Aircraft Technologies STRUCTURES

BOJCAS: Bolted Joints in Composite Aircraft

Structures
Michael McCARTHY

The objective of BOJCAS is to develop advanced numerical design methods for bolted
joints in composite aircraft structures. This is a critical technology supporting the introduction
of composites into the primary structure of large commercial aircraft. The methods
developed have the potential to significantly reduce testing, and hence time/cost
of development, as well as aircraft weight with consequent increase in efficiency. They will
also help to ensure continued safety. This article provides an overview of activities within
the project.

he use of composite materials in joints. Because joints represent potential had limited success in predicting joint

T aircraft structural components

has grown steadily with each
generation of aircraft. From ini-
tial applications in non-struc-
tural parts and secondary structures,
composites have increasingly found use
in primary aircraft structures, particu-
weak points in the structure, the design
of the overall structure tends to follow
from, and be significantly limited by, the
design of the joint. Non-optimal design
of joints can lead to overweight struc-
tures, in-service structural problems
and high life-cycle costs. This is even
behaviour. Consequently, the current
design methods used in industry are
largely empirical and heavily reliant on
expensive and time-consuming testing.
Many of the methods have advanced lit-
tle from those developed during an
intense period of testing in the USA in
larly in light aircraft, commuter planes, more pronounced in composite struc- the 70s and 80s. Their application to
military fighters and helicopters. To tures, since maximum joint efficiencies new, primary structures of commercial
date, their use in the primary structure are at best 40-50%, and at worst consid- aircraft, with increased uncertainties
of commercial aircraft has been relative- erably less. This compares with 70-80% due to new materials and thicker lami-
ly limited. However intensive efforts are for metals and thus detracts from the nates, and increased quantities of mate-
currently taking place on future com- weight advantage of composites over rial used in each test, is likely to lead to
posite wing and fuselage structures, metals. Hence optimising joint efficien-
expensive design cycles and overweight
and the use of composites in primary cy is crucial to realising the maximum
joint designs. With recent developments
structures is likely to increase substan- potential benefits of composites.
in computational mechanics and contin-
tially over the coming decade. Some of the reasons for lower joint effi-
ued increase in processing power, there
Such developments are being driven by ciency in composites are: brittleness
the potential benefits of composites, which means little stress relief around is the potential to develop more
chiefly in relation to reduced weight the highest loaded holes, anisotropy advanced analysis tools which could be
and operating cost. However realising which leads to higher stress concentra- used to optimise joint design, reduce the
the full value of this potential still tion factors, low transverse strength, quantity of experimental tests required
involves many technical challenges. To susceptibility to delamination, and sen- in development, and improve funda-
ensure the future competitiveness of the sitivity to environmental conditions. All mental understanding of joint behav-
European aerospace industry, it is criti- of these factors together with the com- iour, hence ensuring continued safety.
cal that the maximum possible benefits plexity of composite failure modes,
are obtained to meet the challenge from make the analysis and design of com- Project objectives
similar developments elsewhere. It is posite joints far more complex than that
also essential to maintain and improve of metallic joints. Much effort has been The overall objectives are:
current levels of safety. To achieve these put into developing analytical design • reliable and user-friendly analysis-
goals the knowledge base of composite methods for composite bolted joints, based design methods, with improv-
structures behaviour needs to be using both closed-form analytical meth- ed predictive capability which will
extended and advanced design tools ods and numerical techniques such as enable:
need to be developed. the finite element method. However, the (a) a significant reduction in testing,
BOJCAS addresses a critical aspect of majority of models to date have been and hence time and cost of devel-
this challenge, namely composite bolted overly simplistic in nature, and have opment, and

1 AIR & SPACE EUROPE • VOL. 3 • No 3/4 - 2001

BOJCAS

(b) the incorporation of composites Table I. The BOJCAS Partnership.

into the primary structure with
optimal weight savings. Ireland University of Limerick (Coordinator)
• a fundamental improvement in un- United Kingdom Airbus UK, DERA
derstanding of composite bolted joint Germany EADS Airbus
behaviour, especially in primary Sweden SAAB AB, FOI, Royal Institute of Technology
structures, thus contributing to con- Italy CIRA
tinued safety. The Netherlands NLR
Greece ISTRAM
The partnership Switzerland SMR
The consortium consists of three aircraft
manufacturers, four national aerospace
laboratories, two universities, and two a baseline (Tasks 2.1 and 2.2). These bolted composite repairs, which are
research companies. Eight countries are methods include handbook/design essentially complex multi-fastener
represented as shown in table I. chart methods and two-dimensional joints. Improved analysis of repairs will
The start date was February 2000 and finite element methods. Then new glob- considerably reduce testing needed for
the duration of the project is 36 months. al methods will be developed mostly the certification of repair configurations
based on two-dimensional finite ele- and procedures given as standards
ment methods with specialised tech- within the Structural Repair Manual.
Programme content niques to model bolt-hole interaction, Other benchmark structures are aimed
The programme structure is illustrated and validated on the benchmarks at studying the effects of variable bolt
in figure 1. Bearing in mind the needs of (Task 2.3). Figure 2 illustrates one of the patterns, as well as damage tolerance.
industry for preliminary and detailed benchmark structures. This benchmark For comparison with the global meth-
design tools, the following outputs are structure will consist of several varia- ods, the benchmark structures will also
planned: tions on a skin-stringer joint element for be modelled using global-local meth-
• global design methods, for prelimi- a potential hybrid metal/composite ods. These methods are being devel-
nary design of complex, multi-fasten- wing. The structures will be relevant to oped to automatically couple global
er joints; the design of the lateral wingbox in the models with much more detailed local
• detailed design methods for final EU project TANGO and will also models of the bolt regions. More
design of critical joints; address the issue of metal/composite detailed models are needed because
• methods to couple global and detail- joints in a generic way. Another bench- several effects influencing failure are
ed design methods, i.e. to streamline mark structure will be representative of three-dimensional in nature, and cannot
the process of producing a detailed
analysis from a preliminary analysis;
• design guidelines for primary com-
posite bolted joints based on analyses
and tests.
The project is divided into a global
strand (WP 1, 2 and 3) and a local strand
(WP 4 and 5) with the coupled global-
local methods bridging the two strands.
Interaction takes place between the
strands by using the knowledge gained
from the detailed local methods to
improve the global methods. Each
strand contains major testing and analy-
sis components.
At the global level, a series of ‘bench-
mark’ structures representative of com-
plex, primary, multi-fastener joint con-
figurations, will be designed and tested.
Global design techniques will be used to
design and predict the performance of
these benchmarks. Initially, existing in-
house methods will be used to provide Figure 1. BOJCAS programme structure.

2
Aircraft Technologies STRUCTURES

will be processor speed, which can be

expected to be removed within just a
few years. Once validated these models
can also be used to generate design data
for use in preliminary design, with con-
siderably less experimental tests than
are required at present.
In WP 5, an experimental test programme
will be carried out involving smaller-
scale joints than the benchmarks. These
tests will provide further data related to
some of the issues covered by the bench-
mark structures (e.g. composite-metal
joints, bolted repairs), as well as provid-
ing data for validation of the detailed
models. An extensive list of parameters is
Figure 2. Example benchmark structure: skin-stringer joint element
being examined, including variations in
for hybrid metal/composite wing. (Doc. Airbus UK) geometry, loading, materials, lay-ups,
bolt-types, environmental conditions,
bolt-hole clearances, clamping force and
be accounted for by two-dimensional not currently suitable for use in the pre- others. Tests will be extensively instru-
techniques. For example, non-uniform liminary design phase. Work in BOJCAS mented using techniques such as strain
through-thickness stress distributions is aimed at automating the setup pro- gauging, photoelasticity and intrument-
exist in situations involving counter- cess as far as possible, so that the only ed bolts, and detailed fractographic fail-
sunk bolts, non-symmetrical loading, barrier to exploiting these methods fully ure analysis will be performed (figure 5).
bolt bending, or bolt tilting in holes with
clearance, and lead to significant stress
concentrations. This will particularly be
the case with thick primary structures.
The ‘bearing’ mode of failure (in which
the laminate is locally crushed at the
hole) has been shown to be a three-
dimensional phenomenon, involving
through-thickness cracks and delamina-
tions. Such comparisons will also enable Figure 3. Three-dimensional stress distributions in single-lap joint. (Doc. ULIM)
the improvement of global methods via
improved spring stiffnesses and correc-
tion factors for three-dimensional
effects.
At the local level, work is focusing on
the development and validation of
detailed joint models incorporating new
means of determining failure. Figure 3
illustrates the localised nature of the
stress distributions in single-lap joints,
which cannot be accounted for with
two-dimensional methods. Figure 4
illustrates the use of progressive dam-
age modelling to track the progression
of failure in each ply until final failure of
the joint. Such techniques possess the
potential for more accurate failure pre-
diction than the essentially empirical
failure criteria currently in use.
Such detailed models take considerable Figure 4. Progressive damage propagation at different load levels
time to set up and run, and as such are (upper surface of a [90/0-45/45]S8 laminate). (Doc. ISTRAM)

3 AIR & SPACE EUROPE • VOL. 3 • No 3/4 - 2001

BOJCAS

ites in aircraft structural components.

Since joints have such a critical effect on
the safety and efficiency of aircraft
structures, it is vital that the most
advanced design methods are used. A
high potential exists to reduce develop-
ment costs, maximise weight savings,
increase manufacturer and operator
confidence in composites, and ensure
safety of future primary composite
structures. The spin-off potential of the
developed technology is very high in
several other fields such as shipbuild-
ing, space, nuclear, chemical, offshore,
automotive, rail and civil engineering.
Several of the partners in BOJCAS have
been involved for many years in nation-
al programmes on composite bolted
joints. BOJCAS will pool this collective
expertise and provide a European per-
spective on this important topic. ■
Figure 5. Analysis of damage progression using microscopy (FOI).

Finally evaluation and summary tasks ned and predictions have been made
will take place. The global methods are regarding load distributions and failure
aimed at immediate exploitation, and as loads. Fabrication has begun and tests
such will be implemented into the will be complete by the mid-term
industrial partners’ codes of choice and assessment. The specimen tests in WP 5
assessed by those partners. The detailed are under way and will also be complete
methods will be assessed for their by mid-term. Modelling of the bench-
exploitation potential and a path mark structures with global methods
towards implementation will be drawn has been performed, and initial global-
up. Overall, the results from the tests local models will shortly be ready for
and analyses will be used to form comparison. Interim reports on the About the authors:
design guidelines for composite bolted detailed models of the specimen tests Michael McCarthy is Director, Composites
joints. will be supplied at the end of Month 12. Research Centre; and Lecturer,
Department of Mechanical
Current status Conclusions and Aeronautical Engineering, University
of Limerick, Ireland.
In the first year of the project, all the BOJCAS is focused on an enabling tech- [email protected]
benchmark structures have been desig- nology for the increased use of compos-