1

Editorial overview

Steel construction
R Zandonini
University of Trento, Italy

The importance of fire design in the construction

related world is historically fairly high. Far more
recent are the attempts to develop an adequate
philosophy of design, capable of dealing rationally
with the different aspects involved in the assessment
of the reliability of a building system with respect to
the fire event. This was made possible by the growth
of consistent knowledge in the different areas related
to such an event, which are identifying a rather new
branch of engineering: i.e., fire engineering. This
branch is multi-faceted, and implies the use of
multi-disciplinary methods. It is hence quite difficult
to approach it within a Journal, which remains rooted
in the area of structural engineering, despite its
innovative format and scope. The approach adopted
has been a sort of step-by-step path, starting from the
structural performance and moving towards a scene
where structural problems are seen within a broader
perspective. In the very first issue of the Journal a
paper by Moore and Lennon appeared, presenting the
principles of fire-resistant design, and the outcomes of
a series of full-scale fire tests carried out in the United
Kingdom. These tests demonstrated the key features
of the fire response of steel and composite
steel-concrete framing systems. Better performances
than expected were pointed out, indicating the
conservativeness of traditional design methods. In the
following steel construction issue (i.e., issue 1 of the
second volume), a paper by Kruppa was devoted to
the complementary aspect of numerical analysis.
Simulation tools, which are under continuous
refinement, presently allow covering with good
accuracy the various physical phenomena associated
with a fire, such as its development, the temperature
distribution within heated structural elements and the
mechanical behaviour of the structural system at
elevated temperatures. The paper by Bennetts and
Thomas, which appears in this issue, goes a step
further, and intends to contribute to the
understanding of the role of the structural
performance within the fire safety concept. The
introduction enlightens the various fire-safety related
Copyright & 2002 John Wiley & Sons, Ltd.

design objectives, and weights the different hazards

presented by fire, with the structural one being
placed, in many instances, not amongst the heaviest.
This outline leads to identify the risk management
approach as the definite strategy to achieve effective
fire-safety design. Such a global perspective does not
minimize the importance of the structural side of
fire-safety, an area in continuous rapid development.
An update of the advances in understanding and
modelling of the behaviour of steel structures under
fire is then presented, ranging from new mathematical
relations for describing the thermal response of
modern steels to the proposed approaches for the
modelling of the whole structural systems. Particular
attention is given to the performance of steel columns
and composite beams, where numerous studies were
recently carried out.
The two following papers are devoted to
connections. Connections represent a key component
in structural steelwork, which is by its very nature an
assembly of prefabricated elements. The
development of the fastening techniques is strongly
interrelated with the history of steel construction. The
case of the advent of welding can be a clear example
of the dramatic influence of connections on the statics
and aesthetics of steel structures. The importance of
connection fabrication and design increased
significantly, with the competition amongst different
materials, and in particular between steel and
reinforced concrete. In the very first issue of the
Journal a paper by Steenhuis, Stark and Gresnigt
appeared whose subject was Cost-effective connections.
Focus was on the trend to develop fabrication and
erection-friendly connections through a process of
simplification of the detailing, supported by intensive
research work. New connection concepts were also
presented as well as innovative solutions. Besides, the
important topic of the design cost of connections was
covered. In the following Steel Construction issue,
Carter, Murray and Thornton, presenting the North
American advancement on economical steel design,
put forward a similar consideration of the role of
connections in the search for cost-effective steel
buildings. Cost-effectiveness requires that a
satisfactory balance be achieved between the
connection cost and the structural performance. The
Prog. Struct. Engng Mater. 2002; 4:15

EDITORIAL OVERVIEW

concept of semi-rigid joints provided a further

important contribution towards a wider range of
optimal design options. Recent codes, such as the
Eurocodes, incorporated the new philosophy, and
provide an innovative joint classification, based on
stiffness, together with specific design
recommendations for semi-rigid joints. However, an
extensive use in everyday practice of this concept was
prevented by various factors. Among them, the
insufficient knowledge amongst practitioners,
associated with an unclear perception of its inherent
advantages (the additional design burden is indeed
clear), played a substantial role. Numerous studies of
joint behaviour and modelling lead recently to the
development of a powerful method of design, the
so-called component method. The method is fairly
versatile, and, although thought and validated
initially for beam-to-column joints in steel framing
systems, it then was successfully extended to joints in
composite steel-concrete structures, and more recently
to base-plates. A deep understanding and effective
use of the approach would significantly contribute to
bridging the semi-rigid framing philosophy from
research to practice. Jaspart discusses in his paper the
general bases of the method as well as its practical
implementation. The diverse aspects of joint
representation are thoroughly examined, (i.e., joint
characterization, joint classification, joint modelling
and joint idealization), in view of a full appreciation of
the implications of semi-rigid joint design. Different
design strategies are then explored, and the potential
benefits pointed out of a selection and sizing of
connections, which takes advantage from the actual
response of joints. A full calculation procedure is also
provided in accordance with the Eurocode 3.
Reference to the Eurocode is of great usefulness to
show the practicality of the method. However, the
value of the concepts highlighted in the paper is by far
very general, and stresses once again the need of
design innovation in the search for economical, yet
reliable structures. As for all innovative design
methodologies, semi-rigid design is made more
effective when suitable tools are used. The paper
refers in particular to ad hoc developed computer
programs and call the attention once more to the
importance of CAD.

The last paper, by Kurobane, focuses on Connections

in tubular structures. As already stated by the writer
when presenting Packers paper on Tubular
construction (foreword to Issue 1 of vol.2) the
increasing popularity of tubular sections in
architecture-driven projects fostered the substantial
body of knowledge built up in recent years, which
made the area of tubular steel structures to achieve
full maturity. Connections are a key component in
tubular construction, even more than for other
steelwork traditional solutions. Importance and
complexity seems to be the two terms characterizing
them. Their design should hence master the
complexity of the phenomena involved in the
response (yielding, buckling, fracture, section
ovalisation, punching, fatigue . . .), and balance (once
more) reliability with economy of the design process
and of the resulting structure. The paper covers
thoroughly the most recent studies, and provides a
fairly extensive list of references, permitting the
reader to deepen any of the numerous subjects dealt
with. A broad area of tubular connections is covered,
ranging from tube-to-tube connections in 2D and 3D
truss systems to beam-to-hollow section column
connections. Experimental and numerical FE research
work is discussed and new trends in codes and
practice is highlighted. Of particular interest the
development of connection with improved details for
use in multi-storey building in seismic areas. Finally,
some unresolved issues are singled out and put
forward for further studies. The overall impression is
that tubular construction is more and more ready to
accept the challenge coming from the need to fulfil
both architectural and structural constraints: i.e., to
allow pursue of a optimal solution in the 3D space of
aesthetics, safety and economics.

Copyright & 2002 John Wiley & Sons, Ltd.

Riccardo Zandonini PhD

Professor of Structural Steel Design,
Dipartimento di Ingegneria Meccanica e Strutturale,
Universit"a degli Studi di Trento,
Via Mesiano 77,1-38100 Trento, Italy
E-mail: [email protected]

Prog. Struct. Engng Mater. 2002; 4:15

EDITORIAL OVERVIEW
Earthquake engineering and structural
dynamics
A S Elnashai
Imperial College of Science, Technology and Medicine,
London, UK

This issue of JPSM includes three papers on

earthquake engineering, with an emphasis on
practical issues for seismic risk reduction written by
workers with a track record in the subject. The paper
by Jim Beavers, Deputy Director of the Mid-America
Earthquake Center, USA, gives an authoritative
review of US seismic hazard studies and related code
developments, by a man who has been at the forefront
of practical seismic design and code-development for
some 25 years. George Penelis, Professor at the
Aristotle University of Thessaloniki, the author of a
paper on seismic rehabilitation of historical
monuments, has been undertaking such work for the
Greek government for many years and has therefore a
unique insight that has been translated into an
exceptional review paper. The team from Georgia
Institute of Technology, USA, headed by Professors
Jim Craig and Barry Goodno, report on progress and
application examples of response modification
devices for seismic risk reduction where their
experience shows through the insightful text. There
was supposed to be a fourth paper on geotechnical
issues, to give the right balance that I have tried to
maintain, but time limitations did not allow this to be

Copyright & 2002 John Wiley & Sons, Ltd.

the case. We might have two geotechnical papers in

the next earthquake engineering issue, to compensate
for the purely structural nature of this issue.
In selecting the papers, I was constrained by three
considerations. Primarily, I only invite established
experts in the subject matter who also have experience
in application, not only development. Secondly, I
continue to strive to contribute to bridging the gap
between sub-disciplines that contribute to seismic risk
assessment and mitigation. Thirdly, the reliability of
the potential authors in adhering to tight time
schedules is taken into consideration. I have hitherto
been lucky that very few of my authors have
promised and not delivered. We all know how
difficult it is to get sought after experts to commit to
and produce papers to deadlines. I am therefore most
grateful to my three colleagues from Greece and the
USA who have dedicated precious time from their
busy schedule to bring to the PSM readership these
valuable contributions.

Amr S Elnashai PhD

Professor of Earthquake Engineering,
Civil and Environmental Engineering Department,
Imperial College of Science, Technology and Medicine,
Imperial College Road, London SW7 2BU, UK
E-mail: [email protected]

Prog. Struct. Engng Mater. 2002; 4:15

EDITORIAL OVERVIEW

Fatigue and fracture

net-section collapse load at 348C. Therefore, even

large fatigue cracks should not prevent redundant
structures fabricated from high-toughness steels from
yielding and redistributing load. Perhaps the greatest
benefit of high-performance steels is increased
reliability, which may allow more frequent
consideration of efficient but lower-redundancy
structural systems.
Structural engineers involved with buildings are
typically complacent about the possibility of fracture.
However, the fractures that occurred in steel buildings
during the earthquakes in Northridge California in
1994 and in Japan in 1995 were a result of this
complacency. The fracture failure mode was not
explicitly considered. Materials with minimum levels
of fracture toughness were not specified, and the
implications of notches from backing bars and weld
flaws were not considered. As a consequence, when
these structures were eventually loaded, they
fractured in a brittle manner before yielding.
Subsequent to these earthquakes, a great deal of
research has been performed on welded seismic
moment resisting frames.
The third paper in this section, prepared by Lindley,
Bateson, Bannister, and Pike of the Corus Group in the
United Kingdom (formerly British Steel), is a major
contribution to this research particularly aimed at the
potential for fracture or lamellar tearing in thick
jumbo column sections. A series of full-scale tests are
described for evaluating the fracture resistance of
these sections, including cyclically loaded beamcolumn connection tests. The tests show that lamellar
tearing should not be a problem under typical
conditions and that the fracture resistance of the
jumbo column sections was satisfactory. Failures in
these tests occurred in the welds after some level of
plastic strain had occurred. A probabilistic fracture
mechanics assessment is briefly described that was
used to determine a minimum notch toughness
requirement for these column sections.
One interesting finding of these tests was that
fractures still occurred in the beam flange to column
welds, even though notch-tough welding
consumables were used. However, these were not
brittle fractures that occurred at elastic load levels like
the pre-Northridge welded moment frame
connections, but rather fractures that occurred after
some number of cycles at plastic rotations on the
order of 2.0% (often referred to as low-cycle fatigue
failures). It should be noted that further refinements
were made to this connection in research recently
performed for the Federal Emergency Management
Administration (FEMA). Present FEMA guidelines
require the beam web to be welded to the column and
require special weld access hole details. These
improvements have been shown to allow these
welded moment frame connections to achieve even
greater cyclic plastic rotation levels before the onset of
low-cycle fatigue.

R J Dexter
University of Minnesota, USA

Papers published in the three previous volumes of

this journal featuring fatigue and fracture have been
focused primarily on fatigue. As explained in my
overview of these previous volumes, fatigue typically
precedes fracture in dynamically loaded structures
and therefore, from a practical viewpoint in structural
engineering, fatigue is a more important problem.
However, we cannot neglect the possibility of
overload even in statically loaded structures and
therefore we must be equally concerned about
fracture. For example, in the first paper in this
volume, Wojnowski, Domel, Wilkinson and Kenner of
Engineering Systems Inc. (Aurora Illinois) present an
investigation of a bridge collapse. The steel plategirder bridge was under construction when it
collapsed. A detailed investigation is summarized
including documentation of the conditions at the time
of collapse, the recovered components, design review,
and finite-element analyses of the bridge. As it turns
out, the collapse was due to inadequate bracing and
was not attributed to design or materials. However,
this collapse shows that unintended loads may be
applied to any structure and therefore it is important
to have minimum levels of ductility and fracture
resistance.
Setting these minimum levels of fracture toughness
for bridge steel and weld metal has been the subject of
great controversy since the 1970s in the United States.
William Wright, from the Turner-Fairbank Highway
Research Center of the Federal Highway
Administration in McLean, Virginia, explains the
fracture toughness requirements for bridges in the
United States in the second paper in this volume.
Although fracture mechanics test methods are
discussed, the Charpy V-Notch (CVN) test is
relatively easy to perform and therefore it will likely
continue to be the measure of toughness used in steel
specifications. Despite the compromises, these
requirements have served their purpose, i.e. there
have been no catastrophic fracture problems with
bridges since their implementation.
However, many things have changed in the decades
subsequent to the development of these specifications.
Perhaps the most important development is that it is
now possible to purchase steels with very high
fracture toughness with only a slight premium.
William Wright discusses recent developments in the
United States, including the first grade of highperformance steel (HPS) to be fully integrated into
the AASHTO specifications, Grade HPS-485W. This
steel offers upper shelf fracture resistance at service
temperatures. For example, full-scale fracture
experiments show that girders fabricated from HPS485W steel were able to withstand cracks up to 50% of
the tension flange area and still reach at least the
Copyright & 2002 John Wiley & Sons, Ltd.

Prog. Struct. Engng Mater. 2002; 4:15

EDITORIAL OVERVIEW
One of the remaining problems in fracture is the
ability to predict fracture in structural components
using elastic-plastic fracture mechanics parameters
such as the crack-tip opening displacement (CTOD),
which is measured on small three-point bend bars.
Professor Masao Toyoda of Osaka University
addresses this problem in the fourth paper in this
section. In particular, this paper looks at the effects of
mismatch in strength at welded joints on
the fracture mechanics parameters. Alternative
local approaches are described for both ductile and
brittle fracture in welded joints. Finally a procedure
for converting a CTOD in a structural component to a
CTOD in a conventional three-point bend bar is
described. This procedure has been adopted by the
Japan Welding Engineering Society for structures
under large, cyclic, and dynamic loading conditions
such as seismic loading.

Copyright & 2002 John Wiley & Sons, Ltd.

These four articles are written for structural

engineers and do not require special expertise in
fracture mechanics. The reader of these three articles
should come away with more knowledge of the
fracture behavior of structural components that can be
readily applied in design and assessment. Perhaps
more important, however, will be a better
appreciation for the significance of fracture toughness
in the performance of structures.

Robert J Dexter PhD PE

Associate Professor, Department of Civil Engineering,
University of Minnesota, 122 Civil Engineering Building,
500 Pillsbury Drive SE, Minneapolis,
MN 55455-0220, USA
E-mail: [email protected]

Prog. Struct. Engng Mater. 2002; 4:15