelines for HFMI-treated joints. First, available fatigue data was analysed statistically to investigate the effects of high mean stresses and variable amplitude loading with respect to expected fatigue strength improvement. To study the behaviour of residual stresses further, the local material behaviour at the weld toe needs to be known. This was determined by strain-controlled fatigue tests of thin HFMI-treated steel sheet specimens. The resulting data was used in simulations of local stress-strain response at an HFMI-treated transverse attachment to investigate the effects of stress ratio and peak loads on fatigue damage. Finally, allowable peak stresses in HFMI-treated welded joints were discussed based on the numerical results and available experimental data. The statistical analysis showed that the proposed fatigue assessment guidelines fit the current high mean stress and variable amplitude loading data. However, in some cases the proposed allowable stress limits were too conservative. This occurred even when residual stress relaxation was expected based on previous measurements and the current numerical results. The simulations indicated fatigue improvement due to weld toe geometry improvement and strain hardening as the reason. A maximum stress range of 1.2 times yield strength for negative stress ratios and maximum stress ratio of 0.7 were suggested as new design limits. However, further experimental and numerical work is recommended to confirm these limits.


Original languageEnglish
QualificationDoctor's degree
Awarding Institution
  • Aalto University
Print ISBNs978-952-60-6785-8
Electronic ISBNs978-952-60-6786-5
Publication statusPublished - 2016
MoE publication typeG4 Doctoral dissertation (monograph)

    Research areas

  • high-frequency mechanical impact (HFMI), fatigue improvement, high-strength steel, residual stress, strain hardening, notch stress

ID: 18879141