Assessment of residual stress using thermoelastic stress analysis

Abstract

The work described in this thesis considers the application of thermoelastic stress analysis (TSA) to the
assessment of residual stresses in metallic materials. Residual stresses exist within almost all engineering
components and structures. They are an unavoidable consequence of manufacturing processes and may
cause the premature and catastrophic failure of a component when coupled with in-service stresses.
Alternatively, beneficial residual stress may be introduced to enhance the component performance.
Greater knowledge of residual stress and its evolution, will not only provide an opportunity to improve
component manufacture and design, but may allow the potential life extension of current structures
beyond their design life.

Techniques for measuring residual stresses can be divided into two main groups. Destructive
methods involve removing material, measuring the mechanical strain relaxation and back calculating the
residual stress. These techniques are generally cheaper and more portable, but are not appropriate in
many circumstances due to modification and damage to the component. Non-destructive techniques do
not damage the component, but are typically more expensive, less portable and can require complicated
calibration procedures to correctly interpret results.

TSA is a well established non-contacting experimental stress analysis technique that is quick and
portable, and the presence of residual stress is known to modify the thermoelastic response. However,
this change is very small and of the order of the noise threshold and resolution of currently available
infra-red detectors. Several methods for identifying residual stresses from the thermoelastic response
have been suggested and are further explored in this thesis. Significant attention is given to the effect of
plastic deformation on the thermoelastic constant, and the influence of the mean stress on the
thermoelastic response in stainless steel and aluminium.

An investigation of the experimental setup is undertaken to optimise the detector settings, maximise
the thermoelastic signal and minimise measurement errors. For metallic materials, a paint coating is
typically required which may attenuate the response. A study of coating characteristics is presented,
which compares the experimental and theoretical thermoelastic response. The importance of the coating
is highlighted and recommendations for appropriate conditions are provided.

The overall feasibility of applying a TSA based approach to residual stress assessment is considered
by examining residual stresses around cold expanded holes in aluminium plate. Changes in the response
are identified and attributed to the presence of residual stress. Laboratory X-ray diffraction is used to
provide residual stress measurements. These are incorporated into a model of the thermoelastic response
providing good agreement between experimental data and theoretical predictions within the region of
interest. The potential for TSA to identify residual stress is demonstrated, and the study thereby
represents a significant step towards understanding the role of TSA within the field of residual stress.

More information

Identifiers

Catalogue record

Export record