Assessment of fatigue damage evolution in woven composite materials using
infra-red techniques
R.K. Fruehmann, J.M. Dulieu-Barton
*
, S. Quinn
School of Engineering Sciences, University of Southampton, Southampton, SO17 1BJ, UK
article info
Article history:
Received 20 August 2009
Received in revised form 30 January 2010
Accepted 7 February 2010
Available online 12 February 2010
Keywords:
A. Textile composites
B. Fatigue
C. Stress concentrations
Thermoelastic stress analysis (TSA)
abstract
Thermoelastic stress analysis (TSA) is used to study the growth of fatigue damage in single and two ply,
2 Â 2 twill woven composite materials. Test specimens were subjected to a uniaxial tensile cyclic loading
with maximum stresses of 10%, 15% and 20% of the ultimate failure stress. The development of fatigue
damage locally within the weft yarns is monitored using high resolution TSA. The specimens were sub-
sequently inspected using optical microscopy to evaluate the location and extent of cracks. Cracks were
found in the weft fibres, running transverse to the loading direction. It is demonstrated that the lighter
weight fabric is more resilient to damage progression. A signature pattern is identified in the TSA phase
data that indicates the onset and presence of fatigue damage in the composite material.
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Fibre-reinforced polymer (FRP) composite materials with textile
reinforcements have widespread use in both component manufac-
ture and repair of damaged structures [1]. Their popularity is par-
tially due to ease of handling, ability to drape over complex
curvatures, low cost and availability, combined with the typical
composite material features of good mechanical performance in
terms of the strength and stiffness to weight ratios. Twill woven
materials are often used in the surface ply to provide a good sur-
face finish. The use of a textile surface ply, however, provides a
challenge for any surface sensing, experimental stress analysis tool
due to the heterogeneity of the material surface. The small scale
heterogeneity leads to local stress/strain concentrations causing
damage evolution, the significance of which is often difficult to as-
sess [2]. Furthermore, where experimental techniques are used in
conjunction with FEA, homogenisation of the textile architecture
to reduce the cost of numerical simulation means that such small
scale effects are typically ignored, making a comparison between
measured and calculated data difficult.
The current study investigates the use of thermoelastic stress
analysis (TSA) for fatigue assessment in woven FRP composites un-
der tensile loading. TSA is a non-contacting, full-field stress analy-
sis technique with well documented application to the study of
stresses in composite materials [3,4]. The technique is based on
the ‘thermoelastic effect’ where a small temperature change occurs
as a result of a change in stress. The temperature change is ob-
tained by means of a sensitive infra-red detector. It has been
shown that the thermoelastic response (i.e. the output from the in-
fra-red detector) from an orthotropic material such as a FRP com-
posite is dependent on the orientation of the fibres relative to a set
of orthogonal reference axes, which are usually the principal mate-
rial axes or stress axes [5]. Previous work has been conducted using
TSA to study fatigue damage development in laminated non-crimp
FRP composites [6]. The work utilised a global approach to damage
quantification. However, in composites with woven fibre reinforce-
ments, the weave pattern results in two orthogonal fibre orienta-
tions at the surface of the material, with axes that may or may
not align with the reference axes. Moreover, because of the inter-
lacing pattern of woven yarns, resin pockets are formed at the
weave junctions. As a result the thermoelastic response will vary
locally, and even a uniform strain field, as may be expected in sim-
ple uniaxial tension, will result in a non-uniform thermoelastic re-
sponse [7]. However, it is known that even under simple loading
conditions, the strains in a woven composite are not uniform [8].
Hence, the interpretation of TSA data from woven composite mate-
rials which results from both material and strain variations poses a
significant challenge. Therefore, in the current work, both global
and local approaches are employed to evaluate the thermoelastic
response.
Previous studies on the application of full-field strain analysis
techniques to woven composites such as digital image correlation
or interferometry are few [8–10]. These techniques use quasi-static
loading and may require some surface preparation such as the
application of a grating or painted speckle pattern. In TSA, mea-
surements can be obtained from a dynamically loaded component
in practically real time (less than 5 s per measurement) with
0266-3538/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compscitech.2010.02.009
* Corresponding author. Tel.: +44 2380 596522; fax: +44 2380 593299.
E-mail address: janice@soton.ac.uk (J.M. Dulieu-Barton).
Composites Science and Technology 70 (2010) 937–946
Contents lists available at ScienceDirect
Composites Science and Technology
journal homepage: www.elsevier.com/locate/compscitech