Вы находитесь на странице: 1из 223

Structures – Viewpoints and Approaches

Engineering History and Heritage Structural Engineering Documents

Structures – Viewpoints and Approaches

Engineering History and Heritage

The present Structural Engineering Document (SED) is a
compilation of contributions devoted to the vast topic of history
of structural engineering as well as interventions on heritage
structures and structures of high cultural values. Various, some-
times opposed, viewpoints and approaches are expressed
and presented. The rather heterogeneous and controversial

Engineering History
nature of the content of this SED shall stimulate lively discus-
sions within the structural engineering community who needs
to increase the awareness of historical and cultural aspects

and Heritage
of structures and structural engineering. Current structural
engineering methods and practice are only at the very begin-
ning of effective engineering, really integrating historical and
cultural aspects in the assessment of existing structures and
in intervention projects to adapt or modify structures of cultural
values for future demands. Knowing the past is indispensable
Structures – Viewpoints

for modern structural engineering !
and Approaches

Eberhard Pelke

Structural Engineering Documents

Eugen Brühwiler

International Association for Bridge and Structural Engineering (IABSE)

SED15_Cover.indd 1 05/12/17 2:44 PM

About the Authors
To provide in-depth information to practicing stuctural
engineers in reports of high scientific and technical standards
Eberhard Pelke earned his civil engineer diploma
on a wide range of structural engineering topics.
from Darmstadt University. He is Head of the
Department of Bridge Engineering at Hessen Mobil
IABSE Bulletin Board:
Road and Traffic Management in Wiesbaden,
H.Subbarao, (Chair), D. Laefer, (Vice Chair), M. Bakhoum,
Germany. He is a Fellow of IABSE and was from C. Bob, M.W. Braestrup, N.P. Hoej, H.H. Snijder, R.von Wölfel,
2013-17 Chair of IABSE WG9 Construction R. Mor, M.G. Bruschi, I. Payá-Zaforteza, S. Kite, M. Garlock.
The International Association for Bridge and Structural
Engineering (IABSE) operates on a worldwide basis,
with interests of all type of structures, in all materials. Its
members represent structural engineers, employed in design,
Eugen Brühwiler is professor of structural engi- academe, construction, regulation and renewal. IABSE
organises conferences and publishes the quarterly journal
neering at the EPFL—Swiss Federal Institute Structural Engineering International (SEI), as well as reports
of Technology in Lausanne, Switzerland. His and monographs, including the SED series, and presents
activities include modern methods to examine annual awards for achievements in structural engineering.
and enhance structures of high cultural value. With a membership of some 4,000 individuals in more than
He is a Fellow of IABSE and was from 2013-17 100 countries, IABSE is the international organisation for
structural engineering.
Vice-Chair of IABSE WG9 Construction History.
The International Association for Bridge and Structural
Engineering (IABSE) was founded as a non-profit scientific
association in 1929. Today it has more than 300 members iQ
over 0 countries.IABSE’s mission is to promote the exchange
of knowledge and to advance the practice of structural
engineering worldwide. IABSE organizes conferences
and publishes the quarterly journal Structural Engineering
International, as well as conference reports and other
monographs, including the SED series. IABSE also presents
With Contributions From:
annual awards for achievements in structural engineering.
E. Pelke, T.F. Peters, M. Traykova, R. Vergoossen. A. Bögle, W. Lorenz, E. Brühwiler,
N. Janberg, J. Romo, B. Addis, M.J. Beiersdorf, J. Steiner, E. Vianen, R. Spaan,
For further Information:
J.F. Duntemann, B.R. Greve, A. Traykov, D. Partov, N. Winterbottom, B. Heres,
T. Chardakova, D. Wendland, P. van Bogaert, R. Barthel, J. Tutsch, J. Jordan, C. Weber, IABSE
c/o ETH Zürich
A. Kostka, M. Fischer, G. Eisele, J. Seiler, V. Wetzk, Y. Rammer, B. Espion, L. Clarke,
CH-8049 Zürich, Switzerland
M. Bartzsch, K. Geißler, D. Gasparini, W. Vermes, J. Voermans, R. May, Y. Yang, Phone: Int. + 41-44-633 2647
B. Chen, S. Nakamura. Fax: Int. + 41-44-633 1241
E-mail: secretariat@iabse.org
Web: www.iabse.org

SED15_Cover.indd 2 05/12/17 2:44 PM

Copyright © 2017 by
International Association for Bridge and Structural Engineering

All rights reserved. No part of this book may be reproduced in any form or by any means, electronic
or mechanical, including photocopying, recording, or by any information storage and retrieval
system, without permission in writing from the publisher.
ISBN 978-3-85748-154-3

c/o ETH Zürich
CH-8049 Zürich, Switzerland

Phone: Int. + 41-44-633 2647

Fax: Int. + 41-44-633 1241
E-mail: secretariat@iabse.org
Web: www.iabse.org

SED15_title-page.indd 2 04/12/17 8:40 PM


The roots of modern construction historiography reach back to the 19th century when for exam-
ple in France the engineer Auguste Choisy (1841–1909) began to explicitly analyze the con-
struction of historic buildings and to place them in the center of construction history.1 In the
last third of the 20th century, such approaches followed systematically and in an increasingly
professional way. A long arch spans from the works like those by Jacques Heyman (1925) who
interpreted ancient techniques and theories related to vaults by means of modern structural engi-
neering approaches,2 to the historic–theoretical research and publications like those by Karl-
Eugen Kurrer (1952).3

In the meantime, several chairs and professorships in construction history were created and
there is an impressive variety of conferences and publications. Every three years since 2003,
the scientific community gathers at the International Congress on Construction History (ICCH).
There is no doubt that construction history has established and consolidated internationally as
an independent discipline.

Actually, what is construction history? Professor Werner Lorenz, member of the IABSE WG9
Construction History, defines construction history as follows:

Structural engineering is the entity of the practices and products of conceptual design, dimen-
sioning and construction of technical structures and components in the process of the con-
structional designing of the environment. Construction history describes and interprets these
practices and products in their historic sequence. For that purpose, construction history inter-
rogates the products of construction and all associated written and pictorial sources. Both the
historic construction research and the methods of static-constructive and scientific engineering
analyses belong to the methodical cornerstones.

Construction history involves architects, monument conservators, historians and engineers in a

transdisciplinary approach to fulfill scientific, cultural, didactic and also structural engineering
tasks and requirements.4

IABSE WG9 Construction History has the general objective to promote this new science and
to demonstrate its importance for structural engineers. The three main objectives of the WG on
construction history are to:

• increase awareness among structural engineers of historical and cultural aspects of struc-
tures and structural engineering;
• illustrate and propagate the social and technical achievements of civil engineering;
• improve methods and practice in structural engineering by showing ways for systematic
and targeted integration of historical and cultural aspects in intervention projects to adapt
or modify structures of cultural value for future demands.

IABSE WG9 focuses on the role of construction history in the structural engineering practice
and is thus intentionally complementary to the classical construction history as understood by
the ICCH Community. The main concern of WG9 is thus to implement construction history in
the daily work of structural engineers and to demonstrate the importance of cultural values as a
basic design parameter when interventions on existing structures are required.

The present Structural Engineering Document (SED) is structured accordingly. It shall be

understood as an introduction into construction history and how to consider the cultural values
of structures in intervention projects. Although this SED is addressed primarily to IABSE struc-
tural engineers, it may also be useful for nonengineers.

This SED begins with the Editorial written by one of the “deans” of construction history: Tom
F. Peters. Personal statements by several WG9 members testify a surprising variety of ways how
the access to construction history was found and how it influenced professional activities. In the
next chapter, Nicolas Janberg provides a worldwide survey on the activities and contacts in the
domain of construction history. In the following, the papers by Max Johann Beiersdorf and Josef
Steiner are contributions similar to essays on the aspects of construction history.

Twenty-five case studies on rehabilitation and modification of structures form the core material
of this SED. Every case study outlines on a maximum of four pages the cultural values of the
structure and highlights the appropriate measures for its respectful preservation. References
and contact data of the author serve the reader to obtain detailed information. The case stud-
ies obviously range from ancient to modern structures and from medium to high cultural val-
ues, comprising various types of structures. Requirements of cultural heritage shall be taken as
inspiration (and no longer as “hindering constraint”) for better intervention projects on existing
structures. Construction history and cultural values of structures have yet to be understood as
basic structural engineering disciplines.

With the present SED, the IABSE WG Construction History intends to make a significant con-
tribution to modern structural engineering and to provide access to construction history for
practicing structural engineers.

[1] Choisy A. L’art de bâtir chez les Romains Ducher: Paris, 1873.
[2] Heyman J. Coulomb’s Memoir on Statics: An Essay in the History of Civil Engineering
Cambridge University Press: Cambridge, 1972.

[3] Kurrer K-E. The History of the Theory of Structures Ernst & Sohn: Berlin, 2008.
[4] Kurrer K-E. Aufgaben der Bautechnikgeschichte. https://gesellschaft.bautechnikgeschichte.
org/was-ist-bautechnikgeschichte/ Download 01.01.2017.

Eberhard Pelke, Chairman of IABSE Working Group 9 Construction History

August, 2017
Table of Contents

1 History as Educator and as an Aid to Understanding Structural Engineering;

T.F. Peters 1

2 Personal Statements by Members of IABSE WG9 Construction History:

What is Construction History and Why is It Significant for Structural Engineers? 7
Eberhard Pelke 7
Marina Traykova 8
Rob Vergoossen 9
Annette Bögle 10
Werner Lorenz 11
Eugen Brühwiler 12
Nicolas Janberg 13
Jose Romo 15
Ignacio Paya-Zaforteza 15
Bill Addis 16
References 19

3 Engineering History and Heritage Structures around the World—A Survey;

N. Janberg 21
Introduction 21
Survey 21
Current Situation of Construction History around the World 22
Overall Evaluation and Summary 38
References 39

4 (Re)constructing History—How Building Archaeology Can Profit from the

Knowledge of Engineering; M.J. Beiersdorf 41
Introduction 41
Undulating Mud Brick Walls in Pharaonic Egypt 42
Engineering Science Studies 44
Building Archaeology and Construction History—A Fruitful Cooperation 44
References 45

SED_15-1TOC.indd 1 04/12/17 8:50 PM

5 The Many Footprints Left by Martin Bachmann in Pergamon; J. Steiner 47
Dedication 47
Building Z on Pergamon’s Acropolis Hill 47
Repair and Conversion of the Southern Rotunda Next to the Red Hall 49
Assembly of a Monumental Supporting Figure Next to the Red Hall 51
Maintenance of Retaining Walls on the Acropolis Hill 51
Anastylosis of a Palaestra Corner in the Gymnasium 52
Final Remarks 53
References 53

6 Buildings 55

6.1 The Building A of Radio Kootwijk—A Concrete Building from 1920, Ready
for the Future; E. Vianen, R. Spaan 55

6.2 Marina City—The History and Restoration of an Iconic Facade;

J.F. Duntemann, B.R.Greve 61

6.3 Rehabilitation of the Complex Reinforced Concrete Shell Roof Structure

of an Industrial Building; A. Traykov 67

6.4 Maintenance and Strengthening of the Timber Roof Elements in the

Church of St. Dimitar; M. Traykova, D. Partov 71

6.5 Brighton Pier, UK—Innovation in Renovation; N. Winterbottom 77

6.6 Early Iron Structures at the Hermitage in St. Petersburg—Unique

Testimonies to Construction History and the Associated Preservation
Problems; B.Heres 83

6.7 Maintenance and Strengthening of the Cross-Shaped Barracks Building;

M. Traykova, T. Chardakova 89

6.8 Analytical and Experimental Studies on the Technology of Late-Gothic

Vault Construction; D. Wendland 95

6.9 Frost Damage and Restoration of Limestone Domes and Spheres in a

Heritage Building; P.V. Bogaert 101

6.10 The Gothic Tower of Freiburg Minster, Germany: Analysis and Repair;
R. Barthel, J. Tutsch, J. Jordan 107

6.11 The Municipal Public Bath at Strasbourg (1905–1908): A Cultural Heritage

in Reinforced Concrete; C. Weber, A. Kostka 113

6.12 History and Rehabilitation of Reinforced Brick Ceiling; M. Fischer 119

6.13 Reconstruction of the Neues Museum in Berlin; G. Eisele, J. Seiler 125

SED_15-1TOC.indd 2 04/12/17 8:50 PM

7 Bridges 131

7.1 The Necessity for Construction History to Assess Historic Bridge

Bearings; V. Wetzk 131

7.2 Refurbishing of the Nibelungen Bridge in Worms, Germany; E. Pelke 137

7.3 Early Prestressed Steel–Concrete Composite Continuous Bridges

in Belgium; Y. Rammer, B. Espion 143

7.4 Making Rennie’s 1796 Lune Aqueduct Watertight Again; L. Clarke 149

7.5 Restoration of Robert Maillart’s Reinforced Concrete Bridges

in Switzerland; E. Brühwiler 155

7.6 Examination of Two Riveted Railway Bridges over the River Rhine;
E. Brühwiler 161

7.7 The Rendsburg High Bridge across the Kiel Canal; M. Bartzsch, K. Geissler 167

7.8 Steel Viaduct Refurbishment Inspired by the Original Structure and

Its History—The Best Solution with Regard to Structural, Economical and
Heritage Requirements; W. Lorenz 173

7.9 The Main Avenue Bridge, Cleveland, Ohio, USA; D.A. Gasparini, W. Vermes 179

7.10 Renovation of a Historic Railway Lift Bridge; J. Voermans, J. Reusink 185

7.11 An (Almost) Extinct Engineering Heritage Asset—The Case of the

Reichsautobahn Bridges; R. May 191

7.12 Construction Technology of Chinese Woven Timber Arch Bridges;

Y. Yang, B. Chen, S.Nakamura 197

SED_15-1TOC.indd 3 04/12/17 8:50 PM

SED_15-1TOC.indd 4 04/12/17 8:50 PM


History as Educator and as
an Aid to Understanding
Structural Engineering

Tom F. Peters, Professor emeritus, Lehigh University, Bethlehem, Pennsylvania, USA

The following position papers present many valid aspects of construction history. Eugen
Brühwiler discusses the sheer intellectual fascination of historical material, so different
from ours today, and the inspiration that this can generate. Rob Vergoosen sees historical
arguments as a way to convince young engineers that computer programs are tools and not
engineering design at all. Eberhard Pelke points out that history proves that mathematical
methods are also only tools. He also mentions the inspiration that exemplary professional
biographies can exert. Marina Traykova discusses the importance of historical structures to
the definition of national culture, and Brühwiler ties their importance to professional culture.
And the many interesting case studies of built structures that have been presented at IABSE
meetings and symposia over the years illustrate these and further aspects and particularly
their usefulness for informed structural evaluation as relevant for the purpose of restoration
or adaptation. I would like to expand a little on the interest in historical understanding and its
impact on engineering thought and especially on education that both Brühwiler and Werner
Lorenz raise.

Understanding is abstract and at first sight, it seems to have little to do with the daily concerns
of practitioners, but understanding has the ability to change and to expand professional culture
in very concrete ways. I will try to demonstrate through examples how that which we might tend
to dismiss as being mere impractical philosophy is actually quite down-to-earth. Understanding,
as opposed to a simple accumulation of facts and methods that usually counts as professional
knowledge, is deeper and it directly impacts our thinking and approach to everyday problems.
Memory is one of the characteristics that make us human, and history is the concretization of
memory. It situates us in our world, also in our professional world of engineering, and as Jean-
Claude Badoux succinctly wrote and I quoted in the preface to IABSE the first 80 years, “no
history, no memory, no future”. Engineers are the professionals who create the future structure
of our world, and in order to do that, they not only learn from theory, but also from experience,
from mistakes and from successes. That is the value of case studies, which are historical by their
very nature because they deal with completed design and building processes and thus with the

Chapter 1.indd 1 12/4/2017 3:49:15 PM


past. However, we need to acknowledge that when we look back, nothing is absolute: we see
what I call a relative reality. We see facts and their relationships from a constantly changing
standpoint. Werner Lorenz touches on this issue in his contribution.

I compare this to a walk along a riverbank. Looking at the opposite shore, we perceive a land-
scape with trees large in the foreground hiding what appear to be smaller ones farther back.
As we continue along the bank and around bends in the river, our standpoint changes and the
relationships between the trees change: what appeared to loom large previously, recedes and
becomes small, and what was in the background now grows and moves to the fore. The objects
are the same we saw before, but they now stand in different relationships to one another and
to us. The trees are our facts, and they remain the same, but they can look quite different from
another viewpoint. When we hike in the Alps, the mountain we know so well looks unrecogniz-
able from a different standpoint and yet it is a stable, concrete fact—as long as we disregard the
time factor. In reality, only change is constant.

And so it is with case studies, whether of objects or of personal careers in engineering: our
standpoint and our questions change, and with them so do our answers. What was interesting
and valuable a few decades ago is now no longer so, and what did not exist before is now
important. We must realize this in our daily practice when we evaluate a building for demo-
lition, rehabilitation or restoration. It is a delicate balance we tread when we decide which
are important examples of our past. In structural terms, we find ourselves intellectually in
unstable equilibrium. There is no absolute “right” and “wrong”, it all depends on our criteria
and viewpoint. This destabilizes our understanding. We lose our bearings in the world, and
engineers always choose stability over manifestations of instability. We want “real stuff”,
concrete and immutable facts to hold on to, something solid—but our world is not like that.
Intellectuality, like physical instability, makes us move. The role of historical study in educa-
tion is to force us to leave our intellectual “comfort zone” and to make our standpoint more
flexible. Even the body of knowledge known as statics and strength of materials depends on
our viewpoint. The facts are there, but the methods we construct on selected facts depend on
the criteria we choose and the relationships we impose upon them. Flexibility is what makes
a good theoretician, designer or builder. It gives us the ability to react to changing situations.
When presenting the symbolical millionth book to my former university’s library, Galileo’s
1638 Discorsi e Dimostrazioni Matematiche Intorno a Due Nuove Scienze, the then president,
a civil engineer, said of Galileo’s depiction of a cantilever, that the author had made a mistake
in depicting the stresses AB in the cantilever as being constant throughout the cross- section
(Fig. 1).

“Even a man as great as Galileo, the founder of modern physics was not immune to mistakes”,
he told us. The president had missed the point. In fact, Galileo had made one of the great dis-
coveries of his age: he had realized that the force E is external and visible, while the resultant
stress AB is internal and invisible—and he depicted this graphically, showing the two to be
causally related. Far from being simple in his assumptions, Galileo was a genius and had moved
our understanding of structure forward in that one diagram! The facts of the diagram are incon-
testable: the distribution of stress is wrong by our modern understandingi, but the impact of

This is one interpretation of what Galileo may have meant; other interpretations are also possible.

Chapter 1.indd 2 12/4/2017 3:49:15 PM


Fig. 1: Galileo’s depiction of a cantilever

his realization, the “truth”, is another. So “truth” depends on our viewpoint. Had that president
studied the history of his field, he might have realized that this diagram was not a demonstration
of fallibility, but a manifestation of the value of structural history to engineering education. So,
in my view, the IABSE History WG needs to concern itself not only with intellectual delight or
with information that we can use to solve immediate problems, but it should also teach us how
to step back, to reconsider, and to formulate our problems, and to ponder not only what, but
also how we learn.

When talking about this need to engineering educators, the invariable answer is: “Of course
this is important, but our time to educate engineers is short and the curriculum is so overloaded
that we have no possibility to include it.” Is this really so? Do we need to fill students’ heads
with ever more facts and methods that often become obsolete before they enter practice, or
should we rather teach them how to think, how to search out the relevant facts for themselves,
how to put these facts into context in different ways, how to formulate questions, how to react
flexibly to changing situations, how to gain an overview and gain distance to the immediate
“correct”, economic, technical solution? All creativity, whether in engineering or in any other
field, only occurs when one begins to “think outside the box”. History can help to see over the
edge of the “box” that defines and indeed confines our logic. Historical awareness therefore
helps us to search for alternatives to unthinkingly accepted solutions. Is that not the way to the

To illustrate how to step back and question our basic logic, let us compare two bridges from two
different epochs: Hans Ulrich Grubenmann’s Rhine Bridge of 1757 in Schaffhausen (Fig. 2) and
Julius Natterer’s Neckar Bridge of 1976 in Stuttgart (Fig. 3).

Chapter 1.indd 3 12/4/2017 3:49:15 PM




Fig. 2: Hans Ulrich Grubenmann’s Rhine Bridge in Schaffhausen (a) overall view, (b) detailed
Both are trussed wooden structures and have two spans of the same size and the same sharp
bend between them. Both were built for virtually the same load: the first for a fully loaded
18th-century wagon and the second for a modern pedestrian load. The results look very differ-
ent and the first seems structurally redundant and illogical from our standpoint. But despite the
differences between modern and ancient material and connection technology, they are similar
and both logical.

When we understand that in the pre-statics era Grubenmann designed by overlaying one known
simple system that worked with another to extend span and increase loadbearing capacity, we
can easily read his structure as an overlaid configuration of king posts, queen posts and slanted
struts. It now looks less confused, because we understand Grubenmann’s logic. And we can see
what a gifted constructor he was when we notice how his overlaid configuration also formed a
series of superimposed arches, the longest one carrying through from one abutment to the other
over the middle pier. This pier had remained from the washed-away predecessor bridge, and the
client insisted that he reuse it. Grubenmann distrusted its stability, and in order to relieve it as
much as he could, he incorporated that 110 m long arch into his continuous beam. Of course,
looking at it from a classical standpoint, we “know” that the top of a continuous beam works
in tension over the intermediate pier and presume, perhaps correctly, that Grubenmann did not
know this. But we have forgotten an even more relevant fact that Grubenmann as a gifted prac-
titioner knew very well: in a stick structure, we can guide tension and compression in any way
we wish through a structure!

Chapter 1.indd 4 12/4/2017 3:49:15 PM




Fig. 3: Julius Natterer’s Neckar Bridge in Stuttgart (a) overall view (b) detailed view

Now we see the true value of the historical standpoint: it allows us to shift our logic to under-
stand a different form of equally correct logic that preceded our modern, model-based view-
point, and we step back from what we unquestioningly call the “truth” and into Grubenmann’s
“truth” that bases on entirely different criteria. We can use this insight to question our own
logic. It destabilizes our thought processes and makes us realize that our logic does not create an
absolute, but a relative reality. This allows us to create new structures. That is what construction
history can teach us, and its value to our profession is evident.

Chapter 1.indd 5 12/4/2017 3:49:17 PM

Chapter 1.indd 6 12/4/2017 3:49:25 PM


Personal Statements by Members
of IABSE WG9 Construction History:
What is Construction History
and Why is It Significant for
Structural Engineers?

Eberhard Pelke
In 1995, my administration sent me to the “Deutsches Straßenmuseum” (German Road
Museum). I was to help as a “bridge builder” in setting up the “Bridge and Structural Engineer-
ing” part of the exhibition. There, I met the great engineer Klaus Stiglat. He took my by the hand
and taught me, as a practicing engineer, to see not just the here and now of the technical rules
and standards but also the past work of the engineers on whose shoulder we rest. Together we
developed the “History of Bridges – the Age of the Engineers” section of the exhibition. Sixteen
large-sized boards displayed the development of bridge building starting in ancient times using
the four basic bridge types—girder, arch, suspension and cable-stayed bridge. The task was to
elaborate and personify the milestones of bridge building. What was the critical development
step, who was the key engineer in defining the structure—indeed, a question of stance—inform-
ing a society critical of technology about the work of construction engineers via their work and
personality. That was the beginning, purely practical without scientific superstructure and the
call for creating discipline. I am very grateful to Klaus Stiglat, not only for an introduction to
construction history but also for the questioning of and applying the same in everyday engineer-
ing. Even today, to me construction history is not an academic exercise that documents and
archives the past, brought to a scientific community with a presentation at an international con-
gress, but rather a tool for everyday engineering with which to understand structures, to evaluate
and protect them, and to improve them in accordance with their load-bearing behavior. Or, to
detect defects that may result in demolition. Yes, that is the run of things.

A structure and the personality of the engineer are mutually dependent. Engineers do not nec-
essarily arrive at the same optimum. Designing trusses is a process of rational physical crite-
ria and the subjective perception of the engineer. He brought construction critique, a topic in
construction history, back into the limelight. He taught me about that too. And above all, it is
important to him to retrieve the lifework of great engineers from the anonymity of the infor-
mation society. His many interviews as Chief Editor of the magazine “Beton- und Stahlbeton-
bau” (Concrete and Reinforced Concrete Construction) and his book “Bauingenieure und ihr
Werk” (Construction Engineers and Their Works) document construction history brought to
life as a primary source. Together with Karl-Eugen Kurrer, we recalled the lifework of Helmut
Homberg (1909–1990) to the memory of younger construction engineers. Just in time, before

Chapter 2.indd 7 12/4/2017 3:52:17 PM


the last sources of the genial yet uncomfortable engineer are found down the usual track of a
throwaway society. And what emerged: a work exhibition of the construction of major bridges,
which is perhaps more meaningful than some engineer who published diligently and was sent
into retirement with academic honor. Homberg’s last bridge, the cable-stayed QE II Bridge
near London, was opened for traffic after his death (1991): “Working with him was hell, but
the result was compensation for everything”, said Hugh Knox in an obituary. Construction his-
tory brought Karl-Eugen Kurrer who was for long time the link and “Mother of the company”
of the handful of German construction historians. After lengthy, German discussions, we now
have an organization, the “Gesellschaft für Bautechnikgeschichte” (Society for Construction
History) chaired by Werner Lorenz. At last! But what would the history of German construction
be without Karl-Eugen?

I am his foster child too. He took me to Madrid to the first International Congress on Construc-
tion History. He took care of the knowledge regarding the scientific requirements and creation of
discipline in construction history. Through him, I was able to build up my international network.
I built my five contributions to the International Congress on Construction History on his scien-
tific basis. Karl-Eugen always had an open door for his authors, for his construction historians.
He listened, made contributions, argued, agreed or expressed reservations. He was a disciple and
true follower of Fritz von Emperger. Hour-long telephone calls drafted new theories and essays
for his steel construction. Construction history brought to life! As a practicing engineer, I regret
that I have no time for great theories—I would have liked to have had the time to spin a yarn to
the end in a structured manner! Klaus Stiglat and Karl-Eugen Kurrer stand for my understanding
of a lived and practically orientated construction history.

Short Bio
Eberhard Pelke obtained his civil engineering degree from the University of Darmstadt. Since
1990, he is with Hessen Mobil—Road and Traffic Management, and since 2001 Head of the
Department of Bridge Engineering, in Wiesbaden, Germany. He is a Fellow of IABSE and Chair
of IABSE WG9 Construction History. He is also a member of the German Society of Construc-
tion History (Gesellschaft für Bautechnikgeschichte).

Marina Traykova
The cultural and historical heritage is the most important of a nation’s wealth and a major factor
in the accreditation of national identity. Historical buildings and facilities represent one of the
most important parts of historical heritage. Preservation of these buildings and facilities for the
future generations is a way of sustainable development and also an act of culture. The historical
buildings and facilities can be viewed as a work of art representing its time period. The general
concept of construction history relates to the safeguard, recognition and continuation of cultural
and historic traditions in national construction practices for the public interest of the society.
All these construction practices represent the culture and the techniques of time along with
sentiment, intent and conscience of its designer, artist or craftsman. The recognition and the
comprehension of construction history influence the selection of the method, policy and degree
of intervention, and preservation methodology and techniques.

Chapter 2.indd 8 12/4/2017 3:52:17 PM


The most sensitive aspect of construction history is the structural preservation because of the
decisive role of the structure in the global behavior of buildings and facilities. Usually the struc-
tural preservation is related to the solution of the problem “safety versus authentic structure”.
There is no generally applicable answer and the solution should be based on specific conditions.
Finally, the most important is to make a decision which will permit to preserve, to rehabilitate
or to reconstruct the original structure. This decision will enable prolonging the life of the build-
ings and facilities and will give the future generation an opportunity to learn the history and
should be considered as a continuity of history. Construction history is important for structural
engineers and IABSE because of the possibilities to:

• increase professional knowledge and experience;

• make the connection with history and the national traditions;
• learn about the old structures, construction materials and techniques, and their application
in actual situations;
• inspire new design solutions;
• provide continuity of history;
• provide sustainable adaptation and reuse of historical buildings and facilities;
• revive urban areas;
• maintain traditional standards;
• exchange good construction practices.

Short Bio
Marina Doncheva Traykova, PhD, is a Full Professor at the University of Architecture, Civil
Engineering and Geodesy in Sofia, Bulgaria. Her main research activities include rehabilitation
and strengthening of existing buildings, construction history, preservation of heritage buildings
and sustainable design. She is a member of IABSE (Working Groups Construction History
and Earthquake Resistant Buildings), fib and the Chamber of Engineers in Investment Design,

Rob Vergoossen
As a structural bridge engineer with an expertise in existing structures, I am fascinated by the
history of construction. Today, we are faced with a large existing bridge stock which is over-
loaded by the current immense traffic. Fortunately, our bridges are well engineered by our pre-
decessors. With the knowledge, materials and codes of their times, they did a great job. It is our
job to preserve these bridges and make the best out of them. Our young(er) structural engineers
have to be taught the engineering of these structures without the use of computers. With this way
of thinking, residual capacity in a lot of existing structures can be found. Therefore, structures
that are functionally sufficient can be conserved and strengthening can be avoided or minimized.
Society can then profit the most, as costs and hindrance of traffic are minimal. It is important that
we teach our structural engineers the basics of ancient materials, codes and structures. Recent
(new) codes are not meant for structural assessments, as they are a result of the materials and

Chapter 2.indd 9 12/4/2017 3:52:17 PM


knowledge of today. Therefore, there is a need for more flexibility and expert opinions in assess-
ing an existing structure.

Short Bio
Rob Vergoossen got his master’s in structural engineering at Delft University of Technology in
1999. From the start of his career, Rob has worked on existing structures and has been involved
in the drafting of assessment guidelines in the Netherlands for existing structures. Since 2015,
he is a part-time PhD student at Delft University of Technology in this field. Since 2010, he is an
expert in concrete structures at the engineering company Royal Haskoning DHV. At IABSE, he
has contributed as a member of various committees and has authored several papers.

Annette Bögle
History of design of structures is in my opinion an essential part of construction history and
in my practice as a university professor a basic element for teaching creativity and conceptual
design in structural engineering. Regarding today’s built environment and future developments,
conceptual design needs to be a core approach in any academic curriculum to guarantee an
education in sustainability. In many actual building projects like wide span roofs, high-rise
buildings or bridges, an increasing rationalized planning of special engineering structures can
be noticed, whereas the distinct boundary conditions are often neglected. The results are seldom
satisfying from a technical, aesthetical and social point of view. For this, among other things, the
engineers are responsible themselves. This is due to the fact that during the last decades, civil
and structural engineers have focused unilaterally on the rational–analytic aspect of their area
of responsibility and have therefore lost sight of their real and original area of activity, which
arises in a mixture of logical and empirical knowledge. As construction history points out the
dependencies of the boundary conditions and the influence of the designer’s personality, it is
predestinated to mirror the requirements of a holistic design. Therefore, it is now becoming an
essential element in teaching.

First, the clear and self-evident advantage of construction history is that in history, the boundary
conditions of conceptual design—focusing on the dependencies of form, structure, material and
construction—become obvious. Depending on time, knowledge and capabilities of the building
society, only specific structural solutions can arise. Vice versa, historic structures become only
readable if these boundary conditions can be studied. Therewith, a deep understanding of the
different, complex and time-depending boundary conditions arises. It becomes obvious that at a
different time, in a different situation and with different people, the resulting structures are dif-
ferent too. This enables one main aim of studying and teaching history of design of structures:
the transfer of the knowledge of the uniqueness and time dependency of boundary conditions
into the actual discussion of the design of engineering structures. Examples like the Pantheon
enable us to realize that not everything was or is buildable: while it was possible to incorporate a
whole sphere into this building, this ideal/spiritual perfection is not visible from the outside. In
history, the technical boundary conditions of what was possible were often quite limited. Cer-
tainly, the advancing socio-political enlightenment and the progress in science and technology
allow the preference of certain aspects. However, construction history shows us how compro-

Chapter 2.indd 10 12/4/2017 3:52:17 PM


mises had to be made, and how the knowledge and the values of the building society determine
how the interdependency of form and structure is expressed.

Second, construction history can even go a step further; it can also show the passion and the
intension of the personalities behind the designs. Particularly if a new technology or material
arose, the designers and builders explored with courage and creativity the limits of the techni-
cally feasible. All major developments in structural engineering can serve as examples for this.
If we accept the personal influence of the designer on the development of a structure, it is most
logical also to realize the cultural influence. This is also one of the reasons for different design
approaches, visible in different structural styles or in different calculation and dimension-
ing approaches. With the knowledge of the boundary conditions, the dependencies between
form and structure and the influence of the designer’s personality, our young engineers are
enabled to develop their own holistic approach in their designs. Construction history enfolds
engineer’s creativity; this enables new, innovative engineering solutions for actual engineering

Short Bio
Annette Bögle is a professor of structural engineering and Director of the Chair for Design and
Analysis of Structures at the HafenCity University in Hamburg, Germany.

Werner Lorenz
Construction history is a new scientific discipline just in the process of establishing itself, charac-
terized by transcending traditional academic demarcations: it is history of construction, history
of technology, history of science, yet it does not fit either of these categories. “A construction
represents and occupies a sort of boundary between the world of nature and the artificial world
of culture”.1 If we refer to this elegant definition, the role of construction history is no more or
less than to provide an explanation of the action of producing this hinge, in fact construction in
all its facets—thinking/planning, designing, abstracting, modeling, calculating and measuring,
specifying, financing, manufacturing, building, maintaining, reinforcing, destroying, dispos-
ing—in all its cross-linkages, interdependencies, and processes of interchange, right up to the
use of practical applications and technology. Constructing is a multi-faceted activity: It is this
process of construction in its entirety that constitutes construction history. Construction history
tries to understand the way human beings have approached building.

There are three main areas of investigation:

1. People: “There is no other history than that of people” (Lucien Febvre): Because any branch
of history is an examination, first and foremost, of people, and because the act of construct-
ing is deeply human and subjective, the first area of investigation of construction history
is people—master builders, engineers, researchers/academics, entrepreneurs/industrialists,
respectable artisans as well as brilliant mavericks, etc.
2. Institutions and networks: A second area of investigation is provided by the networks people
form and in which they act, structures in which and between which construction happens,
structures which influence civil engineering and which are influenced by civil engineering:

Chapter 2.indd 11 12/4/2017 3:52:17 PM


companies, associations, universities/colleges, national and municipal administrations, last

but not least the military sector.
3. Constructions: The third area encapsulates the sum total of the products of construction, be
they designs, theories or regulatory systems; and of course, and first and foremost, build-
ings, the constructions themselves.

History is interpretation—or it is not history: Construction history is not an exact science; it

is something between engineering sciences and humanities. Thus, construction history is
always—and especially in the IABSE context—endangered by the temptation to develop it into
the strategic narrative of the mighty stream rushing towards one’s own position. The methodical
imperative resulting from that is the obligation to refuse any legitimation, which perceives the
present only as the future of history and distorts history to justify what is. The most exclusive
commitment of construction history is to thwart this temptation by ignoring the term “progress”,
by disguising our common views, by approaching all history with deep humility. History has to
be read anew, again and again, as effects of different practices of construction, as side by side,
hardly consecutively. It has to be acknowledged first of all as a strange world with enigmatic
activities and practices, as the surprising, the unpredictable.

What is the significance of construction history for structural engineers and the IABSE? Of
course, the solution to the problems of today cannot be found in another epoch. The justification
of current practice does not rely on the past—but the past does recover its genealogy. Construc-
tion history read as such, is nothing but an analysis and critique of the present: It is an enrich-
ment for today’s construction practice, which welcomes challenges, courageously questions
current practices, and which develops optimal ways of managing and using materials. On a more
practical level, there is no need to stress that nowadays the field of handling with still existing
buildings, the field of reconstruction and rehabilitation, is one of the largest parts of building
activities. But an intelligent and appropriate intervention in existing structures is hardly imagi-
nable without knowledge of the history of construction. It is a necessary condition for every
present-day builder or designer who is concerned with our built heritage, be it in measuring and
monitoring or restoring and reinforcing. Whether one works with the old or new, in both cases
one needs history. All of a sudden it appears as a practical discipline—and as one of the basic
sciences of civil engineering (for a more detailed investigation see Ref. [2]).

Short Bio
Werner Lorenz graduated in 1980 in structural engineering at the TU Berlin. In 1992, he passed
his doctorate with a thesis about the early history of building with iron and steel in Berlin and
Potsdam. In 1993, he was appointed to the newly created chair of construction history at the
BTU Cottbus where he is active until now. In 1996, he founded his own consulting office for
structural engineering and assessment, which is specialized in the field of structural rehabilita-
tion of historic buildings and bridges.

Eugen Brühwiler
25 Years ago when examining the fatigue life of riveted steel railway bridges, one evening, I
found myself in the archives of the Swiss Federal Railways discovering construction plans,

Chapter 2.indd 12 12/4/2017 3:52:17 PM


notes, photographs and other documents from the 19th century: blue-colored rivet heads on con-
struction plans, black and white photographs from the construction era and treatises on bridge
design methods. I started “whispering” with Ritter, Tetmajer, Culmann and Gubser. This adven-
ture triggered my interest and fascination for history of bridge construction and the question:
where do structures come from? Although I had no education in history of construction during
my engineering studies at ETH Zurich, I quickly realized how indispensable information from
the time when structures were designed and built is when dealing with existing structures of any
age. Construction history deals with history as a basic scientific discipline to study the past of
civil engineering construction and its achievements. Construction history relates to the memory
and discovery of engineering methods and technology by collecting, organizing, presenting
and interpreting information about structural engineering and construction. Construction history
also deals with construction in a broad sense. It is about the fundamental value and importance
of understanding the design and construction of existing structures and their related contexts for
the examination of their today’s performance in view of future use.

Therefore, construction history is

• a basic scientific discipline consisting of rich data to identify and evaluate cultural values
inherent to existing structures and structural engineering;
• an indispensable basis for the examination of the performance of existing structures and
the design of interventions to rehabilitate, strengthen, and modify them for modern use and
demands, while respecting and integrating cultural values;
• about understanding the past to build the future, a rich source of inspiration, providing an
indispensable basis for the design of new structures.

Consequently, construction history is a basic scientific discipline that all structural engineers
should master. It has to become a fundamental course in any curriculum to educate structural
engineers. Most importantly, construction history enhances the general “culture” of structural
engineers, providing them the license to exchange with other professionals, like architects, on
equal grounds.

Short Bio
Eugen Brühwiler is a professor of structural engineering at the EPFL—Swiss Federal Insti-
tute of Technology in Lausanne, Switzerland, since 1995. His teaching and research activities
include modern methods of examination of existing structures and the use of ultra-high perfor-
mance fiber reinforced cement-based composites (UHPFRC) for strengthening of structures. He
is involved in the examination and restoration of bridges and buildings of high cultural value.

Nicolas Janberg
Construction History Can Change Your Life: My first foray into construction history was Prof.
David Billington’s class “Structures and the Urban Environment” which I took as sophomore
at Princeton University. While the class is meant to introduce the concept of structural art to
engineering and nonengineering students alike, it also provides an overview of structural and

Chapter 2.indd 13 12/4/2017 3:52:17 PM


construction history in Europe and the United States beginning with the industrial revolution
through the works of some very important engineers: Robert Maillart, Gustave Eiffel, Othmar
Ammann, Felix Candela, John Roebling and Christian Menn, The way Billington taught the
class changes the way one views the world, as virtually every student of the class will attest to. I
continued studying in the architecture and engineering program at the university, taking classes
in art history and architecture along with the typical structural engineering courses. At the time,
architects were required to learn about the history of their art and profession while engineering
students had the choice. To me it seems like a fundamental difference. Architects learn from
what has been built before in whatever way they choose—often as inspiration for their own
designs or as a reference to what is considered good design—while so many engineers end up
not even knowing whose works could inspire them.

My academic career continued as a combination of structural engineering and construction his-

tory. As a senior, I wrote a bachelor thesis on the historical development of Paris-La Défense. As
a graduate student, I chose to remain at Princeton with Billington as my advisor. I was a teaching
assistant in his class for two years in a row. My master’s thesis explored the evolution of the
modern cable-stayed bridge in Germany and France. I returned to Germany with two degrees
in civil engineering and started working as a bridge engineer at a typical German engineering
office doing detailed design for bridges. When I started, some colleagues were working on
Germany’s longest arch bridge project—the Wilde Gera Viaduct. One of my colleagues—just
as fresh out of school as myself—was proud to explain to me how innovative it supposedly was
to use cable-stays to cantilever the arch without using falsework. I quickly burst his bubble by
telling him that Eiffel had already done this on his bridges. Granted, Eiffel’s bridges were made
of iron while the Wilde Gera crossing was made of concrete, but he could have known about this
had he been taught construction history.

As editor of Structurae, I have continued to keep up with current developments in construction

as much as I have with historic structures. Through this work, I have read about many inspiring
and innovative engineers who have changed the profession in ways no one thought possible. I
have also been lucky to have met some of the ones who are doing it right now. One thing they
seem to have in common is how aware they are not only of the other contemporary designers
and their work but also of their predecessors, often citing them as inspiration. You could argue
that knowing about the past, in fact that knowledge of construction history—as well as a critical
appreciation of structural aesthetics—actually gives them a competitive advantage. I find it para-
doxical and sad that a profession that relies so much on what previous generations have learned
by condensing that knowledge into codes that every practicing structural engineer must adhere
to has actually lost its past and seems to find no practical reason to teach it to the next genera-
tion. It makes me wonder how different the world could look if every bridge designer had been
shown the works of Maillart or Menn during their studies, had learned to appreciate aesthetics
and critique their own and other people’s works in order to make them better.

Short Bio
Nicolas Janberg got his bachelor and masters of science degrees in engineering (civil engineer-
ing) at Princeton University and worked as a practicing bridge design engineer. He created
“Structurae”, the International Database and Gallery of Structures.

Chapter 2.indd 14 12/4/2017 3:52:17 PM


Jose Romo
In my view, a good background in construction history is essential for a contemporary struc-
tural engineer. I think it is not possible to fully understand our present-day work without a deep
knowledge of the evolution of the structural science and its technical application throughout
history. We need to know from where we come, and who were our predecessors and their chal-
lenges, mistakes and achievements. A structure is one of the most perfect elements created by
humankind; a bridge is one of the best examples of human creation where science, technique,
aesthetic judgment, culture, economic, politics and social organization of a historical period
crystalize in a single object. We can understand better previous societies by studying the culture
of construction of a single historical period; furthermore, the knowledge of construction history
enriches enormously the culture of anyone interested in structural engineering.

It is obvious that this knowledge is also a fundamental intellectual tool when an engineer has to
make an intervention in an existing structure. It is crucial to know the structural properties of
the materials used, how the connections among elements were conceived and are working, and
even, which were the codes and loads in the time of designing and therefore expectations we
may have regarding its current structural behavior. As a bridge designer, I have to emphasize
the importance of the deep understanding of our discipline and any other cultural knowledge,
in the crucial moment of the conceptual design of a contemporary structure. This is normally
an individual act where the insight of the engineer plays a fundamental role, and where the
past is usually a tremendous source of inspiration. In many occasions, today’s creativity just
emerges when an old structure is revisited using modern-day materials, resources or tools. With-
out knowing and learning from the past, that enrichment would be lost and as a consequence
the possibilities of progressing could be dramatically limited. I think it is our duty to promote
construction history as a way to preserve the legacy we have received. I strongly believe that
construction history will help the progress, development and improvement of structural engi-
neering and therefore will benefit the society we serve.

Short Bio
Jose Romo got his civil engineering education from the Polytechnic University of Madrid in
Spain. Since 1987 he is with FHECOR Ingenieros Consultores, S.A., now CEO of FHECOR.
He is also a professor of reinforced and prestressed concrete at the Polytechnic University of

Ignacio Paya-Zaforteza
The terms that dominated the courses I took as a civil engineering student were “bending
moments”, “concrete”, “Young’s modulus”, and similar others. Other words such as “society”,
“art”, “conceptual design”, “history”… were missing. Was this a right approach to learn engi-
neering? I think it was not. Structures are multidimensional and as such should be taught and
learned. Prof. Billington stated long time ago in his book “The Tower and The Bridge. The
New Art of Structural Engineering” that structures have a scientific dimension (related to their

Chapter 2.indd 15 12/4/2017 3:52:17 PM


efficiency), a social dimension (related to their cost) and a symbolic dimension (related to their
elegance). In addition, structures also have a value which is not only financial but also symbolic.

The study of construction history provides an amazing way of learning about the multidimen-
sional character of structures. The word “learning” involves looking not only at the best exam-
ples from the past, but also looking at examples which were not so good. If we do so, we will
foster the virtues of our profession, we will not repeat mistakes from the past and we will
promote a more critical profession and citizenship. Many other reasons support the necessity
of construction history. A very important reason to me is the enjoyment that it produces the
discovery of how our predecessors were able to design, build and create a legacy that provokes
admiration, respect, and passion for the work of the engineer. This enjoyment increases expo-
nentially when the structure is visited and seen as a reflection of the society that built it. This
is a wonderful experience that can be shared with family, friends, students and colleagues. So,
please join the club and become a construction history lover!

Short Bio
Ignacio Paya-Zaforteza is an associate professor at the Universitat Politècnica de Valéncia
(UPV), Spain. His main areas of interest are structural art, construction history, structural fire
engineering and the education of the engineer. He has worked at the UPV, at Princeton Uni-
versity and at the TU Berlin. He has carried out extensive research on the works of the Span-
ish engineer Eduardo Torroja and on the links among structural engineering, art, society and

Bill Addis
What use is construction history?i In his presidential talk to the Newcomen Society for the Study
of the History of Engineering and Technology, in 1945, the eminent construction historian Stan-
ley Baines Hamilton proposed six reasons why engineers should study history:

1. The detective interest of tracing knowledge to its source.

2. The opportunity of sharing a disinterested companionship with others who are following
kindred lines, as a member of such a body as our Society [The Newcomen Society].
3. The broadening of interest in engineering from the purely technical to a humane and liberal
field of study.
4. The light which the study of invention and discovery can throw on the working of the human
5. The genuinely recreative form of relaxation which history provides.
6. The contribution which the history of technology can make to the understanding of history
in general.4

This opinion as to why people should study the history of construction, or rather the history of
civil and structural engineering, reflects the background of the author (a research engineer) and

This essay is a reduced version of an editorial published in Ref. [3].

Chapter 2.indd 16 12/4/2017 3:52:17 PM


the time when it was put forward. Things have moved on and there seem nowadays to be six
main views that various people have put forward regarding the possible uses for construction
history, apart from its inherent interest.

The first is that construction history has no direct use. It is a branch of a wider academic disci-
pline of history and thus sits among the humanities. It aims to discover and record what hap-
pened in the past, and to explain this in the context of the contemporary states of technology,
economics and society.

A second view is that construction history is a cousin of the disciplines of history and philoso-
phy of science and history of technology. It thus belongs to the broad spectrum of studies called
“the history of ideas” and seeks to understand the knowledge and understanding in its field, and
to discover the nature and mechanisms of progress in the field. In doing so, it identifies the main
personalities and organizations involved in the discipline, and the part they played in contribut-
ing to progress within the discipline.

A third view is held by many engineers who believe that, by studying engineering failures in
the past, engineers can further their understanding of the behavior of structure and materials
and improve their skill as engineers. Today, this process is sometime called “forensic engineer-
ing” and has been practiced for at least two centuries, and probably very much longer. It has
undoubtedly been of extraordinary importance in certain cases; the collapse of Robert Ste-
phenson’s cast and wrought iron girder bridge in 1847 across the river Dee in Chester is often
mentioned in the many papers and books that have studied bridge failures. The most useful
outcome of this accident was the report that was published by the Royal Commission which
inquired into the cause of the accident.[5]However, while extremely interesting, it is not easy
to see how engineering today would be able to prevent a “similar collapse” since virtually none
of the circumstances of 1847 are relevant today. Similarly with the collapse of the Tacoma
Narrows bridge in 1940 due to wind-induced oscillations. In the investigations that followed
the collapse, it was discovered by the researchers (in the USA) that the Brighton chain pier had
collapsed in 1836 (already well-known history in the UK) and this fact has been used to claim
the potential benefits for engineers to study history. However, as with the Dee Bridge, this is
perhaps making a claim with the benefit of hindsight. In fact, the collapse of the Tacoma Nar-
rows bridge could not have been prevented even if its design engineers had been familiar with
every known details of the Brighton Chain bridge collapse—put simply, the Tacoma Bridge
collapse was largely a result of over-enthusiastic “value engineering” by the contractor who
sought to save money.6
Taking a more positive view of historical events, many engineers also argue that a knowledge
of engineering successes in the past can be a source of ideas that can be applied when design-
ing buildings and structures today—at the very least by avoiding “reinventing the wheel” and,
at best, increasing engineers’ knowledge of precedent beyond their own direct experience. This
“method”, if it can be called that, is now widely practiced in university courses that focus on
design, in construction and other engineering disciplines, under the heading of “precedent stud-
ies”. However, the benefit of this method is largely that the examples provide lecturers with con-
crete examples to illustrate the important principles of engineering design. This is not the same
as studying the history of engineering. Another view taken by many engineers is that studying
the historical engineering structures of the past and the engineers who created them is an essen-
tial part of developing a self-esteem for the profession, identifying role models to inspire young

Chapter 2.indd 17 12/4/2017 3:52:17 PM


engineers and to make nonengineers aware of the engineers’ contribution to society. It is almost
inconceivable that architectural history would not play an important part in the formation of an
architect, or music history in the formation of a musician—both to understand what has already
been done and to identify role models.ii Finally, perhaps the most direct use of construction his-
tory is linked to extending the life of existing structures—whether conservation, rehabilitation,
refurbishment or “retrofitting”. When approaching such work, a series of key investigations
need to be undertaken to determine:

• the design and construction of the original structure;

• the subsequent history of modifications, repairs and previous conservation and refurbish-
• the current state of the existing load-bearing structure and the likely cause of any cracking
or other structural damage and an estimate of its remaining life;
• the current state of the materials in the existing structure and an estimate of their remaining
Together, these can be taken as a detailed construction history of the structure. Compiling such
construction history of a building, bridge or other structure will reveal construction details of
the original structure as well as any subsequent repair, alterations or conservation it has under-
gone during its life. Such information is virtually essential to assess the current condition of the
structure and its materials, and to devise appropriate interventions which respect and conserve
what remains when extending the life of an existing structure. Based on this knowledge, good
conservation and refurbishment of existing structures can be achieved by selecting and apply-
ing appropriate repair, remediation, strengthening, and protection techniques. It is in this area,
perhaps, that the future development of the discipline of construction history may prove most
effective and fruitful. It is worth noting that the Construction History Society in Britain and its
journal Construction History are unique in treating the discipline exclusively as a branch of his-
tory, placing it firmly in the humanities rather than the applied sciences. As it says on its website,
“Construction History is the study of the development of the building industry”. Construction
history societies and congresses based in other countries incorporate both the historical purpose
construction history and the many ways in which it is linked to extending the life of existing
structures including conservation, refurbishment and “retrofitting”. The limited scope which
the Construction History Society embraces is partly explained by the fact that there are over a
hundred societies in Britain which deal with building conservation and architectural heritage,
and many of these have their own journals. Also, the Institution of Civil Engineers in London
addresses engineering aspects of extending the life of existing structures via its website and the
journal Engineering History and Heritage. In summary, there are three main types of interest in
construction history:

• as an independent academic discipline in the humanities, with limited application to the

modern built environment;
• as means to enhancing the engineering professions, both technically and in their standing;
• as an essential part of the process of extending the life of existing structures and conserving
our built environment.

Nevertheless, it seems that the historical content of architectural degrees is, indeed, being reduced in many courses.

Chapter 2.indd 18 12/4/2017 3:52:17 PM


Clearly, there is no single preferred or correct use for construction history, and a single study or
piece of research into the subject may partly fulfill any or all of the uses listed above. Nevertheless,
recognizing that there are different reasons for studying the subject, and that there are different uses
to which the results of such study may be put, can help researchers and authors give their work
a considered focus. When applying for research funding, academics would do well to remember
that funding bodies are increasingly inclined to seek benefits coming out of funded research. The
precise nature of these benefits should be the concern of us all. Above all, perhaps, we need to
develop a robust defense against the oft-quoted sentiment expressed by Hegel in his lectures on the
philosophy of history in the 1820s: What experience and history teach is this—that peoples and
governments never have learned anything from history or acted on principles deduced from it.[7]

Short Bio
Bill Addis is author of many publications on building engineering history including “Building:
3000 years of Design, Engineering and Construction”. He worked for 17 years at Reading Uni-
versity and 15 years with consulting engineer Buro Happold. Recently, he was visiting professor
at Tor Vergata University in Rome, University of Innsbruck, University of San Sebastian and the
ETH in Zürich. Bill is Chairman of the Editorial Advisory Panel for Engineering History and
Heritage published by the Institution of Civil Engineers, for whom he has recently developed
the Conservation Information Resource for Civil Engineers (CIRCE), an information resource
for conservation engineers.

[1] Musso SF. ‘Construction history’ and ‘Construction of histories’. University education and
the future of construction history. In Proceedings of the First International Congress on
Construction History, Madrid, Huerta S (ed), 3 vol., 2003; 1509–-1517.
[2] Lorenz W. From stories to history, from history to histories: What can construction history
do? Constr. History J. 2006; 21: 25–36.
[3] Addis B, Schlimme H. Editorial: The uses of construction history. J. Constr. History 2016;
[4] Hamilton SB. Why engineers should study history. Trans. Newcomen Soc. 1945; 25: 1–10.
[5] Walker Captain Simmons RE. Report to the Commissioners of Railways, on the Fatal Ac-
cident on the 24th day of May 1847, by the Falling of the Bridge over the River Dee. Avail-
able online at http://victoria.cdlr.strath.ac.uk/display.php?page=1&id=SGAV (Consulted
[6] Addis W. Design revolutions in the history of tension structures. Struct. Eng. Rev. 1994;
6(1): 1–10.
[7] GWF H. The Philosophy of History (Trans. J. Sibree) Batoche Books: Kitchener, Ontario,
2001; 19 Available online at http://socserv2.socsci.mcmaster.ca/~econ/ugcm/3ll3/hegel/
history.pdf (Consulted on 20/8/2016). “Was die Erfahrung aber und die Geschichte lehren,
ist dieses, daß Völker und Regierungen niemals etwas aus der Geschichte gelernt und nach
Lehren, die aus derselben zu ziehen gewesen wären, gehandelt haben.” Georg Wilhelm Frie-
drich Hegel (1770–-1831). Vorlesungen über die Philosophie der Geschichte (1st ed 1837).
Available online at http://gutenberg.spiegel.de/buch/-1657/1 (Consulted on 20/8/2016).

Chapter 2.indd 19 12/4/2017 3:52:17 PM

Chapter 2.indd 20 12/4/2017 3:52:17 PM


Engineering History and
Heritage Structures around the
World—A Survey

Nicolas Janberg, Struct. Eng., Berlin, Germany

The idea of this chapter is to provide a quick overview of the state of construction history around
the world. The focus is primarily on the world of structural and civil engineering, of course, but
in some cases, goes beyond into the realms of architecture where architectural history may over-
lap. Much more detailed descriptions of the state of construction history as an interdepartmental
academic discipline are available in other publications.i One of the tools used to determine the
state of construction for this chapter was through an online surveyii of IABSE members which
provided several responses but which are hardly representative of the overall state. Definitions
of construction history also may vary by country so that the responses did not necessarily yield a
single and strongly coherent picture. The results of the survey will be described in this overview
as they were given. As the number of responses was rather small, the survey results are aug-
mented with knowledge from the IABSE Working Group 9 and additional research and sources
wherever necessary and available.

In late 2014 and early 2015, IABSE’s Working Group 9 Construction History conducted an
online survey amongst members of the organization to obtain an overview of the state of

Additional and much more detailed sources on the state of construction history in Europe are found in Ref. [1].
An updated and expanded version of the above is entitled “L’histoire de la construction, Un méridien européen /
Construction History, A European Meridian” and was published online as a work in progress in 2015 at http://www.
histoireconstruction.fr/rapport2015/. It is scheduled for final publication in 2017 by Éditions Garnier.
Several civil engineering heritage country profiles were published in Ref. [2] on Spain and Germany (February 2016),
Japan and Wales (May 2016), Albania and Scotland (August 2016), Ireland (November 2016), Canada (February
2017), and England (May 2017).
The survey is still available online at https://docs.google.com/forms/d/e/1FAIpQLSfMPJCdSz1ZJJP-xJP-izuZugOke

construction history is in the members’ countries. Initially, the chairs of the national committees
of IABSE were asked to personally fill out the survey or ask someone in their group to do so.
Due to the small number of responses, a request was then sent out to the entire IABSE e-mail
newsletter list. This chapter evaluates the responses from the survey.

Survey Questions
The survey included a total of 19 questions and was divided into three sections. The first sec-
tion asked for information on the person answering the questions. The second and largest sec-
tion contained the pertinent questions regarding construction history and engineering education,
while the third section concluded the survey with questions aimed at establishing the current
situation of engineering practice in the member’s country. The complete list of the 19 survey
questions, explanatory notes to the questions and possible and expected responses is provided
in Table 1.

Survey Responses
A total of 21 responses to the online survey were collected. The first responses were recorded
on 4 November 2014, while the last member to fill out the form did so on 24 February 2015.
None of the responses were anonymous, and many of the responses contained answers to all
questions. Some questions did not require an answer if the previous response was negative.
Figure 1 summarizes the number and types of responses per survey question. Positive responses
were counted as such if the answer was “yes” and negative ones where a “no” or “do not know”
was given. Conditional responses included answers such as “yes, but…” and neutral responses
were those where free text was required or other responses were possible.

Countries Represented in the Survey

The 21 responses were from Austria, Belgium, Brazil, Bulgaria, Canada, the Czech Repub-
lic, Finland, France (three responses), India, Liechtenstein, Mexico, the Netherlands (three
responses), Portugal, Spain, Switzerland (two responses) and the United Kingdom. Due to the
duplications in the responses, the survey represents a total of 16 countries mostly from Europe
and the Americas.

Current Situation of Construction History

around the World
Instead of simply reporting the results for each question posed by country, the following will
summarize the responses received by groups of questions surrounding a specific issue in con-
struction history. Also, the responses will be augmented by information available to the working
group through research done in contexts not directly related to IABSE and its construction his-
tory working group as well as personal knowledge available through past activities in forums
dealing with construction history.

No. Question Allowed answers

Additional notes explaining the question
1. Please tell us about yourself
1.1 What is your name and affiliation? free Text
1.2 Could you please provide us with your e-mail address? free Text
1.3 Please name the country for which you are responding to free Text
this survey:
If your responses are specific to a region in your country,
please include this as well.
2. Institutions Dealing with Construction History
2.1 Do Technical Universities offer courses dealing with the Yes / No
history of construction? If so, please specify
2.2 If you answered “yes” to the above question, please free Text
Examples: Master degrees in construction history, build-
ing preservation, etc.
2.3 Is Construction History taught as part of the education Yes, in a dedicated course /
of civil and structural engineers at the bachelor or master Yes, as part of other
level? courses / No, not at all.
2.4 In the education of civil and structural engineering at the Yes, in a dedicated course /
bachelor or master level, are there any courses dealing Yes, as part of other
with existing structures? courses / No, not at all
For example dealing with the assessment, examination,
improvement, or monitoring of existing structures.
2.5 Is there a list of Cultural Heritage structures established Yes / No
by a governmental agency in your country?
This can include buildings, bridges, public works, monu-
ments, etc.
2.6 If yes, does this list include engineering structures? Yes / No
2.7 If engineering structures are included, what kind of struc- Bridges / Tunnels /
tures are on the list? Buildings / Dams, dykes,
etc. / Other (please specify)
2.8 Are engineering structures included as part of ensembles Yes / No / Don’t know
like transportation (road, rail, pedestrian/cycle paths)
infrastructure, energy production infrastructure, etc.?
2.9 Are there awards in structural engineering relating to Yes, there awards dedicated
existing structures? to existing structures /
Yes, but as part of other
awards / No
2.10 Are there associations dealing with Construction History? free Text
If yes, please name them and list the URL of the website
Table 1: Continued

No. Question Allowed answers

Additional notes explaining the question
2.11 Are there museums or galleries related to civil structures Yes, there are museums or
in your country? galleries dedicated to the
subject / Yes, but as part
of a larger or more general
collection / Yes, there are
travelling exhibits / No, not
that I know of

2.12 If there are museums or galleries, please name them: free Text
List the URL of the website if available
2.13 Are there exhibitions on structural engineers? Yes/No
2.14 Please name examples of exhibits held in the past or cur- free Text
rently ongoing:
List the URL of the website if available
3. Current situation of engineering practice in your country
3.1 Please name the 5 most eminent structural/civil engineers free Text
of your country:
If known, please add e.g. year of birth and a significant
structure designed by this person
3.2 Can you name notable projects involving existing struc- free Text
tures in your country?
E.g. regarding the re-use, strengthening, alteration/modi-
fication (extension, addition, upgrading, widening) of
existing structures. If available, please name an Internet
site or article with more information on the project.
Table 1: List of all survey questions

Construction History in Engineering Education

Questions 2.1 through 2.3 of the survey tried to evaluate if the topic of construction his-
tory played any part in the education of structural engineers. As construction history is not
an established academic discipline at this point in time, there are no degree programs at
the graduate nor at the undergraduate level specifically for the history of construction.
There are, however, degrees dealing with the rehabilitation of existing or historic struc-
tures, where historical aspects are of importance. Generally, if engineering students
come into contact with the history of their profession, it is usually by historical summa-
ries as part of other classes, if an educator chooses to outline the development of concrete
as a construction material in a module on concrete construction.iii In countries like Austria,iv

This was confirmed by survey responses from Belgium (Free University of Brussels), Brazil (University of São Paulo),
the Czech Republic, Mexico, the Netherlands, Portugal, Spain, and the United Kingdom.
At the Vienna University of Technology, the master of architecture program at the Institute of History of Art, Building
Archaeology and Restoration offers special modules related to the history of construction.

Number and types of responses per survey question

Number of responses (maximum 21)



1.1 1.2 1.3 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 2.13 2.14 3.1 3.2

Question number
Neutral Positive Conditional Negative
Fig. 1: Numbers and types of responses per question

Bulgaria,v or Germany, it is often the case that architects actually obtain engineering degrees so
that technical universities have schools or faculties of architecture. At least theoretically they
can offer opportunities to engineering students for taking classes in the history of art and archi-
tecture otherwise intended for architecture degrees. However, engineering students rarely are
able to take advantage of these opportunities. A rare example of a combined curriculum for engi-
neering and architecture is the program in “Architecture & Engineering”vi offered by Princeton
University’s Department of Civil and Environmental Engineering. The program was directed by
Prof. David P. Billington from 1990 until 2008 and is overseen today by Prof. Maria Garlock.
The curriculum includes a module entitled “Structures and the Urban Environment”, which
covers the history of structures from the beginning of the industrial revolution to today with
an emphasis on “structural art” as originally defined by Prof. Billington. The class is required
for students in the program but also satisfies laboratory and science requirements for nonengi-
neering students so that it is open to all. Students in the architecture and engineering program
are also required to take classes on the history of architecture through the Department of Art
History, as well as design studios and other classes offered by the School of Architecture in
order to obtain a Bachelor of Science in Engineering. The program does not continue at the

At the University of Architecture, Civil Engineering and Geodesy, there are modules dealing with the history of archi-
tecture or construction such as “History of Bulgarian Architecture”; “History of Modern Architecture”; “History of
Renaissance Architecture”; “History of Medieval Architecture”; “History of Ancient Architecture”. For master degree
students in structural engineering, there is a special module entitled “Diagnostic, Rehabilitation and Strengthening of
buildings”, which also includes a basic outline in Construction History.
https://www.princeton.edu/cee/undergraduate/program-tracks/architecture-and-engineer/ (visited on 29 March 2016).

graduate level, however. The concepts from this program have been transferred to other univer-
sities in the United States.

In France, construction history is more at home in architecture than in engineering institutions

as the former teach architectural history but the latter do not. The Ecole des Ponts ParisTech
(Marne-la-Vallée), formerly called the Ecole Nationale des Ponts et Chaussées, offers a com-
bined degree for “ingénieur-architecte” in cooperation with the Ecole d’architecture de la ville
& des territoires (Marne-la-Vallée) where a module on the history of construction is included.vii
The module is taught at the architecture school, however. Until only very recently, Brandenburg
Technical University at Cottbus and Senftenberg (Germany) was the only technical university in
German-speaking countries with a full professorship in the history of construction technology
(“Bautechnikgeschichte”). The chair has been held by Prof. Dr.-Ing. Werner Lorenz since 1993
and includes the conservation of structures (“Tragwerkserhaltung”) since 2001. Prof. Lorenz
offers various bachelor-level modules on the history of structural analysis, and construction
technology as well as modules dealing with the various aspects involved in assessing existing
structures. In 2016, Prof. Dr. Stefan M. Holzer was appointed to a full professorship of building
research and construction history within the architecture department of the ETH Zürich (Swit-
zerland)viii, though, neither the ETH Zürich nor the École Polytechnique Fédérale de Lausanne
(EPFL) in Lausanne currently teach dedicated modules on the history of construction. At the
master level, the EPFL offers a module on the “Aesthetics of civil structures” (CIVIL-476) ix
taught by Eugen Brühwiler which includes historical examples of works by famous engineers.
At the TU Delft in the Netherlands, to receive a master’s degree in building engineering, the
module CIE4202 “Architectural History of Buildings” is mandatory for all students.x It empha-
sizes the period from 1840 to 1970 and discusses the reuse of existing buildings.xi Historical
aspects are also included in other modules. According to the survey responses, there are mas-
ter-level classes on construction history being taught in Portugal. The same applies to Spain.
However, no specific examples were provided. As reported in “Construction History in the
United Kingdom” (2015) by Bill Addis and James W.P. Campbell,xii the academic teaching of
construction history—especially in engineering—is still not widely established despite much
research and many publications from enthusiasts pursuing this specific interest alongside their
regular professional life or during retirement.

Existing Structures
Given the rising number of interventions necessary on existing and historic structures either
for assessment, preservation, rehabilitation, strengthening or modification, this has become an

Ecole des Ponts ParisTech Web site (http://www.enpc.fr/sites/default/files/files/DE/GCC/Cursus-gcc-archi_2012-2013.
pdf) as of 8 September 2015.
(“Ten professors appointed at ETH Zürich”, published on 11 March 2016).
Course book available on the EPFL Web site at http://edu.epfl.ch/coursebook/en/aesthetics-of-civil-structures-CIVIL-
476?cb_cycle=bama_cyclemaster&cb_section=gc (as of 29 July 2017).
http://studiegids.tudelft.nl/a101_displayCourse.do?course_id=32441 (viewed on 29 July 2017).
http://studiegids.tudelft.nl/a101_displayCourse.do?course_id=12914&SIS_SwitchLang=en (viewed on 29 July 2017).
Chapter published in Ref. [3], work in progress, was available online at http://www.histoireconstruction.fr/rap
port2015/ and is scheduled to be published by Éditions Garnier in 2017.

increasing part of the portfolio not only for architects but for civil/structural engineers as well.
The tasks often necessitate knowledge of historical construction materials, codes, methods, etc.
and thus is the most practical way for practicing engineers to come into contact with construc-
tion history. The survey therefore asked questions about how existing structures are being taught
to engineering students (Question 2.4) on one hand, but also how this is recognized by a more
general public through awards (Question 2.9) and by general knowledge (Question 3.3).

Existing Structures in University Education

The need for the specialized knowledge necessary to deal with interventions on existing struc-
tures seems to have been recognized by academic institutions in many parts of the world. The
following examples were reported by the survey:

• Austria: The Faculty of Civil Engineering at Vienna University of Technology offers a mod-
ule on “Maintenance and restoration of buildings”.
• France: the Conservatoire National des Arts et Métiers (CNAM), a technical school for
engineers, offers a module originally entitled “Connaissance du bâti ancient techniques de
rehabilitation” and now called “Ecoréhabilitation”xiii dealing with construction techniques
for structures built before 1948.
• Germany:
– At Brandenburg Technical University (Cottbus-Senftenberg), there is a chair of “En-
twerfen, Bauen im Bestand” dealing entirely with the design and execution of interven-
tions in existing structures, though it is located within the Institute of Architecture and
not civil engineering.xiv
– The Institut für Bauwerkserhaltung und Tragwerk (founded in 2003) at the Technische
Universität Braunschweig offers two modules on “Bauen im Bestand”xv for a specializa-
tion in building preservation at the master level for civil engineering students.
• Switzerland: At the master level, the EPFL offers two modules on existing structures
(CIVIL-436 and -437),xvi both taught by Eugen Brühwiler.
• The United Kingdom: The following institutions offer postgraduate courses on building
– Cambridge University—MSt Building History
– Bath University—MSc Conservation of Historic Buildings
– University College London—MA in Cultural Heritage Structures

CNAM Web site (http://formation.cnam.fr/rechercher-par-discipline/ecorehabilitation-207844.kjsp), visited on 8
September 2015 and 29 July 2017.
See department Web site at https://www.b-tu.de/fg-bauen-im-bestand/ (in German only, viewed on 29 July 2017).
altung/bauen-im-bestand and https://www.bauwerkserhaltung.tu-braunschweig.de/index.php/menu-lehre-de/
mastervertiefung-bauwerkserhaltung/bauen-im-bestand-ii (in German, viewed on 29 July 2017).
Course books available on the EPFL Web site at http://edu.epfl.ch/coursebook/en/existing-structures-basis-CIVIL-436
and http://edu.epfl.ch/coursebook/en/existing-structures-selected-topics-CIVIL-437 (as of 29 July 2017).
Further courses may be found at https://www.postgraduatesearch.com/pgs/search?course=building-conservation
(URL verified on 29 July 2017).

– Oxford Brookes—MSc in Historic Conservation

– University of Edinburgh—Architectural History & Theory
– University of Central Lancashire—MSc Building Conservation & Regeneration
– Bartlett School of Architecture—MPhil/PhD Architectural History & Regeneration

Without naming specific examples, dedicated modules were reported also from Belgium,
Mexico, Portugal and Spain, while the topic is included in other classes in countries like Finland
or the Netherlands.

Existing Structures in the Public Eye/Awareness

Interventions on existing structures—especially when they are part of the local cultural herit-
age—can often be contentious. Thus, special awards for these kinds of projects also show the
importance that is attributed to them by a more general public. The responses to Question 2.9 on
these special awards showed that there are dedicated awards in Brazil, Bulgaria, France, Mexico,
the Netherlands, Portugal, Spain and the United Kingdom. In countries like Austria, Canada, the
Czech Republic and Switzerland, these interventions are generally recognized through more
general awards. Belgium, Finland, India and Liechtenstein do not seem to have any awards. In
Germany, there are architectural awards for interventions on existing and heritage structures xviii
but no major dedicated engineering awards. The latter are recognized generally through estab-
lished awards. In March 2016, the German Bridge Award (Deutscher Brückenbaupreis) in the
category for road and rail bridges was awarded to the rehabilitation project for the Kocher
Viaduct.xix An honorable mention was given to the rehabilitation of the Mettlach Suspension
Bridge at the 2015 Ulrich Finsterwalder Structural Engineering Awardxx (Ulrich Finsterwalder
Ingenieurbaupreis), while the 2013 edition gave a mention to the rehabilitation of Hamburg
central station. The newly created German Structural Engineering Award (Deutscher Ingenieur-
baupreis) was awarded for the first time in 2016, and projects involving existing structures are
explicitly allowed to compete.xxi The survey’s Question 3.2 asked to name some specific and
notable examples of projects involving existing structures. The following were named, though
the following list is not to be viewed as comprehensive of the respective country’s achievements:

▫ Austria
– Gasometer Wien (www. Gasometer.at)
– Kabelwerk Wien (www.kabelwerk.at)
– Ankerbrotwerke Wien (www.loftcity.at)
– Museumsquartier in den ehem. Hofstallungen (www.mqw.at)
– Burg Perchtoldsdorf (www.burgperchtoldsdorf.at)

Examples: BDA Preis Bayern 2016 “Bauen im Bestand Denkmal” (http://www.bda-preis-bayern.de/nominierungen/
bauen-im-bestand-denkmal.html retrieved on 7 April 2016) or “Respekt und Perspektive“Bauen im Bestand Preis 2014,
awarded by db Deutsche Bauzeitung (http://www.db-bauzeitung.de/aktuell/db-veranstaltungen/respekt-und-perspek
tive-bauen-im-bestand-preis-2014/ retreived on 7 April 2016).
http://www.brueckenbaupreis.de/preis-2016/preistraeger-2016/ retrieved on 7 April 2016.
http://www.ernst-und-sohn.de/ulrich-finsterwalder-ingenieurbaupreis?tab=tab252#rs retrieved on 7 April 2016.
html retrieved on 7 April 2016.

▫ Belgium
– Widening the masonry railway viaduct over the Pede valleyxxii
▫ Bulgaria
– National Museum Complex (Sofia)
– Building of the museum of Sofia’s history (Sofia)
– Museum of the contemporary art (Sofia)
– Trade center “Star Gallery” (Plovdiv)
– Hotel “Ventura” (Varna)
– Ethnographical Museum (Cross-shaped barracks in Vidin)
– Early Christian Basilica of Joan (Sandanski)
▫ Canada
– Rehabilitation of Lions Gate Suspension Bridge (replacement of the entire suspended
structure). Designed by Buckland & Taylor
– Rehabilitation of York Boulevard High Level Bridge. Designed by Delcan (now Par-
– Rehabilitation of the Thousand Islands Bridges. Designed by Delcan/Parsons
▫ The Czech Republic
– Reconstruction of the hall Masaryk railway station in Praguexxiii
▫ Liechtenstein
– Rehabilitation of the headrace tunnels for the Samina power plant
▫ Mexico
– Rehabilitation of the Mexico City Cathedral
– Rehabilitation of the Old Temple of San Agustin in Mexico City
– Rehabilitation of the Old School of Medicine in Mexico City
▫ The Netherlands
– De Meelfabriek (http://www.demeelfabriek.nl/)
– De Hef Bridge (http://dehef.nl/)
– ‘Van Nelle’ Building, Rotterdam
– Rehabilitation of Muider ridge across Amsterdam-Rhine-canal
▫ Switzerland
– Strengthening of the Chillon Viaducts (2.1 km)
– Restoration of several bridges by Robert Maillart
– Railway bridges over the Rhine at Waldshut and Eglisau: upgrading of two riveted
bridge structures from the 19th century

See also Ref. [4].
http://www.konstrukce.cz/clanek/prestavba-nosne-konstrukce-dvorany-masarykova-nadrazi-v-praze/ retrieved on 7
April 2016.

– Widening of the Sierre tunnels

– Raising of Nant de Drance dam
▫ The United Kingdom
– Saint Pancras Station, London
– Bankside Power Station conversion to Tate Modern art museum, London
– Liverpool docks buildings conversion to museums and commercial buildings
– Spitalfields Market, London
– Salt’s Mill near Bradford converted into an art gallery, commercial workshop space,
bookshop and restaurant

Many other projects were submitted; however, these were projects of new and recent structures
and not interventions on existing structures. They were therefore omitted from the above listing.

Cultural Heritage
Four questions of the survey (2.5 through 2.8) tried to establish whether works of civil and struc-
tural engineering are recognized as part of the structural heritage in their respective countries. In
all countries covered by the survey, there exist governmental agencies which list structures (built
environment) as part of their cultural heritage and it can be safely assumed that most countries
do. However, the other questions tried to determine if and to what degree works of structural
engineering are included on these lists and thus considered worthy of preservation and protec-
tion. While all responses confirmed that engineering structures are included in the lists, the types
of structures included in the cultural heritage vary greatly. Buildings are evidently listed in all the
countries surveyed but the responses (due to the nature of question) do not differentiate whether
these buildings are listed for the architectural merit or because of structural aspects (Table 2).

In addition to the survey it should be noted that cultural heritage lists in Spain do include
bridges and most likely other types of engineering structures as the database of the Spanish
Ministry of Education, Culture and Sports lists 138 entries for the keyword “puente” and 29
for “acueducto”.xxiv Germany also lists engineering and industrial heritage structures such as
bridges, factories, etc. In addition to the government listings, the Federal Chamber of Engineers
(Bundesingenieurkammer) has been naming their own historic engineering monuments (“His-
torische Wahrzeichen der Ingenieurbaukunst”), listing a total of 18 structures since 2007.xxv The
American Society of Civil Engineers (ASCE) has also created a list of over 200 historic civil
engineering landmarks both nationally and internationally.xxvi These include Hoover Dam, the
Boston Subway and the Brooklyn Bridge (all United States), as well as the Eddystone Light-
house (UK), Hagia Sophia (Turkey) and Salginatobel Bridge (Switzerland).

Museums and Exhibits

Through museums and exhibits, awareness of the topics relevant to the history of construction
for the general public can be increased, and survey Questions 2.11 through 2.14 were intended

Searches performed on 29 July 2017 via http://www.educacion.es/bienes/buscarBienesInmuebles.do.
http://wahrzeichen.ingenieurbaukunst.de/ as retrieved on 7 April 2016.
http://www.asce.org/landmarks/ as retrieved on 7 April 2016.

Country Types of structures listed Ensembles

Austria Bridges, tunnels, buildings Yes
Belgium Bridges, buildings No
Brazil Buildings No
Bulgaria Bridges, buildings, clock towers, residential towers, No
industrial facilities, mills
Canada Bridges, buildings N/a
The Czech Bridges, buildings Yes
Finland Buildings, stadiums No
France Bridges, tunnels, buildings, dams, dykes, etc., light-houses Yes
India Buildings, caves, forts, monasteries, temples, monuments Yes
Liechtenstein Bridges, buildings No
Mexico Monuments, churches, monasteries, public works Yes
The Netherlands Bridges, buildings, dams, dykes, etc., fortifications, shipyards, Yes
locks, sluices, weirs
Portugal Bridges, buildings Yes
Spain Buildings No
Switzerland Bridges, tunnels, buildings, dams, dykes, etc., retaining walls, Yes
galleries, roads, railways
The United Bridges, canals, aqueducts, tunnels, docks, buildings, Yes
Kingdom fortifications, dams, dykes, etc.
Table 2: Types of listed engineering structures per country

to determine this for each country. Not counting individual structures which have been con-
verted into a museum and display their own history, there are currently no museums dedicated
to the general history of construction or civil/structural engineering. The Museum of Public
Works in Paris (“Musée national des Travaux publics”), which was located at 1, avenue d’Iéna
in Paris, is the only museum that could be classified as such a dedicated museum.xxvii From 1939
to 1955, it showcased around 400 models of bridges, roads, dams, locks, ports, etc. in a didactic
and pedagogic exhibit. The museum eventually had to close because of low visitor numbers.xxviii
However, there are still a number of museums in the surveyed countries which have museums
with a more general focus that in part cover topics related to the history of construction or archi-
tecture. These include but are not necessarily limited to:

▫ Austria
– Ziegelmuseum Wien
– Technisches Museum Wien (www.technischesmuseum.at)
– WienMuseum (www.wienmuseum.at)
– ArchitekturzentrumWien (www.azw.at)

http://www.planete-tp.com/article.php3?id_article=1146 (retrieved on 10 September 2015).
https://fr.wikipedia.org/wiki/Mus%C3%A9e_national_des_Travaux_publics (retrieved on 8 April 2016).

▫ Bulgaria
– Museum of Bulgarian Architecture, Sofia
– Ethnographical Museum, Gabrovo
▫ France
– Cité de l’architecture et du patrimoine, Paris (www.citechaillot.fr)
– Musée des arts et métiers, Paris (www.arts-et-metiers.net)
– Pavillon de l’Arsenal, Paris (pavillon-arsenal.com)
▫ Germany
– Deutsches Museum, Munich (www.deutsches-museum.de)
– Deutsches Architekturmuseum, Frankfurt (www.dam-online.de)
– Deutsches Straßenmuseum, Germersheim
▫ The Netherlands
– Historical forts: The Dutch Waterline (www.hollandsewaterlinie.nl)
– North Sea Flood of 1953: Watersnoodmuseum (www.watersnoodmuseum.nl)
– Pumping Stations:
◦ Ir. D.F. Woudagemaal (www.woudagemaal.nl)
◦ Museum De Cruquius (www.museumdecruquius.nl)
– Fortifications: Kazemattenmuseum (www.kazemattenmuseum.nl)
– Industrial Heritage: Hoogovensmuseum (www.hoogovensmuseum.nl)
– Dredging: Nationaal Baggermuseum (www.baggermuseum.nl)
▫ Portugal
– Museu de Lisboa, Lisbon (www.museudelisboa.pt)
– Museu Nacional Ferroviário, Entroncamento (www.fmnf.pt)
▫ Switzerland
– Swiss Science Center Technorama, Winterthur (www.technorama.ch)
▫ The United Kingdom
– Science Museum (www.sciencemuseum.org.uk)
– British Museum
– Building Centre
– Museum of London
– Transport Museum
– Kirkaldy Museum Southwark
– Menai Heritage Museum
– Ironbridge Gorge Museums
– SS Great Britain
– Brunel Centre
– Waterways Museum

– Fakenham Museum of Gas & Local History

– Amberley Chalk Pits Museum
– Weald and Downland Museum (www.wealddown.co.uk)

Aside from permanent exhibits in the aforementioned museums, there are also temporary/travel-
ling exhibits (held at the above or in other locations) on the topics related to construction his-
tory or on the works of individual structural engineers in the above and other countries. A few
examples are:

▫ Belgium
– 2011: “Bruxelles, sur le traces des ingénieurs bâtisseurs”,xxix held at the CIVA, invited
to explore the city of Brussels through the works of civil engineers.
▫ France
– 1997: “L’art de l’ingénieur”, held at the Centre Pompidou in Paris from June to Sep-
tember, was a landmark exhibit showcasing some of the models of the former Musée
national des Travaux Publics as well as recent achievements in bridge and structural
engineering around the world. The catalogue was a seminal “dictionary” of terms, engi-
neers and structures.
– 2013–2014: “Auguste Perret – Huit chefs d’œuvre! /?”, held incidentally in the build-
ing which he designed for the former Musée National des Travaux Publics, the Palais
d’Iéna. It showcased eight works of this French architect who was seminal in concrete
construction (www.exposiƟonperret.fr).
▫ Germany
– 2009: “Fritz Leonhardt: Die Kunst des Konstruierens”, travelling exhibit on the famous
German engineer.
– 2011: “Félix CandelaKünstler der Konstruktion”, exhibit at the TU Berlin, translated
from a Spanish exhibit (see below) and a set of lectures on shell structures. Also trav-
elled to Munich.
– At the Deutsches Technikmuseum Berlin, a lecture series entitled “Praktiken und Poten-
ziale von Bautechnikgeschichte” (practices and potentials of the history of construction
technology) is held every year and co-sponsored by the VDI working group on construc-
tion technology as well as the BTU Cottbus-Senftenberg.xxx
▫ Italy
– 2010: “Pier Luigi Nervi – Architettura coma Sfida”, exhibit held in Venice and several
other cities across Italy, accompanied by a catalogue.
▫ Spain
– 2010: “Félix Candela: La Conquista de la Esbeltez”, exhibition in Madrid on thin con-
crete shells by Candela, repeated 2013 in Segovia.

A description of the catalogue is available at http://www.ulb.ac.be/wserv2_oratio/oratio?f_context=unibooks&notei
d=606&style=&f_type=view&data-file=bib1 (retrieved on 8 April 2016).
The schedule is available at https://www.b-tu.de/fg-bautechnikgeschichte/wissenstransfer/vortragsreihe.

– 2014–2015: the Spanish group of IABSE is co-organized an exhibit at the Palacio de

Cibeles (Madrid) on contemporary structural engineering (1964–2014) in Spain.xxxi
▫ Switzerland
– The Gesellschaft für Ingenieurbaukunst regularly organized special exhibits about indi-
vidual engineers, offices, companies or topics.xxxii In the past events, the society has held
or sponsored exhibitions on Robert Maillart, Heinz Isler, Christian Menn, Alexandre
Sarrasin, concrete structures, timber structures, railway bridges, hydraulic structures
and tunnels.
▫ The United Kingdom
– “Unseen Hands – 100 Years of Structural Engineering”, Victoria & Albert Museum.
– The Institution of Civil Engineers has held exhibitions on Telford, Brunel, Rennie,
Bazalgette, Ove Arup, etc.
▫ United States
– The Princeton University Art Museum has held exhibits based on research done by Prof.
David P. Billington, his successors and students:
◦ 1972—Bridges & Sculpture
◦ 1974—The Eads Bridge
◦ 1976—The Bridges of Robert Maillart
◦ 1978—The Bridges of Christian Menn (exhibit travelled in USA, Canada)
◦ 1980—Heinz Isler – Structural Artist (exhibit travelled in USA, Japan)
◦ 2003—The Art of Structural Design: A Swiss Legacy
◦ 2008–2009—Felix Candela: Engineer, Builder and Structural Artist
◦ 2011—Fazlur Khan – Structural Artist of Urban Building Forms (hƩp://khan.
◦ 2013—Evolution of German Shells. Efficiency in Form (hƩp://shells.princeton.
◦ 2015—The Art of Spanish Bridge Design including a module entitled “CEE463: A
Social and Multi-Dimensional Exploration of Structures” (hƩp://spanishbridges.

International Congresses

Every three years since 2003, researchers and academics from all affected academic disciplines
gather to meet at an international congress with hundreds of presentations published in the
respective congress proceedings:

Website of the exhibition available at http://www.centrocentro.org/centro/exposicion_ficha/95 (retrieved on 14
September 2015).
The society’s exhibition archive provides a full list at http://www.ingbaukunst.ch/de/veranstaltungen/archiv/
(retrieved on 10 September 2015) as well as exhibition catalogues. http://www.ingbaukunst.ch/de/publikationen/
buecher-zu-ausstellungen/ (retrieved on 8 April 2016).

2003—Madrid, Spain
2006—Cambridge, the United Kingdom
2009—Cottbus, Germany
2012—Paris, France
2015—Chicago, USA
2018—to be held in Brussels, Belgium

There are also national congresses held in Spain, France, Germany, Portugal, the United King-
dom and the United States.

Associations and Societies

Question 2.10 of the survey asked about associations in each country which are either specific
for the discipline of history of construction or related aspects. The following provides a grouped
listing of the societies provided by the survey responses augmented with knowledge available
to the IABSE working group.

Construction History Societies

Most of the respondents actually were not able to name the societies dedicated to construction
history in their respective countries. The current list of construction history societies follows in
the order of the year of foundation.

(a) The Construction History Society (the United Kingdom)

The Construction History Society (CHS) is the central focus for those interested in the
subject of construction history in the UK. It was founded in 1982 and was the first such so-
ciety in the world, by a margin of around ten years. The CHS now has a UK membership of
around 250 and a further 200 or so members worldwide. The first issue of the Construction
History, the Journal of the CHS, journal was published in 1985. This has been published
annually and since 2013, twice yearly. In 2012, it was renamed Construction History, the
International Journal of the CHS. The CHS hosted the Second International Congress on
Construction History at Queens’ College, Cambridge in 2006. Website: hƩp://www.con-
(b) Sociedad Española de Historia de la Construcción (Spain)
Founded in 1996, this Spanish society for the history of construction is the second-oldest
such association worldwide. It holds regular national congresses and was responsible for
holding the First International Congress on Construction History in Madrid in 2003. Web-
site: hƩp://www.sedhc.es/
(c) Associazione Edoardo Benvenuto (Italy)
Founded in 1999, the society was formed to foster research on the science and art of build-
ing in their historical development. Website: hƩp://www.associazionebenvenuto.org/
(d) The Construction History Society of America (USA)
Founded in 2007 as a branch of the UK society, it organizes national meetings and con-
gresses. It also hosted the Fifth International Congress on Construction History in Chicago
in 2015. Website: hƩp://www.construcƟonhistorysociety.org/

(e) Association francophone d’histoire de la construction (French-speaking countries)

The society was founded in 2010 and hosted the fourth international congress held in Paris
in 2012. While headquartered in France, it also covers other francophone countries such as
Belgium and Switzerland. Website: hƩp://www.histoireconstrucƟon.fr/
(f) Gesellschaft für Bautechnikgeschichtexxxiii (German-speaking countries)
Founded in Berlin in 2013, this society also uses language instead of national bounda-
ries to define its membership. It offers researchers and academics from Germany, Austria,
Switzerland, and other countries regular forums for cooperation and exchange of ideas,
methods and research. One of the explicit goals of the society is to establish the history of
construction technology as an academic discipline. Website: hƩp://gesellschaŌ.bautech-
(g) Sociedade Portuguesa de Estudos de História de Construção (Portugal)
SPEHC was founded in 2014 and fosters research and academic exchange on the topic in
Portugal. Website: hƩp://www.spehc.pt/

Other Organizations

The following are additional associations, societies, government departments and other organi-
zations that deal with aspects related to construction history such as preservation of cultural her-
itage, have a vested interested in the topic, or foster the appreciation and awareness of structural
engineers and their work both historically and contemporaneously.

▫ Austria
– Bundesdenkmalamt, Austrian Federal Monuments Office (www.bda.at)
▫ Bulgaria
– National Institute of Immovable Cultural Heritage (www.ninkn.bg)
– Bulgarian National Committee of ICOMOS (icomos-bg.org)
▫ Canada
– Ministry of Citizenship and Culture (Ontario)
▫ Finland
– National Board of Antiquities (www.nba.fi)
▫ France
– Association pour la Connaissance des Travaux Publics (ASCO-TP) (www.planete-tp.
– Association Eugène Freyssinet (efreyssinet-association.com)
– Association des Descendants de Gustave Eiffel (www.gustaveeiffel.com)
▫ Germany
– Bundesingenieurkammer (bingk.de)
– docomomo Deutschland (www.docomomo.de)

The name of the society translates as “Society for the History of Construction Technology”. One of the reasons for
this choice was that a possible German translation for “Construction History” would be “Baugeschichte” but this term
is already established in architecture and would correspond to “architectural history” elsewhere.

– Ingenieur Baukunst e.V. (www.ingenieur-baukunst.de)

– Koldewey-Gesellschaft (www.koldewey-gesellschaft.de)
▫ The Netherlands
– Locks and Dams: Stichting Historische Sluizen en Stuwen Nederland (www.sluizenen-
– Royal Netherlands Society of Engineers: Department of the History of Technologyxxxiv
– Cultural Heritage Agency of the Netherlands (culturalheritageagency.nl and cultureeler-
– Bridges: Nederlandse Bruggenstichting (www.bruggenstichting.nl)
– Water Towers: Nederlandse Watertoren Stichting (www.watertorens.nl)
– Pumping Stations: De Nederlandse Gemalen Stichting (www.gemalen.nl)
– Stichting Bouwhistorie Nederland (www.bouwhistorie.nl)
▫ Portugal
– Ordem dos Engenheiros
▫ Switzerland
– Gesellschaft für Ingenieurbaukunst (www.ingbaukunst.ch)
▫ The United Kingdom
– Association of British Transport & Engineering Museums (ABTEM) (www.abtem.
– IStructE History Group (www.istructe.org)
– Construction History Society (www.constructionhistory.co.uk)
– Institution of Civil Engineers (www.ice.org.uk)
◦ ICE Panel for Historical Engineering Works (PHEW)
– Newcomen Society (www.newcomen.com)
– Association for Industrial Archaeology (www.industrial-archaeology.org)
– The Twentieth Century Society (www.c20society.org.uk)
– Victorian Society (www.victoriansociety.org.uk)
– Georgian Group (www.georgiangroup.org.uk)
– Society for the Protection of Ancient Buildings (www.spab.org.uk)
– International Committee for the Conservation of Industrial Heritage, UK.
▫ United States
– Society for the History of Technology (www.historyoftechnology.org)

Activities in Professional Associations

Related activities within professional associations including IABSE but outside of this working
group should also be noted and mentioned even if they were not part of the survey.

The department’s website is https://afdelingen.kivi.nl/geschiedenisdertechniek/PAG000002476/Home.htmlCultural.

IABSE itself published another SED (no. 12) in 2010, entitled “Case Studies of Rehabilitation,
Repair, Retrofitting, and Strengthening of Structures”. Although these examples focus primarily
on the details of the repair and strengthening methods, construction history is also a part. The
fixed format of the contributions allows for the addition of further examples at a later date in the
online version. The fib (International Federation for Structural Concrete, www.fib-international.
org) was reorganized in 2015, which resulted in the creation of Commission 3 “Existing Con-
crete Structures”, currently chaired by Stuart Matthews. The fib Model Code 2010 includes a
chapter “Conservation of Structures”, and the planned Model Code 2020 is envisaged to deal
extensively with existing structures. Moreover, a task group “History of Concrete” which is
chaired by Manfred Curbach as part of Commission 1 “Concrete Structures” is working on a

Overall Evaluation and Summary

In order to provide a full overview of the status of construction history around the world—or at
least IABSE member countries—the number of responses to the survey sent out by the working
group were unfortunately too small. Thus, the overview was augmented where sensible with
information available to the working group through other sources. While the questions did not
always solicit the intended types of responses, the most obvious conclusion is that there is still
quite a way to go to establish construction history and obtain recognition as a field or discipline
of academic study. While some countries like the United Kingdom, Spain, Germany, France,
Switzerland and the United States have already achieved much in terms of research in the field,
by hosting international congresses on the topic, there is still much work ahead to establish
the field firmly at universities as is the stated the goal of the society for German-speaking his-
torians. In most countries, however, construction history is not established at all. In fact, the
lack of survey responses may point to a general lack of interest by structural engineers in the
history of their own profession and its achievements. It unfortunately also indicates that most
engineers do not see construction history as relevant to their daily work. Part of the problem
may be that construction history is an interdisciplinary field which involves historians, art his-
torians, architects, archaeologists, economists and engineers and often requires knowledge or
research not limited to a single field. While art and architecture history has been established for
much longer, engineers most often still only show interest in their own profession’s history only
once retired as the relevance to current practice is often ignored or misunderstood. However, at
least in some countries this seems to be changing. Only recently for example, the Institution of
Civil Engineers in the United Kingdom created the Conservation Information Resource for Civil
Engineers (CIRCE).xxxv This is a purely online resource intended to help practicing engineers
with work on existing structures as this increasingly becomes a part of the portfolio of structural
engineers. With major interventions on existing structures becoming ever more common, it is
likely that through continued education programs, practicing engineers can be made aware how
knowledge of historic construction materials and processes in design and building can benefit
the design and planning of those interventions.

CIRCE is available online freely to everyone at https://www.ice.org.uk/disciplines-and-resources/best-practice/

[1] Becchi A, Corradi M, Foce F, Pedemonte O (eds), Construction History Research Perspec-
tives in Europe (ISBN 88–88,479–11-2), Kim Williams Books: 2004 (available in open
access at http://www.kimwilliamsbooks.com/titles/out-of-print-books/97-construction-
[2] Proceedings of the Institution of Civil Engineers – Engineering History and Heritage (ISSN
[3] “L’histoire de la construction, Un méridien européen / Construction History, A European
Meridian” 2015.
[4] Schotte K, Stael D, Nagy W, de Pauw B, De Backer H, Van Bogaert Ph. Structural Assess-
ment of the Integrated Steel Fly-overs Widening the Historic Multiple-arch Concrete Via-
duct over the Pede Valley. IABSE Report, vol. 99. 2013; 276–277.


(Re)constructing History—How
Building Archaeology Can Profit
from the Knowledge of Engineering

Max Johann Beiersdorf, Chair of Construction History and Structural Preservation, BTU
Cottbus-Senftenberg, Cottbus, Germany

When, in 1924, Armin von Gerkan, a German architect with many years of experience in archae-
ological excavations, first spoke of “building archaeology” as an independent scientific disci-
pline, he placed it on a level with archaeology and its associated sciences (Ref. [1], 9). However,
with a lack of institutional structures in Germany, the budding building archaeologist was forced
to teach himself and gain the necessary qualifications for his further career through practical
activities (Fig. 1). According to von Gerkan, being an architect in itself involves “familiarity
with materials and design, the verifiable feeling for structural relationships and an understand-
ing of working practices – even those of previous ages” (Ref. [1], 10). With no training oppor-
tunities or structures in the early 20th century, von Gerkan called for a clearer demarcation from
other disciplines through the institutionalization of this subject. As early as 1926, von Gerkan’s
efforts led to the foundation of the Koldewey Society, the “association of archaeological archi-
tects”, which campaigned for the training of young architects in the field of building archaeol-
ogy in Germany and for participation in heritage conservation issues.

Although the training situation for building archaeologists has changed noticeably over the inter-
vening years and is now a separate field of study in Germany, the demands placed on graduates
of this subject have changed little. Only the professional field that feeds the discipline of build-
ing archaeology now covers a much wider area owing to the master course of study, as offered
by several German universities and polytechnics. Whereas in the past building archaeology was
exclusively the province of architects, studies in Germany now attract graduates from archae-
ology, art history, restoration, geography and history in equal numbers, likewise construction
engineering and other disciplines with appropriate background knowledge and, last but not least,
landscape and interior architects. In other words, building archaeology courses now have a wide
appeal and are characterized by greater polyvalence. Opening up this professional field to arts
and engineering graduates brings with it enormous advantages because it does justice to the

modern idea of an interdisciplinary scientific land-

scape. The current situation for building archaeol-
ogy leads to heated debates among experts, with
the shortcomings in training at the construction
history chairs of German universities and the job
prospects for young professionals attracting par-
ticular criticism2. This situation appears to contra-
dict the fact that building archaeology has always
seen itself as an auxiliary science and subdisci-
pline, which in conjunction with subjects from
engineering science and the humanities has the
potential to enhance projects involving widely dif-
fering issues. A project investigating undulating
mud brick walls in Egypt will serve as an example
to show the extent to which building archaeology
can profit from the knowledge and methods of
construction history.

Undulating Mud Brick Walls

in Pharaonic Egypt
Fig. 1: Student surveying the Castor and
Pollux temple in Rome. © Sir John Soane’s The undulating mud brick walls of ancient Egypt
Museum, London date from times between the fourth century BC
and the years of Roman rule. They were built to
protect the country’s most important temples, and their constructional features and enormous
dimensions make them an exception among ancient mud brick wall construction (Fig. 2). They
differ from conventional mud brick walls in that they have sections with concave and convex
brick courses and do not employ any mortar within the walls, except the outer facade.

Fig.2: The undulating enclosure wall of the Amun-Re precinct in the Karnak Temple Complex

Auguste Choisy (1841–1909), a French engineer and historian of architecture, was one of the
first to attempt an explanation for the undulating appearance of the walls. He ruled out that
the wavy shape of the walls is the result of brickwork settlings due to a weak or wet sub-
soil. Instead, Choisy suggested that it prevented the panels from shifting. In his opinion, the
organization of a construction site could be crucially optimized by this technique (Ref. [3],
34–37). During the years 1907–1909, Walter Honroth investigated the undulating enclosure
wall of the Temple of Chnum in Elephantine and proposed, according to Choisys idea, that the
concave segments were raised up first and followed by the convex segments in a second step
(Ref. [4], 39–42). Structural assessment of the undulating design has been a major theme of
research from the late 19th to the middle of the 20th century. This changed from 1962, when
Paul Barguet established his doctrine that the characteristic undulating form is an allusion to
the primeval ocean Nun, the origin of all life (Ref. [5], 32). In a more recent scientific article,
Egyptologist Rosanna Pirelli asked whether the design of the enclosures had technical reasons
or rather mythology being the source for their appearance, concluding that the walls gain no
structural benefits from this technique (Ref. [6], 77). So, according to Pirelli, the reason behind
the wavy form is seen as purely symbolic. Henceforth, there have been two camps: one seeing
the design of these walls as serving structural functions and the other, following Barguet, favor-
ing a mythological reading.

Contrasting with this, however, a dissertation at the Chair of Construction History and Struc-
tural Preservation, BTU Cottbus-Senftenberg (W. Lorenz), postulates a primarily constructional
explanation. The work focuses on two main aspects. First, there is to be an investigation of the
extent to which the characteristic undulating form of the walls can be explained by the process
of their construction. Although omitting the mortar inside the walls certainly speeded up the
work, at the same time, however, it weakened the masonry bond. The thesis put forward is that
this disadvantage was supposed to be compensated for by forming concave and convex seg-
ments7. During construction, the concave segments (type 1) were very likely built shortly before
the convex segments (type 2) for structural reasons. Owing to the fact that neither cranes nor
scaffolds were used during construction, the materials had to be distributed via ramps. Ideally,
the wall segments themselves would have served as ramps, and so the wall was built out of these,
piled up, as it were (Fig. 3).

The second focus of the study is the structural analysis of a finished undulating wall. The
individual bricks of a concave segment slide toward the middle, and so shearing apart of the
segments corners is prevented. On the other hand, the convex segments exert pressure on the
adjacent concave segments (Fig. 4). This stress state probably lent the walls greater stability
despite the lack of mortar.

Fig.3: Potential reconstruction of an undulating mud brick wall built in stages


Engineering Science
In order to verify the main theses,
two in-depth studies are proposed.
First, a master thesis in conjunc-
tion with the Chair of Statics and
Type 1 Dynamics, BTU Cottbus-Senften-
concave berg, will investigate to what extent
Type 2
the idea of a state of prestress
convex resulting from the construction
process can be confirmed by more
Type 1 accurate finite element modeling
(FEM). This method of computa-
tion will enable the sliding behav-
Fig.4: Flow of forces within the concave and convex seg- ior of the mud bricks assembled
ments of an undulating mud brick wall without mortar to be determined
based on known material parame-
ters. The results can be analyzed by
comparing them with calculations for conventionally built mud brick walls employing mortar
and horizontal courses. A series of investigations will be carried out together with the Research
and Materials-Testing Institute (FMPA) at BTU Cottbus-Senftenberg in order to determine the
material parameters required. Especially interesting for the later calculations is the determina-
tion of the friction coefficients, the deformation behavior of the mud bricks, the compressive
strength, the flexural tensile strength and the modulus of elasticity. The parameters derived will
permit a calculation of the stress state in the undulating walls which is much more accurate
than can be obtained from literature. The options available to the materials-testing institute and
in FEM go beyond those that building archaeology examinations of a structure can achieve
and considerably enhance the significance of any statements regarding the possible structural
advantages of the undulating walls.

Building Archaeology and Construction

History—A Fruitful Cooperation
The study of a single undulating mud brick wall purely from the point of view of building
archaeology criteria is undoubtedly worthwhile. But combining this with the techniques of con-
struction history, engineering science and archaeology brings about promising synergy effects
and, in the end, new findings. In light of this development, the construction history takes on
a special role: tracing and understanding design, the process of designing forms the heart of
this discipline and, at the same time, the common ground with building archaeology (Fig. 5).
Construction history can provide important momentum when it comes to questions regarding
not only the constructional nature of a structure, but rather its durability and renewability, also
the procurement, processing and use of building materials as well as questions regarding social
consequences and the interdependencies of a structure or the conditions under which those
structures were built.

• Measuring and documentation • Models of structural behaviour

• Identification of building and load paths
phases • Studies of the construction
• Reconstrution process

Building Construction
archaelology history

Archaeology Engineering

• Prospecting, surveying
• Excavation • Structural analysis
• Age determination • Materials-testing
• Material analysis • Ground investigation

Fig.5: Contributions of the disciplines involved in the project and their mutual interactions

Seen in the light of increasing specialization, construction history can exploit the full potential
of classical studies through an interdisciplinary dialogue with building archaeology and its asso-
ciated disciplines. This interdisciplinary approach is in no way alien to building archaeology; in
fact, it is its quintessential feature. At best, the dialogue with construction history allows it to do
justice to Armin von Gerkan’s description once again: developing a scientifically sound feeling
for structural relationships and the working practices of previous ages.

[1] Gerkan A v, & Boehringer E. Von antiker Architektur und Topographie: Gesammelte Auf-
sätze Kohlhammer: Stuttgart, 1959.
[2] Wulf-Rheidt U. Zur Lage der Bauforschung an den Universitäten in Deutschland: Vor-
trag, gehalten auf Anfrage des Deutschen Archäologen-Verbandes e.V. am 15. Juli 2002 in
Greifswald. Accessed 24 March 2016, 2002. hƩp://www.koldewey-gesellschaŌ.de/de/
[3] Choisy A. L’art de batir chez les egyptiens Rouveyre: Paris, 1904.
[4] Honroth W, Rubensohn O, & Zucker F. Bericht über die Ausgrabungen auf Elephantine in
den Jahren 1906–1908. ZÄS 1909; 46: 14–61.
[5] Barguet P. Le temple d’Amon-Re à Karnak: Essai d’exégèse. Recherches d’archéologie,
de philologie et d’histoire 21 Institut francais d’archéologie orientale: Le Caire, 1962.
[6] Pirelli R. Once more on Undulating Walls in Ancient Egypt: Mythological Reasons or
Technical Requirements? In Egyptological Studies for Claudio Barocas, Pirelli R (ed)
Napoli, 1999; 55–95.
[7] Beiersdorf MJ. Undulating mud brick walls in ancient pharaonic Egypt. In Proceedings of the
Fifth International Congress on Construction History: Chicago 3rd–7th June 2015, Friedman D
(ed), vol. 1, 3 vols The Construction History Society of America: Raleigh, NC, 2015; 171–179.


The Many Footprints Left by Martin
Bachmann in Pergamon

Josef Steiner, Struct. Eng., Ingenieurgruppe Bauen (until 2012), Mannheim, Germany.
Corresponding member of DAI (German Archaeological Institute).

The architect and architectural researcher Dr.-Ing. Martin Bachmann, since 2006 second direc-
tor of the german archaeological institute (DAI) in Istanbul, participated in many of the DAI’s
archaeological sites in Turkey. Last July, Martin Bachmann died unexpectedly. This article is
dedicated to him, to his achievements, and to the many visible traces left by him, particularly
in Pergamon.

Building Z on Pergamon’s Acropolis Hill

Since 1926, the year the Istanbul Department was founded, Pergamon has been one of the DAI’s
largest and most significant excavation sites. In addition to the archaeological excavations and
the careful safeguarding of antique buildings in danger of collapse, the institute’s tasks include
long-term protective measures for valuable finds, as well as restorations projects that aim to
offer visitors at least an impression of the original feel of an antique building. One such project
that has been widely discussed among experts is the reconstruction, completed in 1996, of parts
of the Sanctuary of Trajan. Situated on the acropolis hill in Pergamon, the sanctuary is visible
from a great distance. The project leader at this time Klaus Nohlen reports in Ref. [1]. With
Ref. [2], Wolfgang Radt published a comprehensive work on the history, the buildings and
the excavators of Pergamon. At the beginning of the 1990s, during excavations to investigate
antique housing construction on the acropolis hill, remains were also uncovered of a large, two-
story peristyle house. In the north section of the building, excavations revealed very impressive
colorful mosaic floors from the time of the Roman Empire, as well as Hellenistic stucco on a
hill-facing wall. It was immediately obvious that these extraordinary finds must be protected and
presented to visitors in an appealing way. This led to Martin Bachmann’s first large footprint
in Pergamon: the new Building Z. Together with Andreas Schwarting, he showed that it was

Fig. 1: Building Z in 2004 on the acropolis Fig. 2: The impressive interior of Building
hill (Courtesy: J. Steiner) Z (Courtesy: A. Schwarting)

Fig. 3: Well-protected mosaic flooring in Fig. 4: Plan of the Red Hall on the large
Building Z (Courtesy: J. Steiner) Roman terrace (Courtesy: DAI)

possible not only to be an architectural researcher, but also to develop a relevant, eye-catching
concept of modern protective structure and museum architecture (Figs. 1 and 2).

On the antique foundations, a clear building structure was created that not only meets “func-
tional and conservation requirements”,3 but also offers visitors a highly impressive feel for
the space as they are guided through the rooms on walkways to view the rescued works of
art. The walls were mostly constructed from antique stone material; the steel girders of the
lightweight roof construction and the slat-like façade elements were manufactured by a local
metal worker. The building, inaugurated in 2004, has a similarly dominating position half-
way up the acropolis hill as its antique predecessor. An extensive publication documenting
the history of Building Z and its valuable artworks was completed by Wolfgang Radt and
Martin Bachmann just a few months ago.4 It was during the planning for Building Z that I
met Martin Bachmann. Over the last 14  years, this initial contact developed into a coop-
eration between architect/architectural researcher and civil engineer that was characterized
by great mutual understanding. The following is a brief look at some of our joint projects
(Figs. 3 and 4).

Fig. 5: The area around the Red Hall (Courtesy: DAI)

Repair and Conversion of the Southern Rotunda

Next to the Red Hall
The original construction on the acropolis hill comes from Pergamon’s Hellenistic heyday in
the second century BC. In the Roman period, the focus of construction moved to the flat area
at the foot of the acropolis hill. An impressive urban focal point is provided by the Red Hall,
constructed as a temple complex by Emperor Hadrian in the second century AD on a 275 m
by 90 m terrace designed for the purpose. The main building, an imposing construction unique
in Asia Minor, with 2 m thick walls made of small tiles, was first a temple, then a Christian
church in Byzantine times, and later became part
of the town with houses and an olive factory in the
south courtyard. The Red Hall is flanked by two
rotundas, with substructures below and topped by
half domes. After contemporary construction was
discontinued in the 1930s, the DAI was allowed
to use the southern rotunda as a depot for finds.
At the beginning of the 21st century, comprehen-
sive damage documentation by the DAI led to the
introduction of a maintenance project by director
of excavations Felix Pirson (Figs. 5 and 6).

The aim was to repair the damage and to later

Fig. 6: Steel construction for the new ceil- use the southern rotunda for displaying large and
ing above the Roman vaulting (Courtesy: middle-sized finds. The project lead was Martin
M. Bachmann) Bachmann. I met with him on site in December

2005 to discuss the next steps toward solving the

problem. The damage to the antique substructure
ceiling in the rotunda was serious. This ceiling
consists of mushroom-shaped rounded vaulting
with a central pillar and a generatrix in the shape
of a quarter-circle. The vaulting is made mainly of
Roman concrete, is extremely delicate at its upper
outside edge with a thickness of only 180  mm,
and rests horizontally on the ring-shaped, 2  m
thick surrounding wall. About a third of this thin
outside area of the vaulting had collapsed, due to
Fig. 7: Completed new ceiling in the south- the weight of heavy finds. We soon agreed that
ern rotunda (Courtesy: M. Bachmann) the use of modern methods, for example gunned
concrete, would not appropriate for an antique
element. In addition, the aim at antique excava-
tions is always to use simple building methods,
to deploy local craftsmen and to keep costs, for
example for special building techniques, at a min-
imum. My proposal to leave the antique ceiling
in its current state, only stabilizing its damaged
edge while creating a new level with a 300 mm
space in between, met with general agreement.
In 2006, a new depot was created in the cross-
vaulted substructure adjoining the rotunda to the
south, where the many finds could be stored. The
Fig. 8: Rotunda following restoration at uprights and beams of the supporting structure
the time of the work on the large support- with a mono-pitched roof are also elements of
ing figure (Courtesy: J. Steiner) the heavy-duty shelving in the interior, and for
the façade, Martin Bachmann again turned to the
slatted construction used at Building Z, which is
clearly visible from the Red Hall. Once the surface of the mushroom vaulting had been cleaned,
the new ceiling in the rotunda was constructed in summer 2007. Constructed to handle a live
load of 10 kN/m2, the ceiling consists of a central foundation, a steel construction set on it con-
sisting of six steel profiles arranged in a star formation and resting on ledges in the surrounding
wall, and of cross profiles to shorten the ceiling span. Bonded anchors were welded on to lost
shuttering made of steel sheets. Then it was reinforced and the power-trowelled concrete slab
was cast on to it. The ceiling is edged by sheet metal around its perimeter with a gap of 50 mm
from the antique wall, such that it might give the impression of a floating plate, as the six sup-
porting beams in the surrounding wall are not immediately obvious. An opening has been left
in the damaged part of the antique vaulting that allows visitors to view or walk round the lower
floor with its ring-shaped antique vaulting (Figs. 7 and 8).

The next job was to repair the damaged exterior wall facing the courtyard, and the dome over the
rotunda. The wide crack in the wall and its direct continuation in the dome had occurred because
where the ring-shaped antique wall straightened at the point of transition to the courtyard, the
thickness of the wall was reduced by about 50%, thus creating a predetermined breaking point
in the supporting structure. At the thinnest point in the supporting structure, therefore, a single

wide crack appeared early on. The same can be seen in the northern rotunda, which is used as
a mosque. We agreed that this ancient crack did not present a danger to the dome and that there
was no need for any complicated repair work. The surface of the dome was cleared of a thick
layer of soil and protected with lead sheeting. For more details, see Ref. [5]. In the last few
years, we also worked to close the gaps in the cross-vaulting of the substructure under the ter-
race, which adjoins the rotunda.

Assembly of a Monumental Supporting Figure

Next to the Red Hall
The courtyards between the Red Hall and the two rotundas were originally partially covered by
angled or U-shaped roofing. On the courtyard side, the architraves under the beams of the roof
construction did not rest on pillars, as would normally be the case, but instead on monumental
supporting figures in the form of human figures of white marble with arms and heads in dark
marble. The discovery of well-preserved fragments led to the decision to reassemble a support-
ing figure for which a large part of the original material was still in existence. This reassembled
figure would give visitors an idea of the size of such a sculpture and of the enormous space
under the former roof and the impressive sight of the Red Hall. To guarantee the structural
stability of the 8 m high supporting figure in the face of wind and earthquakes, the individual
components—in spite of their great weight—had to be well connected to each other. To do this,
we used threaded stainless steel rods and compos-
ite mortar. The supporting figure was completed
in 2013 and is an additional attraction for tourists
on the Red Hall terrace, who all too often are hus-
tled past the Red Hall and straight to the acropolis
hill by their tour operators. The particularly gen-
erous support for the Red Hall project that was
provided by the Studiosus Foundation needs to be
mentioned at this juncture. The anchoring mate-
rial for the assembly of the supporting figure was
provided by Fischer company (Figs. 9 and 10).

Maintenance of Retaining
Walls on the Acropolis Hill
In view of the settlement of the Acropolis Hill, it
was necessary to divide the area into terraces that
were realized by erecting retaining walls. These
walls, approximately 1–1.5 m thick and up to 8 m
high, are built using multiple leaf masonry. Using
standard geotechnical procedures, their stability
can be verified only with unrealistically high soil
Fig. 9: The supporting figure on comple- friction angles and cohesion values. Due to their
tion (Courtesy: M. Bachmann) heavy self-weight, many of them have withstood

the horizontal pressure of the soil for many cen-

turies, to which they are subjected even when the
cohesion and friction values are high. But sev-
eral of the walls have collapsed, and from time to
time partial collapses still occur today, whether
due to a change in soil pressure, or due to the
effects of water from the slopes, or a combination
of both. Below Building Z on a large terrace is
the upper Gymnasium, the site of the final large
project initiated by Martin Bachmann. The dam-
age to the retaining walls there showed that very
few header bricks had been provided between
the two outer masonry leaves. For walls with this
kind of masonry, however, these header bricks
are of great importance, because they increase
considerably the stability and thus the collapse
load. As the craftsmen repaired the damaged
areas, we consistently ensured that header bricks
were added. The multiple leaf masonry in walls
that looked unsafe but had not yet collapsed were
Fig. 10: Restoration of the annular retain- also linked by drilling into them and adding steel
ing walls between the Odeon and Building rods that were cemented in. This improved their
Z (Courtesy: J. Steiner) stability. We decided—not only for financial rea-
sons—to forego the use of soil clamps or other
civil engineering techniques such as prestressed ground anchors. What is more, in preparation
for such measures it would first be necessary to perform comprehensive geotechnical investiga-
tions to determine the geotechnical characteristics of the soil behind the walls.

Anastylosis of a Palaestra Corner

in the Gymnasium
The large area of the upper terrace of the Gymnasium lies
about halfway up the acropolis hill, but hardly attracts any
visitors. To redress this imbalance, the Pergamon excava-
tion decided in 2012 to initiate an ambitious project. The
main focus of the project is the reconstruction of the north-
west corner of the Palaestra directly in front of the Odeon.
The idea is to exemplify the development in height of this
former hall-like peripheral development from the Roman
period. A basic foundation for such a reconstruction is the
existence of a minimum of 70–75% of the original antique
structural components. Figure 11 shows a computer-devel-
Fig. 11: 3D model of the planned oped model of the reconstructed Palaestra corner made of
reconstruction of a corner of the existing antique and supplementary structural components.
Palaestra (Courtesy: DAI) The completed Palaestra corner will be 12 m high (Fig. 12).

The building site equipment includes

a gantry crane that can carry loads of
10 ton. Due to the difficulty in access-
ing the building site by transport
vehicles, the weight of the individual
parts of the crane had to be limited
to 100  kg. A reliable metal worker in
Bergama made the components and
assembled the crane in autumn 2014.
With the lifting of the 8 ton corner pil-
lar with a heart-shaped cross-section, it
has already passed its first test. Due to
its excellent stability, this pillar is the
main supporting element of the struc-
Fig. 12: Gantry crane for the reconstruction of the
ture. However, for the final proof of
Palaestra corner (Courtesy: J. Steiner)
stability, a structural and dynamic anal-
ysis is required, using a suitable spatial
framework program. This is currently being completed under the supervision of Prof. Gerhard
Eisele at Potsdam University in the form of a master’s thesis. This project was close to Martin
Bachmann’s heart. He fought for it to be realized, as a member of the DAI committee, following
the fundamental conservation requirements. Financial support for the project is provided by the
Studiosus Foundation, but additional external funding will be required to complete the project.
With the tragic death of Martin Bachmann, the project may have lost its leader and its impetus,
but Felix Pirson, the director of the Istanbul Department of the DAI, and all those involved
intend to continue the work as Martin Bachmann would have done, and complete the project as

Final Remarks
The footprints left by Martin Bachmann in Pergamon have become a broad path. He was
involved in particular in preserving the historic wooden houses in Istanbul, and when he talked
about his research in Oenoanda, you almost had the impression that he was making the stones
talk. Martin Bachmann was an excellent advocate of practice-based architectural research, had
comprehensive knowledge of the construction of antique buildings, and favored pragmatic deci-
sions. Architectural research and the DAI have lost one of their best people.

[1] Nohlen K. The partial re-erection of the Temple of Trajan at Pergamon in Turkey. Convers.
Manag. Archaeolog.. Sites 1999; 3(1&2): 91–102.
[2] Radt W. Pergamon: Geschichte und Bauten einer antiken Metropole, Radt W, Bachmann M
(eds) Wissenschaftliche Buchgesellschaft: Darmstadt, Special Edition 2011.
[3] Bachmann M, & Schwarting A. Pergamon Bau Z, Schutzbau über römischen Mosaiken
Universitätsverlag Eckhard Richter & Co.: Dresden, 2005.

[4] Radt W, & Bachmann M. Bau Z in Pergamon, Architektur und Wanddekor Verlag de
Gruyter, 2016.
[5] Bachmann M, & Steiner J. Erhalt und Instandsetzung mit einfachen Mitteln: Der südliche
Rundturm der Roten Halle in Pergamon. Die Bautechnik; 90(9): 593–601.


The Building A of Radio Kootwijk—A
Concrete Building from 1920, Ready
for the Future

Erik Vianen, Struct. Eng., Phd; Vianen Bouwadvies bv, Nuenen, The Netherlands
Ron Spaan, Struct. Eng., Spaan Erfgoedadvies, Doetichem, The Netherlands

Historical Context
The buildings of Radio Kootwijk were developed to establish independent contact with over-
seas foreign countries by the Dutch Government. During World War I, it became clear that
independent contact was important. Therefore, in 1918 it was decided to build this facility in a
remote location in the Netherlands, away from any existing cities. Since its construction, Radio
Kootwijk has been at the center of radio communication for over 75 years (Fig. 1).

Architectural Design
After the decision by the Dutch Government in June 1918, Maria Julius Luthmann was appointed
as the architect for the buildings and surrounding area. The young architect designed the build-
ings in coproduction with Gesellschaft für drahtlose Telegraphie Telefunken. The ground plan
of the broadcasting building was prescribed by the technical installation. Besides the ground
plan, the need for fire resistance was an important aspect considered in the design of the build-
ing. The new building material, reinforced concrete, was chosen for the construction. Luthmann
designed the Kootwijk broadcasting building keeping in mind a previous broadcasting building,
the Nauen project and used more reinforced concrete in the design (Figs. 2 and 3).

Structural Design
Dr. ir. Jan C. Emmen, who had graduated with specialization in reinforced concrete in 1915,
was the structural engineer for this project. During the design of the broadcasting building, Jan
Emmen had to incorporate the requirements of the Telefunken, which heavily influenced the
final design. These requirements were:

Fig. 1: Exterior of Building A in 2015

Fig. 2: Ground plan, 1920 Fig. 3: Cross-section of Building A, 1920

1. The roof structure should not be above 14.50+P, and the maximum height of the trusses
would be 13.50 m.
2. The inner mold should be a large open area within which a 15 ton crane with a 20 m span
could be operated.
3. The reinforcement at the top of the structure should meet the minimum requirements.

With these requirements in mind, Jan Emmen designed the concrete structure of the broadcast-
ing building. The structure was divided into four sections, using vertical dilation joints in the
building from the top until the foundation. This created four separate structures: a tower with
rooms at the front, a broadcasting hall (two sections) and a rear façade with additional areas.
The building with the separated structures is able to minimize the effects of shrinkage and tem-
perature differences.

The Tower
The concrete structure of the tower is an open structure with the columns at the corners of the
tower as the carriers. The dimensions of these columns turned out to be large because of the
stability forces. The arch structures of the roof form a combined structure with the columns.
During the first test period with the Telefunken installation, the reinforcement of the concrete
experienced high electric induction. To prevent further influence of high temperature on the
reinforcement, copper plates were used in the room to create a Faraday cage.

Broadcasting Hall
The concrete structure of the broadcasting
hall was built with three-hinge frames. The
design calculations turned out to be labo-
rious for Jan Emmen, as he did them all
manually. To simplify the work, he calcu-
lated only the symmetric loads for the rein-
forcement of the frames. To support the
perpendicular frames, the support beams
of the roof were used (Fig. 4).

The Rear Façade

The concrete structure of the rear façade is
formed with columns that are clamped to
the foundation, detached from the broad-
Fig. 4: Interior of broadcasting hall in 2012 casting hall structure. This creates the flex-
ibility for the temperature differences in the broadcasting hall structure. The supporting areas
are simple concrete structures.

Realization of the building started on 16 June 1920 by Braat Delft and Internationale Gewapend
Betonbouw I.G.B. Breda. As the building was located in a remote undeveloped area, temporary
access (rail)roads were provided for delivering equipment and material. Temporary housing was
also necessary. Local sand and gravel were used in the concrete structure. During a short period
of preparation, work was started on the foundation. After completion of the basement, plans for
the underground cable network were changed; therefore, the basement was filled with sand. One
year after the first concrete casting (9 August 1920), Building A was ready (16 August 1921). The
exterior of Building A has a characteristic look by virtue of the chiseling of the concrete struc-
ture’s surface. By chiseling the concrete surface, it was possible to hide the wooden boards used
in the structure. The chiseling however created the possibility of developing hairline cracks in
the building. To avoid corrosion of the reinforcement, a 35 mm concrete cover was realized. An
additional effect of this uneven surface was the growth of moss and algae. The final concrete sur-
face would stand for a long period of time, till a rebuilding during 1958–1961. The three years of
remodeling gave Radio Kootwijk a modern look. The construction company used concrete spray-
ing equipment, which sprayed 20 mm of concrete on all exterior walls. The old concrete casting
lines and hammered surface were thus replaced by a sleeker look with fine-grained surface.

Use in 21-Century
On 31 December 1998, use of the building came to a formal end. In the period till 2010, differ-
ent studies for new uses were carried out. The building has now become a national monument,
which makes putting it to new use difficult, considering its value as a monument. Between

2011—2013, restoration of the tiled floor was conducted, and also an elevator for disabled peo-
ple within the existing structure was constructed. The elevator is located close to the foundation
of a tower column, and the lift pit is connected to this foundation. After an investigation of the
concrete structure of the building, it was concluded that the concrete, created almost 95 years
ago, is still in very good condition. In spite of the limited experience and knowledge of rein-
forced concrete during the building period, the structure is still in good condition and the build-
ing capable of standing for a long period of future use.

The Intervention Concept

After 85 years of service with the KPN, Koninklijke PTT Nederland (Royal Dutch telecommuni-
cation company), in 2004 Building A was transferred to the DLG Dienst Landelijk Gebied (Ser-
vice Rural Areas). The DLG maintained
this building for five years until 10 Decem-
ber 2009, when Staatsbosbeheer (Federal
Forestry Services) took over ownership
and management. Staatsbosbeheer and
three other government authorities—the
RCE, Rijksdienst voor Cultureel Erfgoed
(Federal Service Cultural Heritage), the
province of Gelderland and the commu-
nity of Apeldoorn—confirmed their inten-
tion to realize the restoration following the
Fig. 5: View over the area in 2016 (courtesy: Jan principles of cultural heritage as described
van Dalen) in the document “Hallo Bandung…, Hier
Radio Kootwijk” (Fig. 5).

At the moment, both internal and external work has been done for Building A and its two
annexes. The external work concerned the replacement of the roof and the cleaning and
repairing of the crystal cement granite outer wall. The internal work includes restoration of
the tile flooring in the radio hall, repainting the colors of the walls back to the original and
installing an elevator to provide disabled people access to the radio hall. Additionally, several
other rooms were restored and optimized for new use as well. The restoration process is being
coordinated by R. Spaan. A long period of investigation and realization were done under his
supervision, with support from architect Jan van den Burg. Analysis for implementing the
elevator and required modifications to the reinforced structure were done by E. Vianen. Build-
ing A is now ready for its new future and visits can be planned or arranged. Also, events like
concerts, symposia and exhibitions take place and have found a great platform in the radio

Into the Future

The buildings, created almost a century ago, stand tall and can still be used for a long time with-
out any serious modifications. During the use of the buildings, there have been many external
influences, which include demolitions during World War II. Despite the limited knowledge of

the designers about reinforced concrete

structures, this building will stand the test
of time in future. Therefore, it is a great
tribute to the structural engineer Emmen.
With the recently executed restoration,
the building is now equipped for future
long-term use. By changing the surround-
ing area into a public domain, the build-
ing is also more accessible. Therefore,
the former closed character is over. The
building is open on special days for the
public; on other days, the building is in
use for concerts, symposia and exhibi-
Fig. 6: Interior in 2015 tions (Fig. 6).

[1] Spits P. Radio Kootwijk, monument in gewapend beton, ISBN 978-90-75365-92-4
Aneas: ‘s-Hertogenbosch, 2008.
[2] Emmen IJ. De werken in gewapend beton voor het station voor draadloze telegrafie op het
Kootwijksche Zand. De Ingenieur 1923; 12: 213–237.
[3] van der Pluijm C. Radio Kootwijk. Biografie van een zendstation en een dorp in het hart van
de Veluwe BDUMedia: Barneveld, 2014 ISBN 9789087882167.


Marina City—The History and
Restoration of an Iconic Facade

John F. Duntemann, Senior Principal; Brian R. Greve, Associate Principal; Wiss, Janney,
Elstner Associates, Inc., Northbrook, Illinois, USA

The twin towers of Marina City were the tallest reinforced concrete buildings in the world
when they were completed in 1962. The design and construction of Marina City was an impor-
tant milestone during the evolution of concrete high-rise construction during the mid-20th cen-
tury, and the unique modern design served as a model for mixed-use developments that is still
employed today. The construction of the towers utilized innovative design and construction
techniques. Significant concrete deterioration was identified on the facade in the 1990s, and
repairing these conditions on a high-rise building in a dense urban setting presented unique
challenges. This paper reviews the history of the design, construction and restoration of these
iconic towers.1

Marina City in Chicago was designed by Bertrand Goldberg Associates. The twin high-rise tow-
ers are part of a mixed-use complex that consists of five distinct structures including the towers,
a 10-story office building and a theater building. Described as “the City within the City”, the
complex originally included residences on the upper portion of the towers, a parking garage
on the lower floors of the towers, an office building, a marina, a theatre, an ice skating rink, a
bowling alley, restaurants and retail space. While common today, this mixed use of residential,
office and retail with parking was an innovative concept in urban planning when Marina City
was built. Figures 1 and 2 are ground-level views of the Marina City complex and towers, circa
1964–1965 and today. The towers are 63 stories tall (178  m) and were the tallest reinforced

concrete buildings in the world when they

were completed. Each tower has three
plaza levels that connect the buildings, 19
parking levels with 896 parking spaces,
and 40 residential levels that contain 896
apartment units. The circular design uti-
lizes an innovative structural center core
surrounded by two concentric rings of
columns and 16 radial floor beams extend
from the core to the exterior columns, and
curved cantilevered balconies extend out
from each column.

Construction of the towers began in 1960
and was completed between 1963 and
1964. The tower foundations consist of
reinforced concrete caissons drilled to
a depth of 35  m. Three concentric rings
of caissons were used to support the core
wall, inner columns and outer columns.
The concrete core is supported on a cen-
tral ring of eight caissons, and the core
Fig. 1: View of Marina City circa 1964–1965 was cast utilizing a high-strength con-
crete mix design that provided sufficient
strength in 24 h to support continued con-
crete placement above.2 The core construction advanced much faster than the floors as shown
in Figs. 3 and 4.
The innovative construction utilized curved fiberglass forms, which were continuously reused
to increase the speed and ease of assembly. The formwork, reinforcing steel and concrete place-
ment were completed in one day on each floor. The work crews alternated between each tower
to pour a new floor every two days. Figures 5 and 6 show the floor construction in progress.
Concrete was delivered to the site by trucks and lifted by tower crane to each floor level. A spe-
cially designed electric conveyor belt system was used to distribute the concrete on each floor.2
In addition, lightweight concrete was utilized for the beams and floors.

Evolution of Concrete High-Rise Design

The first concrete high-rise building constructed in the United States was the Ingalls Build-
ing in Cincinnati, Ohio. The Ingalls Building was completed in 1903 and was 15 stories tall.
However, following the construction of the Ingalls Building, many high-rise buildings were
constructed using steel, and few concrete buildings exceeding 20 stories were constructed until
approximately 1960. The Marina City towers were the tallest reinforced concrete buildings
in the world when they were completed in 1962, and their construction renewed interest in
using reinforced concrete for high-rises. The use of concrete for tall buildings was initially

limited due to challenges related to the

delivery and placement of the concrete.
The concrete used for the Ingalls Building
was mixed on-site. The ready-mix indus-
try began using revolving-drum mixer
trucks in the 1930s, which significantly
improved concrete delivery by avoiding
remixing of the concrete once it arrived
on-site. Trucking technology continued
to improve and the ready-mix industry
expanded significantly in the 1940s and
1950s in the United States. However,
placement of large quantities of con-
crete remained a challenge until mobile
hydraulic pumps were developed in the
1960s, and the technology continued to

As the delivery and placement methods

improved, concrete material options
also advanced. The use of lightweight
concrete was particularly attractive for
use in high-rise building construction.
The additional material cost of light-
weight concrete is partially offset by the
reduced weight and related cost savings
for the columns and foundation. The
development of high-strength concrete
Fig. 2: View of the towers in 2015 also allowed for smaller column sizes
in high-rise buildings, which increased
the usable floor space. As the concrete
delivery, placement and materials improved, structural
engineers developed a variety of structural systems to
utilize concrete in high-rise construction. The Ingalls
Building utilized a beam-column frame system with
a two-way floor slab system. Shear walls were intro-
duced in 1940 and typically utilized a center core wall
to resist lateral loads. A combined system called a shear
wall-frame interaction system was a significant devel-
opment in the design of tall concrete buildings. This
system utilizes a combination of a central core or shear
walls with the beam-column or slab-column framing in
the building.3

The circular center core of Marina City serves as a shear

wall and was designed to resist seventy percent of the
Fig. 3: Concrete core construction lateral load. Following the completion of Marina City

Fig. 4: Shoring of the concrete floor slabs

in 1962, continued innovation resulted in

the construction of many concrete high-
rises. Fazlur Khan of SOM developed
the framed tube structural system in the
1960s. The tube structure utilizes the
three-dimensional framework of the entire
building to resist lateral loads. The devel-
opment of these structural systems was a
key component that facilitated the design
of tall concrete structures. The DeWitt-
Chestnut Apartment building in Chicago
was completed in 1965. This building is
the first known building engineered by
Fig. 5: Floor slab construction Khan as a structural tube. Khan went on to
develop additional variations of the tube
concept including the tube-in-tube system. The Brunswick Building in Chicago was completed
in 1965 and has a tube-in-tube system consisting of a framed tube with an internal shear wall.
Water Tower Place in Chicago was completed in 1975 with a height of 262 m. The structure con-
sists of a reinforced peripheral framed tube with interior steel columns. The One Magnificent
Mile building in Chicago was completed in 1983 and was one of the last buildings engineered
by Khan. This building utilized a concrete bundled tube system, similar to the steel system used
for the Sears Tower.3

The mass of a reinforced concrete structure is a significant advantage in high-rise building con-
struction due to its increased stiffness and vibration damping characteristics. The increased stiff-
ness and damping help reduce building motion to make building movement less perceptible to
the occupants. Concrete is still the material of choice for many tall, slender towers.

Facade Deterioration and

Significant concrete deterioration was
identified on the facade of the Marina
City towers in the early 1990s. An inves-
tigation revealed that the concrete had a
high chloride content, which contributed
to the corrosion of the reinforcing steel
which, in turn, caused cracking, delami-
nation and spalling of the concrete. The
source of the chloride was from the chem-
ical admixtures used to achieve high-early
strength and accelerate construction.
Major concrete repairs were performed
on the facade and balconies at that time,
and in some cases, extensive full-depth
concrete repairs were required as shown
in Fig. 7. The demolition and repair work
Fig. 6: Placement of concrete floors was performed using suspended scaffold-
ing. The repair areas were sandblasted
prior to concrete placement to clean the
reinforcing steel, remove surface con-
tamination and roughen the concrete
surface. The reinforcing steel was coated
with an epoxy paint to inhibit future cor-
rosion, and supplemental reinforcement
was installed where required to anchor
the repair areas. Polymer-modified con-
crete was utilized for the repair areas. The
advantages of polymer-modified concrete
include low permeability, good bond to
prepared concrete surfaces and increased
Fig. 7: Extensive concrete repair on balcony slab durability. The polymer-modified con-
crete was mixed on-site and placed in
formwork installed on the building facade. Following the completion of the concrete repairs,
waterproofing membranes were applied to a portion of the balconies and an architectural coat-
ing was reapplied to the facade to reduce the moisture content of the concrete, thereby reducing
the ongoing corrosion of the embedded reinforcement.

The facade repair project presented many challenges including suspended scaffolding (swing
stage) access and difficult debris control. The commercial property below including the res-
taurant, hotel, theater and parking garage were open for business throughout the repair project.
Extensive canopy protection was required to permit the continuous use of the commercial
property below as shown in Fig. 8. Several additional facade repair projects have been per-
formed since the initial repair project was completed in 1991; however, the amount of concrete

repairs required has been significantly

reduced. The concrete repair work and
application of waterproof coatings has
significantly reduced the rate of concrete
deterioration and the amount of repairs
required, thereby preserving this impor-
tant architectural and structural engineer-
ing landmark.

The high-rise towers of Marina City are
iconic structures that represent an innova-
Fig. 8: Canopy protection at the base of the towers tive and durable architectural and struc-
tural design. The efficiency of concrete
repairs and application of waterproof coatings have preserved the structure and extended the
useful life of these iconic buildings.

[1] Duntemann J, Greve BR. Marina City—The history and restoration of an iconic façade.
Proceedings, IABSE Conference—Structural Engineering: Providing Solutions to Global
Challenges, 23–25 September 2015, Geneva, Switzerland, 2015; 482–483.
[2] Marjanovic I, & Ruedi Ray K. Marina City—Bertrand Goldberg’s Urban Vision Princeton
Architectural Press: New York, 2010.
[3] Ali M. Evolution of concrete skyscrapers: from Ingalls to Jin Mao. Electron. J. Struct. Eng.
2001, 2001; 1(1): 2–14.


Rehabilitation of the Complex
Reinforced Concrete Shell Roof
Structure of an Industrial Building

Alexander Traykov, Struct. Eng., and Prof., University of Architecture,

Civil Engineering and Geodesy, Sofia, Bulgaria

Brief Description of the Presented Case Including

Project Aims and Challenges
This case study presents the investigation and design of the rehabilitation of a storage structure
for the cement industry and in particular of its roof. The storage was designed and constructed
within a two-year period between 1956 and 1958 as part of a cement plant in the northern part of
Bulgaria. The storage is approximately 500 m long and is divided into seven separate structures
(indicated as blocks in the following text) by expansion joints (Figs 1 and 2).

The main structure is a single-span frame in transverse direction and multiple-span frames in
longitudinal direction. The structural span width in transverse direction is 33 m, and the span
width in longitudinal direction is generally 6 m. The roof structures of three of the storage blocks
are steel trusses with a cover made of trapezoidal steel sheets. The roof of the other blocks is a
reinforced concrete (RC) shell structure. The RC shells span 33 m in transverse workshop direc-
tion and 6 m in the longitudinal direction. The shell structure is thin, elegant and unique for Bul-
garia. It represents efficient implementation of the theoretical advantages of shell structures. The
shell is divided by means of RC arch ribs supported by steel elements (two channels of 180 mm)
for resisting the tension in transverse direction at each column position. The shell is additionally
divided by secondary beams/ribs in the longitudinal direction in order to increase the stiffness
and to minimize the shell thickness. These beams are considered as the final transverse rib of
the shell. The storage has been operational for decades without need for additional maintenance
during the long period of use. Material corrosion and structural damages caused mainly by crane
operation have deteriorated its condition and have raised a reasonable question about the struc-
tural safety and reliable behavior of the structure. A team led by the author of the present study
carried out an extensive investigation of the storage structure. It included visual observation,
laboratory testing of the materials both on site and on samples, and computer modeling of the
entire structure and its parts. Calculations and design checks were carried out according to the

Fig. 1: General view of the building

Fig. 2: Sections of the roof

present-day codes that are valid in Bulgaria. Even so, the roof structure theoretically displayed
perfect behavior. That behavior contrasts sharply to the real condition of the roof. The design-
ers team considered two options—to replace the RC shell roof by means of steel trusses or to
preserve the original roof structure. The owner chose to preserve the original structure.

An investigation that involved collecting specific details for rehabilitation of the main ele-
ments—shell, joints and beams—was carried out. The design was adequate and followed the
main principles of sustainable engineering and construction, preserving a structure of historical
value, obtaining economic advantages and protecting the environment.

Influence of CH Analysis Results on the Project during the

Assessment Phase and in the Design of the Intervention
If the importance of construction history (CH) must be considered in the decision of interven-
tion procedures, the period of construction of the particular structure as well as the main ideas
behind the use of the structure’s type must be taken into account. A comprehensive review

of the theory, modeling, construction and use of various types of shell structures is presented
in Ref. [1]. The application of a particular type of shell structure built of specific material is
considered as a result of complex influence of historical, social, scientific, technological and
economic factors. Different aspects of the construction history of roof shell structures are stud-
ied in Refs. [2–4]. Different aspects of the modeling, calculations and investigation, including
academic, of specific shell types are presented in Refs. [5,6]. There are also a number of studies
on the rehabilitation of concrete shell structures—see for example Refs. [7–10], some of them
with a lot of references to other works.

Project-specific Experiences and Lessons Learned

As discussed in Ref. [3], the construction of thin shell concrete structures ended abruptly
at the end of the 1970s. One reason was the higher cost in comparison with other structural
systems. The other reason was the theoretical uncertainties about the structural behavior of
the shells. However, computational methods and the efficient development of software allow
engineers to model the structural behavior of shell structures in a very realistic way, including
the material and geometrical nonlinearities. Those developments, as well as the developments
in concrete technology and innovations in relation to its materials, have given new life to the
idea of constructing new shell structures and rehabilitating existing concrete shell structures.
The most important lesson learned is that it is possible to extend significantly the life of con-
crete shell roofs taking advantage of their perfect static behavior. That could be achieved by
means of using materials and construction methods that comply with the original idea of their

Choice of Strengthening Method

Significant efforts were devoted to study the construction period of the roof, its structural type
and construction material (Fig. 3). Plenty of information was collected in order to choose the
most appropriate techniques for strengthening. Different options were considered, including the
replacement of the original RC shell structure by means of steel frames roof. Innovative meth-
ods, some of which are presented in Refs. [7, 8], were studied as well. In those works, the appli-
cation of innovative strengthening method
using textile RC is described. The method
consists of applying three layers of fine-
grained concrete and textile carbon fabric,
adhering to a rough, sandblasted concrete
surface. However, strengthening by using
any kind of fiber-reinforced plastic (FRP)
fabric was not approved because of the
large area that is necessary to be rehabili-
tated and the greater requirements during
the construction. Considerations about the
maintenance included the risk of any dam-
ages to the structure as well, for instance
the provision of waterproofing on top of
Fig. 3: The roof from inside the building the structure.

Computer modeling of the entire storage building and a separate model for the roof alone were
created by means of SAP2000 software. The results reveal the amazing behavior of the roof.
The bending moments in the shell structure are practically zero for vertical loading due to self-
weight, dust and snow. The above considerations led to the choice of a most “classic” method
for reinforcement, which appeared most suitable for the project. A layer of small-size aggregate
concrete with a single rebar mesh on a cleaned and rough surface was applied. Anchoring of the
reinforcement mesh to the shell ribs was used for better connection. As a result of the structural
investigation and design, a unique structure has been preserved. It combines theoretical achieve-
ments realized in a clear and simple structural form. Post the rehabilitation, the storage will be
in use for a long period.

[1] Rutten HS. Forty years of theory, design and construction of thin cells. Heron 1986; 31(1).
[2] Booth LG. The design and construction of timber hyperbolic paraboloid shell roofs in
Britain; 1957–1975. Constr. History 1997; 13.
[3] Peerdeman B. Analysis of thin concrete shells revisited: opportunities due to innovations
in materials and analysis methods. Master’s Thesis Report, Delft University of Technol-
ogy, Faculty of Civil Engineering and Geosciences, Delft, 2008.
[4] Ochsendorf J. Engineering analysis for construction history: opportunities and perils. Key-
note Lecture, Proceedings of the Second International Congress on Construction History,
Cambridge, UK, 2006.
[5] Williams Portal N, Lundgren K, Walter AM, Frederiksen JO, Thrane LN. Numerical mod-
elling of textile reinforced concrete. Proceedings of the VIII International Conference on
Fracture Mechanics of Concrete and Concrete Structures, FraMCoS-8, Spain, 2013.
[6] Galant JAL . Cylindrical thin concrete shells. Master of Science Thesis, KTH, Stockholm,
Sweden, 2009.
[7] Ortlepp R, Weiland S, & Curbach M. Rehabilitation and strengthening of hypar concrete
shell by textile reinforced concrete. In Excellence in Concrete Construction through In-
novation, Limbachiya MC, Kew HY (eds) Taylor and Francis Group: London, 2009 ISBN
[8] Weiland S, Ortlepp R, Hauptenbuchner B, Curbach M. Textile reinforced concrete for
flexural strengthening of RC-structures—part 2: application on a concrete shell. Special
Publication, DOI: 10.14359/20149, 2008.
[9] Brühwiler E, Denarie E. Rehabilitation of concrete structures using ultra-high perfor-
mance fibre reinforced concrete. UHPC-2008: The Second International Symposium of
Ultra High Performance Concrete, Kassel, Germany, 2008.
[10] Meleka NN, Safan MA, Bashandy AA, & Abd-Elrazek AS. Rehabilitation of elliptical
parabolic reinforced concrete shells with openings. Asian J. Civil Eng. (BHRC) 2013;


Maintenance and Strengthening of the
Timber Roof Elements in the Church
of St. Dimitar

Marina Traykova, Prof., PhD, Struct. Eng., University of Architecture, Civil Engineering and
Geodesy, Sofia, Bulgaria
Doncho Partov, Prof., PhD, Struct. Eng., University of Structural Engineering and Architec-
ture, Sofia, Bulgaria

Brief Description of the Presented Case, Including Project

Aims and Challenges
Saint Dimitar is a Bulgarian Renaissance church (Fig. 1). It is located in the eastern part of
Kuystendil city. It was built between 1864 and 1866 through the initiative of the Kuystendil-
based teacher Dimitar Stoyanov-Dimitry and carries the name of Saint Dimitar, Miracle
worker of Solun (Thessaloniki). The creator of most of the icons in the church was the famous
Samokov-born artist Ivan Dospevski, a representative of the Samokov Art School. The church
is a three-nave pseudo basilica. The funding for the construction and decoration of the temple
was provided by wealthy citizens, craftsmen, teachers and religious representatives. In 1865,
the second school was founded. The initiator and main teacher was Dimitar Stoyanov-Dimitry.

Most of the buildings constructed until the beginning of the 20th century in Bulgaria are made of
masonry, mainly stone masonry (exterior and some interior walls), and timber elements (floors,
roofs, ceilings, interior and some exterior walls). According to Ref. [1], if properly connected,
these elements promote good global behavior: the masonry walls support the floor beams and
roof trusses, which act as horizontal braces, inducing a more uniform distribution of stiffness
and loading throughout the structure. This type of structure is disseminated all over the country
and represents one of the most applied structures of Bulgarian heritage buildings, justifying the
increasing interest in its preservation as a memory of culture and identity. Unfortunately, most
of these structures are seriously damaged and abandoned, requiring urgent interventions.2 In this
field, and particularly for old timber structures, many authors, respecting the International charts
and ICOMOS,1 recommend techniques using traditional materials such as wood and steel. These
techniques, consisting of the addition of timber elements and steel plates, have been applied in
the rehabilitation of old buildings with very good results.3 One example is the strengthening of

Fig. 1: Current view (courtesy: authors)

the old timber roof structure of the church St. Dimitar (Kyustendil city, Bulgaria) through tech-
niques using traditional materials, which will be presented in this paper.
The church is rectangular in plan, and some of the typical sections are presented in Fig. 2.
The roof structure is made entirely of timber elements. Inside, two rows of timber columns,
passing through the arch of the roof structure and reaching the inclined beams of the roof of
the building, are constructed. On the columns at the level of the floor and above the level of
the roof, ceiling beams and longitudinal beams are situated, respectively, bearing the weight
of other parts of the roof structure. At the level above the stone walls, two unoccupied galler-
ies with timber floor beams are located. At one end, floor beams are supported by the stone
walls (Fig. 2). The inner end of the beams is based on the main longitudinal beam, which in
turn is supported on timber columns. Above this level, the timber columns continue (Fig. 2).
Above the galleries, columns are united using one purlin, which lies on them at the distances
marked on digital axes. The purlins rest on columns by means of a pillow-like bearing. The
slope of the roof (32°) is formed with inclined ribs that lie at the lower end on the stone walls.
Usually, the main goal for a structural engineer is to improve the load capacity of specific ele-
ments, but in the case of timber structures, moisture is one of the most important influential
factors for all the physical–mechanical properties of wood, and it creates favorable growing con-
ditions for agents responsible for wood degradations. Wood is highly susceptible to the attack
of biological agents (fungi and insects). In analyzing this aspect, it is important to realize that
if fungi attacks are linked to the moisture content of the wood, insects can attack in any condi-
tions. According to Ref. [4], the stability of the existing roof timber structure of the church is
commonly studied using the general methodology of structural mechanics, the theory of timber
structures, the theory of materials applied to the wood and the technology of the timber con-
structions. The most common causes of failure of timber structural systems are inadequacy of
configuration (geometry of the structure, sizing of the members, kind of connection of the mem-
bers, bracing, etc.) in relation to the both static and dynamic actions, in addition to slenderness,
instability, defects of the wood laid in place, severe biotic damages, accidental factors and splits.
The strength properties considered in the design of timber elements are related to the bending
moment, tensile, compressive and shear forces. Other strength properties, like torsion, creep or
fatigue resistance, are less important but may also be analyzed.2 According to Ref. [1], the need
for interventions on timber structures is usually related to the existence of damages or alterations
of use, increasing the loads. Consequently, in order to ensure safety and simultaneous proper

Fig. 2: Typical cross-sections of the church (design project drawings, 2015)

performance of the structure, it is necessary to intervene with regard to the damage of structural
elements through rehabilitation. The decision about the type of interventions should be taken
only after a rigorous and careful survey of the structure.3 According to the results obtained in
the survey and the circumstances of each situation, the intervention for a particular element
or structure can take two different paths: rehabilitation or substitution. The rehabilitation of
timber structures can be performed using different techniques. When choosing the techniques
and materials to be used, there are, among others, two criteria linked to heritage protection that
should be respected: compatibility and reversibility.

There are many different ways of using traditional rehabilitation techniques in old timber struc-
tures, namely (a) the fixation of timber pieces or thin steel plates, with varied configurations,

to the sides of the element; (b) the intro-

duction of thin steel plates in the inte-
rior of the element; (c) the installation
of steel belts around the element; and (d)
the installation of new structural elements
(timber or steel) parallel to the existing
elements. Laboratory investigations are
usually undertaken to increase knowledge
of the efficiency of these techniques and to
provide useful information to the design-
ers about the suitability of each tech-
nique.1 The use of timber elements in the
rehabilitation of old timber structures is a
Fig. 3: 3D model of the structure very common solution. On the other hand,
solutions with steel elements are com-
monly used in interventions for old timber structures, particularly for timber floors, leading to
an increase in strength and stiffness. Timber structures show a very complicated deformational
behavior, mainly because of the viscoelasticity property of wood, due to the nature of the mate-
rial and the longitudinal position of the fibers.4

Total strengthening of beams can be achieved with or without increasing their transversal dimen-
sions. To change the cross section, new elements of wood or metal are added above, below or
laterally to the existing members. When retaining the initial beam dimensions, strengthening
involves the insertion of steel profiles, which have a lower height than the actual beam.3 In the
considered case study, the rehabilitation of the timber beam supports was performed through
fixation with a steel triangle and steel bolts. Anchorage of purlins to the outside wall was recom-
mended using steel plates and steel bolts. Strengthening of the damaged structural element with
steel is also achieved using steel plates and steel bolts.3 After the rehabilitation of the supports
and the removal of the heavy load from the roof surface, the existing deflections of the purlins
were reduced through a system of struts. It was also necessary to partially substitute the dam-
aged elements using prosthesis, with the fixed connection between the new and the existing part
of the timber elements using M16 threaded rods and screws.3 Numerical analysis of the timber
roof of the church shows the potential of three-dimensional computer modeling and simulation
of the building structure. However, how the science of engineering and advanced computer
modeling, including finite element method, can be used to increase the understanding of a whole
structure is demonstrated, along with how to apply this method for the analysis of historical
building structures (Fig. 3).

Influence of Construction History Analysis Results on the

Project in the Assessment Phase and in the Design of the
The construction history analysis in this case requires the rehabilitation of the existing building
for future use (as a church) while retaining the original structure. The purposes of the project,
after the assessment of the structure, are to replace, to rehabilitate or to strengthen the damaged
elements with minimal changes to the original position and view.

Project-specific Experiences and Lessons Learned

1. This article demonstrated that the failure of the timber structures is very peculiar. The full
understanding of the different types of failures of an existing structural system can greatly
help to interpret the real behavior. This is the reason the three-dimensional (3D) model is
preferred to perform global analysis.5
2. This structure is typical for National Heritage Buildings and is related to the specific pe-
culiarities of the local construction traditions. It is very important to preserve the original
structure as evidence of the construction tradition of the region and to continue the history
of the place. Special interest in preservation represents the icons in the church. Their high
value is related to one of the famous art schools in Bulgaria, the Samokov Art School.

The original timber structures represent a very significant part of the historical buildings in
Bulgaria. Research on their behavior is the only way to provide the best design details for reha-
bilitation and strengthening.

[1] Drdácký M. Historic roofs and timber frames, state of the art studies. Proceedings of the
ARCCHIP Work, vol. 4, Prague, 2006; 313–531.
[2] Straka, B., Novotný, M., Krupicová, J., Šmak, M., Vejpustek, Z. Konstrukce šikmých střech
Grada Publishing: Praha, 2013.
[3] Mönck W. Schäden in Holz-Konstruktionen, Analyse and Behebung DDR: Berlin, 1987.
[4] Vinař J, & Kufner V. Historické Krovy, Konstrukce a Statika Grada Publishing, 2004.
[5] Petkov M. Numerical Model of the Church (Software “TOWER”), 2015.


Brighton Pier, UK—Innovation
in Renovationi

Nigel Winterbottom, CEng, MICE. Senior Structural Engineer, PM Group, Whiteley,

Fareham, UK

Brighton’s Piers
The Victorian piers remaining around the coast of the UK are iconic structures that represent
more than just the skill of the engineers that built them. After the technical and commercial
challenges of originally building the piers, they have often become an integral part of the local
community and history. Within harsh marine environments, piers will always require mainte-
nance to contribute to their safety, and managing these costs can be fundamental to the survival
of the pier. Today’s engineers play a vital role in helping to sustain these structures. The 537 m
long Brighton Palace Pier (now just known as Brighton Pier) was opened in 1899. Designed by
Richard St George Moore, the pier also featured cast iron screw piles. It has undergone many
changes and events during its life. In 1940, a section was removed as a precaution against enemy
invasion in the war. In 1973, during demolition of the landing stage, a barge collided with the
pier head, and significant damage was caused. In 1995, a large, piled extension to the pier head
was undertaken. The owner, to his credit, has demonstrated a responsible approach to planned
maintenance of the pier structure, realizing that this is necessary to ensure viable long-term busi-
ness on the pier (Fig. 1). Repair work was funded entirely by the private owner and received no
public or charitable funding.

The Boat Deck

The central Victorian core still retains its cast iron piles and columns, although the bracing and
most of the deck structure has been replaced. In the 1930s, an area known as the “boat deck”
was added, approximately 35  m by 15  m on plan. A single-story building was added over

This essay is based on a paper published in Engineering History and Heritage, which should be consulted for full details
in Ref. [1].

this area, probably around the same time.

The 36 columns of the boat deck were
rolled steel joists. In the west–east direc-
tion, the column-to-beam connections
were haunched for stability, whereas in
the north–south direction, the frame had
cross bracing. At some stage, strengthen-
ing of the columns had been undertaken,
with plates welded between the flanges
to form closed rectangular sections. The
area immediately under the pier deck
Fig. 1: Brighton Pier; 537 m long (supported by the beams) was in use as a
workshop and beer store, accessible by a
steel stair to a walkway. A survey of the condition had identified ongoing corrosion of the boat
deck steelwork along with missing bracing and poor state of the concrete slab of the workshop.
The workshop also had a history of damage from wave impact. One of the piles had been poten-
tially identified as broken at the sea-bed level by a dive survey. The structure of the building
above was in a generally good condition. An innovative solution was required to replace piles
and substructure while keeping the pier and entertainment buildings above open during the
restoration. This was achieved by devising a means of working on existing structural elements
under the pier deck.

Concept Design
A concept was established to replace the existing boat deck structure while maintaining
the use of the building above. To keep the existing building supported, use of the existing
structure underneath, rather than any temporary structure, appeared sensible. Although its
condition was poor, the existing structure did not show any immediate signs of distress
and was considered sufficiently robust in the short term. However, the existing structure
itself, particularly given the relatively closely spaced columns, was a significant obstruc-
tion to many forms of access and construction. Piling through the roof of the building was
not favored as this would require a large-reach marine plant with attendant weather risk as
well as disruption to the building use. Piling from within the building would also be far too
disruptive. Rather than seeing the existing boat deck structure as an obstruction, a decision
was made to view it as an asset. Could it be used to support a temporary working platform
under the pier deck? The initial reaction was “Yes”. The structure was already supporting a
workshop and beer store (in addition to the building). If these were removed and replaced
by a temporary working platform, this would not represent significant additional load. With
the Victorian columns at a diameter of 300 mm and the 1990s pier head extension columns
at a diameter of 600  mm, this gave a good idea of the likely diameter of new piles and
columns. The rotary method that seemed obvious was the Victorian solution of a “screw”
pile. The column above could then be fabricated in short sections (to suit the headroom)
and spliced together under the pier. To complete the concept design, the columns would be
cross-braced, and a grillage of steel beams would be used to support the deck and building
above. Finally, the temporary platform and remaining original boat deck structure would
then be removed (Fig. 2).

Horatio’s bar Timber decking
Existing and joists on steel
+12.86 mCD concrete support beams

MHWS Existing
+6.5 mCD walkway
1990s steel pile
MLWN 1930s Victorian
Bed level +2.0 mCD piles cast iron
–2.5 mCD piles

1990s 1930s boat deck Original Victorian

pier head extension extension pier head

(b) Horatio’s bar

+12.86 mCD working platform

+6.5 mCD

400 dia
MLWN columns in
Bed level +2.0 mCD sections
–2.5 mCD

New works

Steel support Horatio’s bar
framework on
new columns +12.86 mCD

+6.5 mCD
MLWN jacket to
Bed level +2.0 mCD columns
–2.5 mCD

Fig. 2: Construction sequence: (a) existing section; (b) new helical piles; (c) construction com-

Survey and Investigation Works

The client had no record drawings for the boat deck structure or building above. Lack of safe
access and the complexity of the existing steelwork meant that conventional surveying tech-
niques were very limited. A decision was therefore made to use a three-dimensional (3D) laser
scanning technique. The scanner was placed on a conventional tripod in several positions around
and under the structure. The scanner generates a “point cloud” of information, each point being
a point on the structure with known 3D coordinates. The information from each location was
stitched together to form a cloud for the whole structure. This was then postprocessed to form
a 3D computer model of the structure. This process enabled geometry to be determined even in
areas where physical access was not possible.

Tender Design
Loading: For the ground floor of the existing building, an imposed loading of 5 kN/m2 was used
(the same as the pier deck is generally designed for). Current and wave loading was based on
a 400 mm diameter column, with an additional allowance of 50 mm for marine growth. As the
design life was 50 years, the design allowed for the sea level rise over this period. Nonbreaking
and breaking wave loads on the columns were calculated using Morison’s equation as the basis.
The total nominal breaking wave load on one column was calculated as 117 kN. Over the 14 m
height of the column, the peak load of 11 kN/m occurred at 2 m below the top. The breaking
wave load was applied to the first (seaward) row of piles, and the nonbreaking wave load was
conservatively applied to the other rows.

Analysis: An elastic 3D frame analysis was used to determine load effects in the columns, piles
and bracing members. One model was established with column fixity assumed at bed level in
order to generate bed-level loads for the subsequent geotechnical pile design. A second model
had pile fixity assumed at a depth of 46 pile diameters below the bed, in addition to an allowance
of 0.5 m for scour. This model was used to determine load effects for the structural design of the
columns, piles and bracing members, as well as to check global deflections.

Member design: Although a pile design for tender was prepared, the final design would be the
responsibility of the contractor. The contractor would also be permitted to submit alternative
tender designs. The tender design was based on a steel helical or “screw” pile. The maximum
working load in compression was 1285 kN, with a coexistent moment of 145kNm at bed level.
The geotechnical design considered the bearing capacity of the helices or skin friction capacity
based on a cylinder equal in diameter to the helix diameter. Seventeen existing building columns
that required resupporting had been identified. The columns supported relatively large areas of
concrete roof slab. Typical nominal dead loads were in the range of 200–300 kN, and this was
similar for imposed loads. Two generic cases required resupporting. The first was where existing
building columns were supported by the top of boat deck columns. In this case, short lengths of
twin new transfer beams would be welded to the sides of the boat deck columns and would be
supported between the new longitudinal beams. In the second case, the existing building col-
umns were supported by existing boat deck beams. Here, the concept was to span a new transfer
beam under and between the new longitudinal beams and provide a packing member between
the transfer beam and the existing boat deck beam. Both concepts involved retention and reuse

of local parts of the old structure. These areas had been inspected to the extent possible, although
it was recognized that full inspection would not be possible until the works commenced.

Approvals by statutory bodies: The pier is a grade II* listed building, and listed building consent
was required for the project in accordance with the Planning (Listed Buildings and Conserva-
tion Areas) Act 1990 (1990) https://www.legislation.gov.uk/ukpga/1990/9/contents (chapter 9).
Heritage issues were discussed, and it was highlighted that it was important to protect the pier
from degradation, that similar materials would be used and that the form of the new construction
was similar to the original Victorian construction. It was highlighted that the use of the tempo-
rary platform, use of prefabricated elements and, in particular, the use of a rotary pile installation
method all minimized the risks to the marine environment. For the formal process, the client
employed a specialist consultant to deal with communication with the council and with English
Heritage. The specialist’s report highlighted that the boat deck structure did not possess particu-
lar architectural or artistic interest and was supported by a designer’s technical report describing
the proposed works and what demolition work was required. The technical report made exten-
sive use of rendered images from the 3D model to help describe the project and show it from
various viewpoints. The application was successfully approved with no conditions.

The contractor established a site compound on the pier. Physical access and weight restrictions
along the pier had been highlighted in the tender documents. The contractor also negotiated
with the local council for a small storage area on the forecourt area of the pier. Deliveries and
removal of materials were generally made out of opening hours of the pier. However, during qui-
eter periods and in agreement with the client, limited access was allowed during opening hours.
The temporary platform was installed using roped access techniques. The platform comprised
a grillage of steel beams supported by a lower level of existing boat deck beams. The platform
had steel open-mesh flooring, and similar flooring from the original walkway was reused to
form edge protection. The platform was designed for an imposed load of 2.5 kN/m2 and a plant
load of 45 kN plus a 20 kN weight of pile (held by the plant) in any single bay. The contrac-
tor’s alternative pile design was a bored composite pile. Like the helical pile, this not only
had the advantage of rotary installation but also offered further benefits. Without the helices,
significantly less torque was required, leading to more confidence in installation with a smaller
rig. Combined with a more conventional design basis and a higher factor of safety, it was also
considered acceptable to remove the need for a preliminary pile, thus further reducing costs.
Following the installation of the transverse and longitudinal beams, steel wedges were used on
top of the longitudinal beams to resupport existing floor beams. For resupport of the existing
building columns, local blast cleaning, inspection and measurement were carried out. Deflection
was monitored through dial gauges, and all the work was successfully carried out, with only a
few millimeters of settlement typically occurring, well within the limits set.

Program and Budget

The original program was for 36 weeks. The final duration was 42 weeks, and completion was
scheduled for 1 October 2012. Two of the additional weeks were attributed to poor weather.
However, if the work had been carried out from marine vessels, longer weather delays would

have almost certainly resulted. The additional four weeks were due to complexities in pitching
and boring early piles, changes in the jacking sequences due to unknown corroded steel and
changes for the retention of the temporary platform. However, the innovative concept of work-
ing under the pier ensured that the increase in duration did not significantly affect the client’s
business. In terms of budget, cost increases were balanced against cost reduction associated with
not removing the temporary platform, and the final cost of £26 million was only 1% over the
original contract sum.

With a need to maintain and protect the heritage value of ageing piers in a harsh marine envi-
ronment, engineers need to find appropriate solutions. This project has demonstrated the use
of several innovative design and construction techniques. The use of laser scanning, temporary
working platform, sectional bored composite piles and fabric abrasion jackets all contributed
to the success of the project and meeting the client’s key requirement of keeping the building
beyond fully operational.

[1] Winterbottom N. Brighton Pier, UK—innovation in renovation. Proc. Inst. Civil Eng., Eng.
History Heritage 2014; 167(EH2): 100–110.


Early Iron Structures at the Hermitage
in St. Petersburg—Unique Testimonies
to Construction History and the
Associated Preservation Problems

Bernhard Heres, Chair of Construction History and Structural Preservation, BTU

Cottbus-Senftenberg, Cottbus, Germany

The State Hermitage Museum in St. Petersburg is one of the world’s most important museums
of fine art. It is housed in a former tsarist palace complex, the buildings of which, with their
baroque and classical façades, are key landmarks in the city (Fig. 1). Less well known, however,
is the fact that these buildings offer a unique tour of one of the most exciting chapters in the his-
tory of construction—the emergence of building with iron in the first half of the 19th century.
The exhibits—in this case hidden from the public—are the many surviving iron structures sup-
porting the floors and roofs. These date from the years 1838 to 1851, and their existence can be
wholly put down to the devastating fire at the Winter Palace in December 1837. Fire-resistant

Fig. 1: The State Hermitage buildings (from left to right): the Large or Old Hermitage (1771–
1787) with the New Hermitage (1839–1851) behind, the Small Hermitage (1764–1775) and the
Winter Palace (1754–1762) (courtesy Sinjaver)

forms of construction were therefore chosen for the immediate rebuilding of the palace. That
meant using iron for the roofs and most of the suspended floors, especially those with long
spans. This decision was then also applied to the building works carried out in the following
years throughout the palace complex, for example, during the conversion of the Small Hermit-
age (1842 and 1850s), for building the New Hermitage (iron structures 1843–1844) and during
the conversion of the Old Hermitage (1850s). The result, in terms of numbers and variations,
is a unique collection of early examples of European construction with iron. Since 2002, BTU
Cottbus and Karlsruhe Institute of Technology have been compiling a systematic inventory of
the items and analyzing them with respect to their significance for the history of construction
and various levels of context in a joint research project.1,2 This research was given particular
practical relevance by the full refurbishment of the roof structure to the Winter Palace, which
was carried out at the same time.

Rebuilding the Winter Palace—A Chance for Experiments

in Building with Iron
Every conceivable means was employed to rebuild the Winter Palace as fast as possible after the
fire, and the fabrication and erection of the iron structures for this prestigious project were largely
completed within a year. Immediately
after the fire, the building commission
responsible appointed Matthew Clark, the
head of the state-owned Alexandrovsky
Works in St. Petersburg, to carry out the
design and construction of prototypes for
the roof and floor structures (for informa-
tion on Clark, see Ref.  [3]). Following
successful loading tests in the spring, fab-
rication and erection work was conducted
in stages up to the autumn of 1838. The
beam types Clark developed for rebuild-
Fig. 2: Elliptical beams from 1838, section and ing the Winter Palace were described by
elevation both Clark himself and contemporary
writers as significant innovations, not just
for Russian iron construction (see Ref. [4]). Even from a modern viewpoint, these solutions are
impressive in various ways. The great number of “elliptical beams” built into floors spanning
up to 15 m represents perhaps the first large-scale use of such lightweight solid-web beams in
building (Fig. 2). Made up of roof sheeting and angles riveted together, their structural effi-
ciency—the ratio of self-weight to load-bearing capacity—is similar to today’s I-sections. With
the help of detailed models, finite element method (FEM) has been used to verify their load-
carrying capacity for the new loads of modern museum operations.5

The beams spanning almost 20 m, designed by Clark for the ceilings over the halls, are still in
place today. This represents an imposing example of the “doubling” of load-bearing structures,
very common at that time, which can be regarded as a way of finding or securing a suitable
solution (Fig. 3). With a lightly cambered top chord made up of iron flats strapped together and
a chain as a bottom chord, the result is a fish-belly-like girder with threaded bars to retain the

form. In addition, an arch rises from spring-

ing points near the walls to support the top
chord at mid-span. The incredible amount
of work needed for the fabrication and erec-
tion of this multicomponent, overall almost
2  m deep, structural system can probably
only be explained by the enormous impor-
tance of this construction project. Different
systems consisting of trussed rafters and
truss-like girders made up from these were
developed to suit the various spans of the
roof structures. With no purlins, these were
closely spaced to carry the roof battens and
Fig. 3: View of the hybrid suspended floor struc- roof covering (Fig. 4). The girder designs
ture from 1838 are very similar to the Polonceau (or Fink)
system, which was developed at the same
time, and are also used over the large halls
and enfilades of rooms. Contrasting with
that, at transitions and along the edges of
the building with its many irregularities,
there are parts of the load-bearing struc-
ture where it is obvious that improvisa-
tion on site prevailed, where the details
also deviate from the standard solutions.
Instead of the outwardly modern pinned
and bolted connections that are already
common, the load-bearing members here
were joined together by, in some cases,
Fig. 4: Typical Winter Palace roof structure—
the very simplest of clamped and push-fit
Throne Hall, 1838–1842
connections—a clear reference to exam-
ples from timber buildings and the very
early examples of construction with iron, specifically in Russia. Thus, the Winter Palace build-
ing site represented a great opportunity for experimentation, with designers and fabricators
devising what seemed to be the best answers in each case in light of the conflicting interests of
tradition, innovation and transfer.

“Watching” the Emergence of Early Building with Iron

The designers of the load-bearing structures for the Winter Palace were treading new ground in
many ways. It is, therefore, hardly surprisingly that defects or damage appeared in some cases.
One part of the research work involved analyzing archive materials, chiefly the findings on site.
This revealed, for instance, various corrections and additions, particularly to the roof structures,
which were implemented not long after erection. The reasons for this were severe deforma-
tions of the rafters as early as the first winter (1838–1839, see Fig. 5). It is therefore possible to
trace how the fabricators tried to solve the associated problem of insufficient buckling strength
of the rafters step by step. Two strategies can be seen here: the arrangement of additional or

more effective lateral bracing and the step-

Add. bracing 1842 by-step development of cross-sections
less prone to buckling, either assembled
from various semifinished products or in
the form of the first T-sections rolled in
Russia. Both strategies were combined in
exemplary fashion in 1842 in the form of
the roof structure, modified by several new
Bracing 1838 components, re-erected over the Throne
Room in the Winter Palace (Figs. 4 and 5).

Further examples allow us to “watch”

Fig. 5: Roof structure of Throne Hall: rafter with the development process of the structure
typical lateral deflections and lateral braces and the details of building with iron. For
1838–1842 instance, comparing just the form of the
roof trusses of the various fabricators
shows that they were already close to the Polonceau system that emerged around 1840 in France
and then later in Germany. However, closer analysis reveals small, yet very distinctive, dif-
ferences that indicate a triangulated truss system, pointing more to intuition than intention.
Evidence for this can be found in, on the one hand, the differences in the arrangements of the
members and, on the other, a critical detail: the continuous bottom chord, fabricated in one piece
with a relatively large cross-section and complicated connections to the struts. The trussing of
round bars and gusset plates so typical of Polonceau trusses did not appear until the final phase
of construction (Fig. 6).

The collection of iron load-bearing structures in the buildings of the St. Petersburg palace com-
plex is a good example of the main stages in construction with iron in the 1830s and 1840s—
from the early “exploratory” phase right up to the establishment of this form of construction.
The chance to be able to trace the development of important elements in design language spe-
cific to structural steelwork in the second half of the 19th century here in one place is what
gives this collection its special value. In addition, many findings of the survey, for example, the
numerous stamps of the foundries that supplied the sections, indicate the significance of Russian
iron production and fabrication in the Ural Mountains since the beginning of the 18th century.
What emerged out of this situation were numerous iron load-bearing structures that constitute
one of the effective lines of tradition here.6

Fig. 6: Evolution of trussing details (from left to right): Clark, 1838 (Winter Palace); Nobel,
1844 (New Hermitage); Baird, 1851 (Old Hermitage)

The Difficulties of Preserving Delicate Historic

Load-bearing Structures
The risk of buckling in the rafter sections and the resulting deformation problems were the rea-
sons for extensive rebuilding work in many areas of the roof of the Winter Palace as early as the
1880s. Many different upgrading strategies were used, ranging from leaving the original mem-
bers intact to complete replacement, depending on the specific situation. Since then, almost all
the load-bearing structures, with their flaws, have fulfilled their function up to the present day.
The reason for the latest full refurbishment was the need to replace the existing roof covering—a
single layer of sheet metal with a number of disadvantages. The new, much heavier roof covering
called for structural calculations. The outcome of these, based on conventional modeling assump-
tions and design methods, called for wide-
spread strengthening. The initial proposals
would have reshaped the existing structure
almost completely and at the same time
resulted in the loss of considerable historic
material. Following a heritage protection
report, during the construction phase, the
two contractors appointed to do the work
modified the solutions that had been origi-
nally envisaged for the roof areas for which
they were responsible, but loss of historic
material was still unavoidable. In accord-
ance with the two concepts implemented,
almost all the load-bearing members above
the rafters were removed—precisely those
many different bracing elements that were
especially important features of the suc-
cessive improvements to the lateral bracing
Fig. 7: Rafters strengthened with large angle sec- back in the 19th century. One of the two
tions both sides contractors strengthened the rafters with
large angle sections on both sides and thus
avoided welding the historic iron material
(Fig. 7). The other contractor strengthened
the rafters by welding small steel flats to
the top surface of each rafter and reused
some of the old roof battens, and there-
fore, the modifications remained much
closer to the look of the original (Fig. 8).
These conversion works reveal the specific
difficulties when dealing with early iron
structures. Their particular weaknesses—
for example, the often delicate members,
Fig. 8: Rafters strengthened with small steel flats like the rafters in this case—represent real
(painted red) welded to the top surface of each challenges when it comes to preservation
rafter (Throne Hall, see Figs. 4 and 5) and further use.

[1] Lorenz W, Fedorov S. Die Eisenkonstruktionen in den Gebäuden der Staatlichen Eremitage
St. Petersburg – Historische Bauforschung mit ingenieurwissenschaftlichen Schwerpunkt.
In Bericht über die 47. Tagung für Ausgrabungswissenschaft und Bauforschung, 16. bis 20.
Mai 2012 in Trier, Koldewey-Gesellschaft, Stuttgart, 2014, Bachmann M, et al. (eds), 2014;
[2] Lorenz W, & Heres B. Eiserne Eremitage. Bauen mit Eisen in Russland der ersten Hälfte
des 19. Jahrhunderts Ernst&Sohn: Berlin, 2017 (in preparation).
[3] Fedorov S. Matthew Clark and the origins of Russian structural engineering 1810–40s: an
introductory biography. Constr. History, J. Constr. History Soc. 1992; 8.
[4] Fedorov S. Rebuilding St. Petersburg’s Winter Palace in the context of Early European steel
structures 1838–1850s contemporary sources and documents. In Nuts & Bolts of Construc-
tion History, Carvais R, et al. (eds), Proceedings of Fourth International Congress on Con-
struction History, 3–7 July 2012, Picard, Paris, vol. 3, 2012; 203–214.
[5] Häßler D. The ‘Elliptical Beams’ in the Winter Palace in St. Petersburg – Structural as-
sessment of a lightweight structure. In Proceedings of Eighth International Conference on
Structural Analysis of Historical Constructions, Jasienko J (ed), Wroclaw, 15–18 October
2012, vol. 3, 2012; 262–272 (based on a master thesis at BTU Cottbus).
[6] Lorenz W, Heres B. The Demidov Ironworks in Nevyansk (Ural Mountains) – iron struc-
tures in building from the first half of the 18th century. In Proceedings of Fifth International
Congress on Construction History, Bowen B, Friedman D, Leslie T, Ochsendorf J (eds) ,
Chicago, 3–7 June 2015, Construction History Society of America, Chicago, 2015, vol. 2,
2015; 505–517.


Maintenance and Strengthening of
the Cross-Shaped Barracks Building

Marina Traykova, Prof., PhD; Tanya Chardakova, Assistant Prof., PhD; University of
Architecture, Civil Engineering and Geodesy, Sofia Bulgaria

Brief Description of the Presented Case Including

Project Aims and Challenges
The Cross-Shaped Barracks building is a cultural monument and one of the 100 National
tourist attractions in Bulgaria. It is situated in Vidin, which is a Bulgarian city on the Danube
River. The building was constructed in 1801 by local ruler Osman Pazvantoglu for the deploy-
ment of the Ottoman troops (Fig. 1). The designers, if there were any, are unknown. After the
liberation of Bulgaria from the Ottoman Empire, the building was used as a court and barracks
of Bulgarian troops. The building is famous for its specific cross shape in plan (Fig. 2). The
building structure is of brick and stone masonry (Fig. 3). The foundations are constructed as
dry stone masonry. The roof and the floors are timber structures. The building had not been
maintained properly, and a settlement of the south section corner had occurred due to moist-
ness of the soil and caused substantial cracking of the walls. Moisture was observed in all the
walls up to 1 m above the ground, which had caused damage to the external plaster as well
as several cracks in the internal walls. The building is situated in a region of grade 7 seismic
activity according to the Bulgarian seismic code. The project aimed to rehabilitate the exist-
ing structure and adapt the building to an ethnographic museum for the region. Before the
rehabilitation of the walls, the masonry was processed using polymer materials to act against
the moisture and in order to fill in the cracks. The strengthening itself was designed to be car-
ried out by means of steel elements to be anchored to the walls and sequentially sprayed with
concrete (Fig. 4a). The chosen design solution provided for the best possible preservation of
the architectural authenticity.

(a) (b)

Fig. 1: Final view: (a) exterior and (b) interior

Fig. 2: Layout of the building (structural design project drawings, 2010)

A design solution where the foundations are strengthened by a new reinforced concrete (RC)
structure of waterproof concrete was adopted (Fig. 4b). This included adding an air channel
around the building in order to provide for proper insulation and aeration of the foundations.

Fig. 3: Typical cross-section of the structure (design project)

(a) (b)

Fig. 4: Strengthening (a) of the brick masonry with sprayed concrete and (b) of the foundations

When construction activities began, serious new cracks were found in some of the masonry
walls. An additional RC wall in the building with a new foundation was designed for rehabilita-
tion of the damaged masonry walls (Fig. 5).

Influence of CH Analysis Results on the Project

in the Assessment Phase and in the Design of the
The design solution and the techniques for rehabilitation were decided according to the results
from the construction history (CH) analysis of the building: special research on the history of
the region and the construction traditions. The aim was to preserve the original architecture and
structure.[3] Hence, only the most crucial interventions were undertaken, and seismic strength-
ening was considered too invasive. Additionally, the chosen structural interventions were pre-
ferred because of economical considerations.

Fig. 5: Strengthening of the brick masonry with additional RC wall (design project of the
authors, 2012)

Project-Specific Experiences and Lessons Learned

The specific experience from this project and lessons learned can be summarized as the

1. Adaptation of old buildings is the best solution for reviving a region and creating the
original atmosphere. The adaptation of existing buildings ensures the best sustainable ap-
2. Preservation, rehabilitation and strengthening of existing structures require the application
of corresponding remedial measures. The final choice should be made according to the fol-
lowing: condition of the building, compatibility with the original structure, avoidance of
historical forgery and transparency of the interventions.
3. Selection of the best design solution for historical buildings requires a very rational and sci-
entific approach, starting with a study of structural integrity and determination of the causes
of deterioration.
4. Sometimes, innovative technologies for retrofitting demonstrate that those solutions are not
suitable in all cases. Technical and economical reasons can guide the final solution for ret-

The adaptation of the early 19th-century building of the Cross-Shaped Barracks in Vidin for
an ethnographic museum is a positive example of the proper adaptive reuse of monumental
structures. This, along with improper or rather lack in maintenance of this fine piece of historic
architecture, led to the need for strengthening of the building. In order to preserve its original
appearance while assuring its continued safe use, some inconspicuous strengthening interven-
tions were undertaken. The aim was to stop the active process of foundation movement due to
excessive moisture and the subsequent cracking of the masonry walls. The presented case study
again shows that heritage buildings are an important link between the present time and history.
They reflect the social, cultural and economic experiences of the past. The protection of heritage
buildings is however related to many structural problems. The majority of historical buildings
were built essentially following empirical rules and usually were not specially designed for both
gravity and lateral actions. At the same time, the authenticity should be necessarily preserved
for the future generations. The described design solution confirms that decision making for the
type and range of interventions is very important and is usually specific to the different cases.

[1] BDS EN 1998-3: Design of Structures for Earthquake Resistance. Assessment and retrofit-
ting of buildings.
[2] Ordinance “РД- 02-20-2 for design of buildings and facilities in seismic areas”, 27.01.2012
(Current Bulgarian Code).
[3] Traykova M , Chardakova T. Research on the seismic behavior and the retrofitting of the
buildings from the Bulgarian Cultural Heritage. Proceedings of the Sixth International Con-
gress “Science and Technology for the Safeguard of Cultural Heritage in the Mediterranean
Basin”, Athens, Greece, 22–25 October 2013, 2013.


Analytical and Experimental Studies
on the Technology of Late-Gothic
Vault Constructioni

David Wendland, Chair of Christian Art in Late Antiquity and the Middle Ages, Technische
Universität Dresden, Dresden, Germany

This paper outlines how analytical and experimental studies in construction history have led
to a fuller understanding of the construction of “cell vaults”, which has enabled restoration
of such vaults to be undertaken with greater confidence. According to common belief, cell
vaults were invented around l470 for the construction of the Albrechtsburg, the new palace
of the princes of Saxony at Meissen (Germany). These vaults that sometimes have stone ribs
but in many cases only have sharp groins
are characterized by their folded surface
with ridges between the groins (Fig. 1),
creating ceilings with subtle patterns of
light and shadow. The construction mate-
rial (apart from few exceptions) is brick
masonry. Their design corresponds to that
of late-Gothic net vaults, but the pattern of
their groins and arches is often much more
complex than that of conventional rib
vaults. Cell vaults rapidly became popu-
lar first in Saxony and then spread all over
central-eastern Europe, including Bohe-
Fig. 1: Cell vault in the second floor of the Albre- mia, Poland, the Baltic, Prussia and even
chtsburg at Meissen (Germany). The folded sur- beyond. During more or less a century,
face of such vaults are built with brick masonry hundreds of them were built in palaces,
(Courtesy: D. Wendland) town houses, convents and churches.

This paper is a reduced version of ‘Research on “cell vaults”: analytic and experimental studies on the technology of
late-Gothic vault construction’ published in Ref. [1].

We learn nothing about how these vaults were built from sources and contemporary records,
just as little why they were introduced and why they became so popular in such a vast area.
We do not even know what these vaults were called—the names we use are of modern origin.
Therefore, it is necessary to search for other ways in order to find out how the cell vaults were,
or might have been, actually built five centuries ago.

Construction Principles of Late-Gothic Cell Vaults

Experimental Archaeology Correlated with Detailed Surveys and
Analyses of the Archaeological Evidence
In order to formulate hypotheses on the construction process that is not otherwise documented,
its reproduction according to the method of experimental archaeology is the most promising.
In the current research on the construction of cell vaults”, experiments were carried out that
were closely linked with detailed surveys of original vaults—particularly one vault where the
masonry texture was visible in large portions. The curves, surfaces and single courses in the
masonry texture were the subjects of three-dimensional (3D) surveys and geometric analyses
that were carried out by means of a software tool for reverse geometric engineering. Using these
analyses, the construction process could be characterized; principle statements on the temporary
auxiliary structures needed for the building could be made; and construction principles based
on the archaeological evidence could be formulated. The methodology of such analyses has
been developed in earlier research by the author.2,3 By putting these hypotheses into practice
in experiments carried out in full scale, they could be evaluated and refined, and the practical
experiments could generate new observations about the vault.

Analyzing the Construction Principles of Cell Vaults

From the geometric features of the curves, surface form and masonry texture, it is possible to
learn about the building process and the principle features of the centering system. First, the
geometry of the groins shows that, during construction, these must have been supported by sin-
gle centering arches and not by a continuous formwork. Their curves can be described in simple
geometric terms, in contrast to the shape of the vault surfaces. All arches describe circle seg-
ments in vertical planes with discontinuities at the intersection points, which also provide some
indication of the whole system of centering arches. Such a layout of the arches is also common
in the contemporary rib vaults, and it corresponds to what we know about the geometric design
of late-Gothic vaults.4,5 The ridges or inward-folded groins, in contrast, describe curves with a
complex geometry: the 3D curves demonstrate variations in the direction and radius of curva-
ture, in addition to the abrupt changes of direction. From this, we can conclude that they were
not guided by centering but resulted spontaneously in the process of the construction of the
vault surfaces, which therefore must have taken place free-handed. The analysis of the masonry
texture proves that the analyzed vault was actually built free-handed. This is on one hand clear
from the spatial position of the masonry courses, which, over great portions, are inscribed in
tilted planes that are parallel—a typical feature of the masonry texture of vaults built without
formwork (Fig. 2).

Fig. 3: The exposed masonry texture of the

analyzed vaults shows the continuity of the
courses over large portions across the dis-
Fig. 2: Model simulation showing that the
continuities of the vault surfaces. This does
angle of the courses at the groin (V-shaped
not conform to modern masonry rules and
connection in front) is not determined a pri-
contradicts the current description models
ori but depends on the development of the
but turned out to be very practical in the
curves of the formeret arches (in the back)
experiments (Courtesy: D. Wendland)
(Courtesy: D. Wendland)

Finally, the masonry texture in these folds is fundamentally different from Ungewitter’s descrip-
tion because, in spite of what in modern terms we would claim to be a regular bond pattern, the
courses usually run continuously across them (Fig. 3).

Experimental Reconstruction of Late-Gothic Cell Vaults

The experiments were carried out in collaboration with an academy of historical craftsman-
ship, reproducing two vaults in full scale with bricks and mortar according to the origi-
nal. The used bricks have the same large
format as the original and are produced
in the traditional manner—their poros-
ity is of benefit for free-handed vaulting,
and they can be easily cut to shape with
an axe. The mortar has been reproduced
according to the analysis of the original
mortar. In these experiments, the hypoth-
eses of the construction principles as
developed from the surveys turned out
to be practicable, confirming the obser-
vations made above. The geometric
Fig. 4: Experimental reconstruction of the ana- problems in the masonry texture due to
lyzed late-Gothic cell vault where the hypotheses the curvature of the vault surface have
on the construction principles and building pro- occurred in the experiments in the same
cess could be verified and refined (Courtesy: A. manner as observed in the analyzed vault
Gosch) (Fig. 4).

The continuity of the masonry texture over large portions, which we could observe and which
we believe to be a principle of late-Gothic vault masonry, was even surprisingly easy to repro-
duce in practice, just by slightly bending the beds and sometimes cutting the corners of the
bricks. The axe to cut bricks is needed very often in the process (once more contradicting Unge-
witter’s theory), but the traditional bricks are cut easily as they are rather soft and not brittle.
The main difficulty for everybody was to pull apart the modern rules of bond pattern. These
rules, although considered “historical” or “traditional”, have been formulated only in the course
of the 19th century and are not relevant to earlier historical periods. As already pointed out,
from the archaeological evidence, the construction principles that were used by the masons of
the late-Gothic vaults could be characterized and, in the experiments, could be demonstrated to
be very practical in execution. In reality, these principles also make much sense because a very
good coherence throughout the masonry is obtained with the continuous texture. This can be
illustrated by the fact that cell vaults built in the 19th century with modern criteria often show
cracks that almost never occur in the original late-Gothic vaults. In conclusion, we can now
describe the difference between late-Gothic and modern construction rules, and we are capable
of reproducing late-Gothic vault masonry with high fidelity.

In the presented study, a methodology of analyzing historical construction is proposed, which
enables the formulation of statements on construction processes and design criteria in historical
structures that lack documentation. This can be relevant to conservation or restoration by pro-
viding better knowledge of complex structures, as well as in basic research on construction his-
tory. The improved knowledge of the construction principles and details illuminates the original
construction principles of late-Gothic vault construction, which has been perceived through the
optics of modern interpretation as established in the 19th century until now. Finally, through the
collaboration with an academy of historical craftsmanship, this knowledge is disseminated to
the technicians and craftsmen operating in the practice of restoration.

The research on cell vaults has been carried out within the research project “Form, Konstruk-
tions- und Entwurfsprinzipien von spätgotischen Zellengewölben—‘reverse engineering’ und
experimentelle Archäologie”, B. Klein, D. Wendland, Technische Universität Dresden (Ger-
many), funded by the Deutsche Forschungsgemeinschaft since 2009.

[1] Jasienko J (Ed.). Structural Analysis of Historical Constructions DWE: Wrocław, 2012.
[2] Wendland D. Traditional vault construction without formwork: Masonry pattern and vault
shape in the historical technical literature and in experimental studies. Int. J. Archit. Herit.:
Conservation Anal. Restoration 2007; l(4): 3ll–365.
[3] Wendland D. Lassaulx und der Gewölbebau mit selbsttragenden Mauerschichten. Neumit-
telalterliche Architektur um 1825–1848 Imhof: Petersberg, 2008.

[4] Müller W, & Quien N. Virtuelle Steinmetzkunst der österreichischen und böhmisch- säch-
sischen Spätgotik: Die Gewölbeentwürfe des Codex Miniatus 3 der Österreichischen Natio-
nalbibliothek in Wien. Imhof: Petersberg, 2004.
[5] Wendland D. Zum Bau figurierter Gewölbe – eine Anleitung im Werkmeisterbuch des Rod-
rigo Gil de Hontañón. In Werkmeister der Spätgotik: Personen, Amt und Image, Bürger S,
Klein B, Schröck K (eds) WB: Darmstadt, 2010; 244–272.


Frost Damage and Restoration of
Limestone Domes and Spheres in a
Heritage Building

Philippe Van Bogaert, Prof., PhD; Ghent University, Ghent, Belgium

Domes and spheres have been used extensively in heritage buildings during baroque and rococo
periods as well as in 19th-century buildings. Used as ornaments and as coverings and roofs,
these curved elements add to the monumental character of imposing buildings, which are pres-
ently considered a part of the cultural heritage. The number of historic buildings containing
domes and spheres is impressive and includes some of the world’s most famous structures.
Most of these ornaments and coverings are made of limestone because of its excellent quality
and durability. However, because limestone is a sedimentary rock, consisting mainly of calcium
carbonate, it may deteriorate due to acid rain and frost. The latter introduces cracking, allowing
water ingress and subsequent further cracking. Depending on the crack width, further mois-
ture ingress is fostered, and cracks grow. In the present paper, this progressive effect is being
assessed by numerical simulation. Obviously, all types of limestone blocks and ornaments are
prone to degradation due to frost. However, a brief survey of various degradations shows that
curved shapes are more vulnerable. The reason for this is yet to be found. The ratio of exposed
surface to volume of a sphere is not significantly different from the value for an equivalent cube.
However, if rainfall is considered from a single direction, for instance, vertical, the relative
exposure ratio of spheres is three times larger than that of cubes. This might give some indica-
tion of the larger degradation of curved surfaces.

Southern Pressure House, Antwerp

The Southern Pressure House, in Antwerp, consists mainly of two long buildings, the first of
which contains the machine room and the steam hall and parallel storage depots for coal and
oil.1 The boiler room and the warehouses are separated by a corridor that opens to a court-
yard. The second building houses offices, homes for personnel, a repair workshop and a forgery
house. Of all these, the imposing unit containing the accumulators for water pressure (Fig. 1) is
the most valuable. This is mainly a brickwork building decorated with limestone façade blocks,

which were intended to enhance the view

as the city particularly wished to improve
the appearance of the neighborhood.

The front view of the building clearly

shows the two vertical parts containing
the water accumulators. The entrance door
and ground level appear rather massive;
the twin columns are richly decorated,
suggesting the cylindrical accumulators
inside, and the narrow openings underline
the vertical orientation of the building.
Both towers are covered by domes carved
from single limestone blocks. An individ-
ual dome is shown in Fig. 2. The domes are
supported by white sandstone cylindrical
masonry. The total height of the structure
reaches 24.5 m. Figure 2 also shows some
of the heavy cornices, column capitals and
decorating spheres. The highly decorated
industrial building was highly appreciated
in the second half of 19th century and dis-
plays the wealth and prosperity of the port
Figure 1 Southern Pressure House: front view of Antwerp in industrial times. The style of
this building may appear to be excessively
heavy, but the exterior form clearly shows
the particular purpose of the building. In
addition, the domes entirely correspond to
the inside equipment.

Although the installation was decommis-

sioned in 1977, the whole complex was
listed as protected heritage in 1979. About
20 years back, a complete restoration was
undertaken, and the buildings became a
center for performing arts. This lasted
until five years ago when the theatre com-
pany ceased to exist. Since then, events are
being organized in the 4000  m2 building
Figure 2 Top of the vertical part with the dome with its seven rooms, except for the front
unit (Fig. 1) which remains unused. The
activities include seminars, smaller concerts, receptions, discussions and workshops, as well as
art studios. During the inspection of the heritage building in January 2012, several limestone
ornaments and blocks were found to be heavily cracked. In addition, during the 1985 restoration,
inferior products were used at some locations. After some of the limestone debris fell, a second
inspection was organized in December 2012. As both inspections were conducted in winter, it
was hoped that frost was at maximum and recent degradation would be detected. Cracks reached

a width of 10 mm, and some parts needed

immediate removal as there was a threat
of debris falling on to the footpath below.
The inspections also showed that some
parts of the structure were already missing.
These findings have urged the owners to
apply for necessary funding for a restora-
tion program and to install safety measures
in order to avoid endangering passers-by
below the building. However, the larg-
est concern applies to the domes as these
also suffer from cracking. Figure 3 shows
a detail of the cracked state of the domes.
Fig. 3: Degradation of the dome It is clearly seen that the limestone surface
is eroded in horizontal layers. This seems
to be characteristic of domes and spheres.

Simulation of Dome Cracking

The domes are relatively small and consist of hemispheres with an inner radius 1.22 m and an
outer radius 1.40 m, the thickness being 0.18 m. The base of the dome consists of a flat ring
with an outer radius of 1.63 m, thus providing a larger support and stiffening the dome. The
idea was to constitute a model of volume elements and to introduce cracking as soon as tensile
stresses exceed the tensile strength of the limestone. The tensile strength value of 5.67 MPa is
considered rather low, far below the average compression strength of 140 MPa or a character-
istic value of 126 MPa. This type of approximation allows reasonable assessment of the influ-
ence of cracks on dome resistance. In further steps, nonlinear material characteristics may be
considered, although it is believed this will not consistently modify the results. The approach
consists of introducing cracks into the model, assuming subsequent water ingress and allowing
the effect of frost. The latter will start the progressive effect, the issue being how far this process
may continue and eventually lead to the destruction of the dome.

Modeling of the effects of frost proved to be delicate. A first approach considers that an initial
crack is completely filled with water. As the water freezes, its volume increases by 9%, thus
causing an internal pressure in the crack. This internal pressure needs to be applied to the crack
surface. The main issue is to identify the magnitude of the pressure. This, and other approaches
based on the Washburn equation,2 all disregard the actual process of freezing in natural stone.
The latter has been researched more extensively in Ref. [3]. Although the research is considered
to be idealized, it enumerates the various phenomena involved during freezing. The model is
based on the importance of the flow of water toward the solidification front of ice, as well as on
the existence of thin films separating the ice and the surrounding stone, as mentioned earlier.
The thickness of these films varies from 15 to 30 Å. These films also exert an attractive force
on the pore water and a disjoining pressure that pushes the ice and the stone apart, which is the
pressure we are seeking. The fastest damage growth rate occurs in the range from −4 to −15°C.

In Ref. [3], the case of a spherical cavity has been studied using the Gibbs–Duhem equation
to describe the change from liquid to solid state, the van der Waals interactions and the Clau-

Fig. 4: Cracking of the dome wall

5 sius–Clapeyron equation. The porosity of the

Stress (MPa)

stone has been taken into account, and three
στ examples have been worked out. It appears
σχ that dense rock-like granite can certainly
crack due to freezing water as the pressure
crack depth (mm) rises up to 25  atm or 2.533  MPa. However,
–10 10 30 50 70 90 110 130 150 in the case of sandstone, in a spherical cavity,
pressure would rise to 12  atm or 1.22  MPa,
and this value also corresponds to limestone.
In porous stone, the main pressure rise is due
to the pre-melting stage when the ice is very
Fig. 5: Evolution of stresses with crack depth
close to the stone. The flow then reverses,
and the water flows toward the solidification
front. This explains why the pressure rises sufficiently to cause cracking only in the case of very
impermeable rock. Consequently, the research in Ref. [3] has been adopted for the present simu-
lations. The pressure in a spherical cavity is most probably higher than that in a longitudinal
crack, which has a larger surface area in contact with air. Consequently, an internal pressure of
1.22 MPa has been applied to a crack on the outer surface of the dome. This crack runs along a
quarter of the circumference of the dome, its depth being increased stepwise. Looking at a cross-
section of the dome, the local meridian stresses can be found as shown in Fig. 4 for crack depths
of 1/20, ¼ and ½ of the dome thickness.

The former cracks have been assumed to exist before water ingress and freezing. The stresses
shown in Fig. 4 may either cause excessive tensile stress at the crack tip or may crush the crack
opening at the surface. In addition, the crack may be widened due to internal pressure. In all
of these cases, the crack will increase until these quantities decrease below a critical level. The
diagram in Fig. 5 summarizes the tensile and compression stresses as a function of the crack
depth. The upper blue line shows the tensile stresses, which in all cases is below the tensile
strength of 5.67 MPa. This simply implies that the cracks will not increase due to frost, and
there is no progressive deterioration. This also applies to the compression stress, which is two
to three times lower than the tensile stress. According to these results, water ingress and subse-
quent frost cannot be responsible for damage to the domes. In addition, the evolution of crack
width with the depth has been summarized in the graph in Fig. 6. This illustration also shows
that the crack width is moderate. Hence, the model has indicated that cracking of the limestone

crack width




crack depth (mm)
–10 10 30 50 70 90 110 130 150

Fig. 6: Evolution of crack width with depth Fig. 7: Developed crack

ornaments is probably not due to freezing of water in the cracks. Calculations were continued
until 80% of the dome wall thickness was cracked, and yet no progressive effect was seen. At
this particular point, the crack width grows to about 0.3 mm. When all loads are considered, the
dome still remains stable. The deformations at this stage are shown in Fig. 7, the crack opening
being clearly visible, including the effect at the tip inside the stone. In view of these results, the
crack length has not been varied, nor has the exact location been modified, and further analysis
becomes irrelevant.

As the present state of deterioration of the domes and other limestone parts of the Southern Pres-
sure House in Antwerp cannot be due to frost, the cause must reside in chemical attack due to
increase of CO2, thus favouring the dissolution of the material. Due to the nature of limestone,
the calcium carbonate may dissolve if the pH of rainwater is sufficiently low because of CO2
content. Therefore, refurbishment should preferably make use of soft lime mortar to close the
surface cracking. Deeper and larger cracks, endangering further decay and disintegration of the
parts, may be treated by injection of epoxy-based products. Former calculations have shown
that injection pressure may easily reach 5 bars (0.5 MPa). However, these epoxy-based prod-
ucts are harmful to the limestone as they have little permeability and can become rather rigid
in the stone. Hence, they should be used in-depth only and after the surface treatment of cracks
with lime cement is complete. After successive inspections and temporary measures to avoid
accidents, the restoration of the Pressure House is presently being considered. The project will
require more extensive evaluation of the various repair methods. However, it may have become
clear that water ingress and frost are not the main reasons for the degradation.

Recent inspections have revealed that important limestone parts of the Southern Pressure House,
a heritage building in Antwerp, show large cracking and may fall from the building. Water
ingress and frost are thought to be the cause. The domes show horizontal cracks that might be
of particular concern during future restoration. An extensive numerical model has been used to

predict cracking due to frost. Various approaches to model the effect of water ingress and frost
have been considered. Among these, the approach based on the successive stages during ice
growth, the presence of a water film separating ice from the surrounding stone and the internal
water flow toward the ice front seems the most successful. Application of this approach clearly
shows that the effect of frost is incapable of fostering progressive cracking as both the tensile
stress at the crack tip and the compression stress at the surface are sufficiently small. In addition,
crack width may increase with depth, and this quantity also remains small.

[1] Himler A. The Southern Pressure House and 7 other hydraulic stations” (in Dutch). Monu-
menten en Landschappen 1985; 4(6): 8–28.
[2] Fisher L. An experimental study of the Washburn equation for liquid flow in very fine capil-
laries. J. Colloid Interface Sci. 1979; 69(3): 486–492.
[3] Vlahou I, & Worster M. Ice growth in a spherical cavity of a porous medium. J. Glaciol.
2010; 56(196): 271–277.


The Gothic Tower of Freiburg Minster,
Germany: Analysis and Repair

Rainer Barthel, Prof. Dr.-Ing. Partner, Barthel & Maus, Beratende Ingenieure GmbH,
München, Germany
Joram Tutsch, Struct. Eng., Dipl.-Ing. Technische Universität München München Germany
Joseph Jordan, PhD, Struct. Eng., Dr.-Ing. Partner, Barthel & Maus, Beratende Ingenieure
GmbH, München, Germany

The construction of the Freiburg Minster began around 1200 in the Romanesque style, and from
1230 onward, the master builders adopted the Gothic style. The tower was completed around
1330. It is 116 m in height and has survived until the present without major damage or altera-
tion. The outstanding feature of the tower is the filigree construction of the octahedral-pointed
spire, which has no internal supports (Fig. 1). Since its construction, no major alterations were
undertaken, and no strengthening measures have been required. It survived several catastro-
phes, including one of the most destructive earthquakes in Northern Europe in 1356. It also
survived the bombing raids of November 1944. Nevertheless, the masonry and the tracery had
to be repaired at regular intervals, for instance, in the 1920s and 1960s. In 2009, a scaffold was
erected once again after small pieces of stone fell onto the viewing platform. Originally, the
intention was to undertake local repairs and stone conservation, but further cracks were discov-
ered, which proved to be much more dangerous than expected. In particular, the cornerstones
of the main struts of the octagon, which are structurally essential, showed several cracks. The
crucial questions were what were the causes of this cracking and what kind of remediation
measures would be necessary.

The lower part of the tower is square in plan and very solid, with only a few openings. At 40 m,
there is the so-called star gallery and the belfry. Above the belfry, at a height of 55 m, there
is an internal platform. From there, the tower is octagonal and free from any internal struc-

ture—there are no walls, only pillars and high

tracery windows. At the height of 72 m, there is
a further gallery on the outside. Above that level,
the octagonal spire rises. The spire is subdivided
into a solid base, seven stories of tracery elements
between levels 1 and 8 and the pinnacle incorpo-
rating the finial. Circumferential wrought-iron
ties are situated on eight levels.

Damage to the Structure

When the first cornerstone was removed, it was
found to be completely destroyed; what remained
was a mere stump. Numerous cornerstones
showed similar damage, and some were found
to have been repaired rather crudely in former
times. It is difficult to repair such a cornerstone
in an adequate way. The vertical load on a cor-
nerstone is up to 400 kN. The initial assumption
was that corrosion of the iron ties was the cause
Fig. 1: Freiburg Minster: View from the of the cracks, and indeed, in some chords 1–2 m
octagon level inside the spire long, significant corrosion of the ties was found.
Water had infiltrated through open joints between
the coping stones and was trapped in the tie channel. Thus, water freezing could also cause dam-
age. However, corrosion of the iron ties does not sufficiently explain all the cracks and defects.
Generally speaking, the damage inside the tracery was not found to be alarming. There is little
more than a few displacements and minor spalling at some joints (Fig. 2a). Fortunately, only a
small number of stones had developed cracks parallel to the outer surface of the tracery panel,
which would be dangerous. While there are some cracks that seem to be due to tensile stresses,
more often, the cracks indicate bending (Fig. 2b). All the visible damage and deformations were
recorded. Especially important was the need to identify the repair measures carried out in former

(a) (b)

Fig. 2: Typical damage inside the tracery. (a) Relative displacement; (b) bending crack

repair campaigns. Substantial repairs were undertaken in the 16th century, in the 19th century,
in 1920 and again in 1963. The quality of these measures differs greatly. A full 3D computer
model of the entire structure was built. Every stone, every joint and every metal piece is mod-
eled as a separate element; every crack and other anomalies are also fully modeled. All historical
alterations are also included. The model was created using a photogrammetric survey that was
completed by hand measurements.

Structural Analysis
Several different methods of modeling the structure were tested in order to find an appropriate
way to conduct structural analysis calculations. The first approach was a finite-element model
using simple plate elements for the tracery panels with a reduced stiffness. A more complex
model was also created consisting of beam elements and special gap elements at the joints in
the tracery, with separate truss elements for the iron ties. This model is very cumbersome, and
the results are not easy to check. Therefore, a much simpler model was also developed to gain a
better understanding of the overall statical behavior. The tracery panels are substituted by simple
truss elements, and their stiffness was derived from a detailed analysis of the various types of
tracery under vertical and lateral shear forces. The truss elements can transfer only compressive
stress; in addition, the stress–strain relation in compression is nonlinear.

Dead Load
The total weight of the spire is 6310 kN. For comparison, the total weight would be 9470 kN if
the structure were of the same thickness but with solid masonry walls. The tracery panels are not
only ornamental—they support the weight of the entire structure above. The load paths through
the tracery panels result in a concentration of forces along the vertical edges. The pattern of
the tracery causes horizontal forces that provoke a certain increase in the circumference of the
octagon. On the other hand, the overall shape of a pyramid gives rise to horizontal compres-
sive forces, which hold the structure together. The resultant tensile forces in the wrought-iron
ties therefore remain small. One feature of the construction is particularly surprising. The only
structural elements connecting the tracery panels with the struts are the iron ties. However, the
struts and the tracery panels do not have the same stiffness in the vertical direction. Different
displacements due to elastic deformation or shrinkage can thus cause significant shear forces
in the ties and the adjacent stones. The tie becomes a “shear pin”, which can give rise to local
stress peaks and tensile stresses in the stones. This may well account for the vertical splitting of
numerous stones in the horizontal chords adjacent to the struts.

Wind Load
Because there is no information on assumptions regarding wind loads on such a structure, wind
tunnel tests were conducted on a physical model. Wacker Engineers undertook experiments on
a 1 : 100 scale (600 mm) model of the entire octagonal tower and spire. Further models at 1 : 40
scale were made of smaller sections to determine the distribution of forces around the circum-
ference of the spire. A peak velocity wind load of 1.21 kN/m2 is assumed, which corresponds to
a velocity of around 44 m/s with a 50-year return period. The resultant wind force on the struc-
ture is astonishingly large; indeed, the force is much higher than on a spire that is not perforated.

The wind force corresponds to a force coefficient of roughly 1.4. The forces acting parallel to
the surface are particularly high due to the roughness of the surface. The suction forces on the
structure are relatively small. The forces in the ties and their connection to the masonry are
much larger than those due to the dead load. At the fixings, the maximum tensile force is 226 kN,
which is nearly three times that under self-weight. The tracery panels parallel to the wind direc-
tion act as shear walls. At the cornerstones, two effects occur. First, the forces in the ties have to
be diverted, causing forces inward; second, a proportion of the forces have to be anchored in the
stone. Both effects lead to tensile stress across the full thickness of the stones.

Seismic Loads
Vibration measurements were conducted on the model and determined that the vibration period
of the first and second eigenmode is around 1 Hz. Assuming the seismic zone and the founda-
tion behavior according to Eurocode 8, the equivalent static lateral seismic load was found to
be higher than the total wind load. A story-by-story or even stone-by-stone assessment of the
effects of this load within the statical model is currently being implemented.

Material Tests
A great deal of materials testing was undertaken. Mechanical properties of the original and new
mortar, the lead, the sandstone and the wrought iron were determined. In addition, some full-
size copies of structural elements were tested in the laboratory. Some nondestructive testing was
also undertaken to investigate internal cracks in the stone. One result of fundamental importance
was revealed by this series of tests; it was found that the coefficient of thermal expansion of the
historic sandstone and of the wrought iron is the same, indicating that these two materials fit
together very well.

Two repair philosophies were discussed. The first proposed using external ties or even steel
cables to embrace the whole structure horizontally. These ties are positioned on the outside of
the structure at the level of the chords and post-tensioned to eliminate the tension forces in the
original ties. No drilling is required. At the cornerstones, steel brackets are required to bear the
forces. The cornerstones are thus loaded in three-dimensional compressive stress. This ensures
that the repair of the damaged cornerstones is easier, and more of the original cornerstones can
be retained. The chords are also loaded in compression, which assists the flat-arch action under
local wind loading. This repair philosophy aimed for the minimum of interventions and altera-
tion to the historic structure. However, this solution would generate entirely new load paths
inside the chords due to the post-tensioning. In statical terms, this would be a major change; it
could cause further unpredictable displacements, and even damage, due to the inhomogeneity of
the structure arising from the numerous repair measures completed in the past.
A second philosophy was developed that will maintain the original statical situation, and this is
the approach now being implemented. The stones and the joints will be repaired locally and, if
necessary, strengthened to meet the local requirements. Twelve cornerstones show small cracks,
and a frame made of titanium was designed to hold them together and to provide adequate
safety under vertical and horizontal loads. Eight cornerstones are badly damaged and must be

removed and replaced by new ones. Split

stones located in one third of the horizon-
tal chords have to be replaced in order to
achieve sufficient shear stiffness—a precon-
dition for the flat-arch action (Fig. 3). All
historic repairs that do not meet the stati-
cal requirements have to be replaced. When
implementing this kind of in-depth repair, it
is possible to strengthen the overall struc-
ture and to increase substantially the robust-
ness and level of security. However, if the
original ties do not meet the required level
of safety, additional ties are inevitable. The
Fig. 3: Repair plan: example for chords of sto- final decision depends on the investigations
ries 6 and 7 (the wrought-iron tie ring and stone mentioned above, which still have to be
replacements are dark colored) executed. The ties would consist of titanium
with a diameter of 20 mm and would not be
post-tensioned. Thus, they will not change
the stress distribution and load paths in the original structure. Two alternative ways of introduc-
ing these new ties are being considered. They could be positioned inside the upper part of the
chords and anchored in the struts directly above the cornerstones. This would require drilling
holes at least 35 mm in diameter over a length of about 5 m. Alternatively, to minimize drilling,
the ties could be positioned directly above the chords, necessitating penetration of three or four
tracery stones. The diameter of the drilling hole could be reduced to 24 mm. These ties would
hardly be seen from below. Apart from being resistant to corrosion, a further advantage of tita-
nium is that its Young’s modulus is roughly half that of steel, which reduces the change of forces
in the ties due to temperature variation. These additional ties will be inserted, if necessary, after
the repair of all the stones. The implementation of the various repair measures began in 2014.
Due to the construction of the scaffolding, the work progresses from top to bottom. Currently,
the repair measures have been completed in the upper third of the spire. One of the remain-
ing major challenges is the design of the temporary steel support structures that will allow the
removal and replacement of several cornerstones (Fig. 4).

The spire on the tower of Freiburg Minster is an invaluable monument of architectural and
construction history. The analysis of the construction is an interdisciplinary research effort. The
tracery is a filigree masterpiece of medieval art, and at the same time, it is a substantial structural
element that must transfer both vertical and horizontal forces. Even the repair measures from
the 15th and 16th century are witness to the level of technical development and understanding.
Thus, it is worth assessing the construction to an extent that goes far beyond the normal in order
to avoid repair or strengthening measures, that would alter the historic substance. The tower
has proven its stability for nearly 700 years. Such a statement is frequently used to justify the
argument against undertaking any strengthening measures. However, 700 years existence does
not prove that the tower is still safe or will continue to be safe in the future. The statical calcula-
tions and the tests show that the overall structure and the dimensions of the structural elements

Fig. 4: (a) Three-dimensional (3D) model of repaired stereotomy at a cornerstone. (b) 3D model
of temporary steel structure for replacement of cornerstones

correlate more or less with the forces determined theoretically. However, the reliability of the
numerous structural elements cannot be proven completely. Stress concentrations or a weak
sediment layer in the sandstone or slag inclusion in the wrought-iron ties cannot be excluded.
In addition, the repair measures during previous centuries led to a much larger total number of
stones and joints than in the original structure. That has weakened the structure and makes it
impossible to assume the original level of safety. Even after repairing all the damage that was
detected, it will not be possible to achieve the level of safety of the original structure. Thus,
extreme wind loads or a large earthquake can be a real threat. It will not be possible to strengthen
the structure to an extent that removes all the risks completely, according to our contempo-
rary understanding of safety, without implementing major strengthening measures. This would
destroy the original monument. The proposed strengthening by means of additional ties could
reduce the risk of major damage or collapse substantially. The final decision of whether this is
acceptable must be made by consideration of the various values and points of view regarding
historic structures.


The Municipal Public Bath at
Strasbourg (1905–1908): A Cultural
Heritage in Reinforced Concrete

Christiane Weber, Prof., PhD; Leopold-Franzens-Universität Innsbruck, Innsbruck, Austria

Alexandre Kostka, Prof., PhD; Université de Strasbourg, Strasbourg, France

In Strasbourg in Alsace, France, a municipal public bath has survived till this day almost in
its original form: Les Bains Municipaux.1 The complex was conceived as a “municipal public
bath” in the years prior to World War I. At the time, Germany had annexed Strasbourg as the
“Reichsland Elsass-Lothringen” as a consequence of the Franco-German War of 1870–1871
and the Frankfurt Peace Treaty. On 17 June 1901, the municipal chief architect Johann Karl
Ott presented the first proposal to build the public bath, suggesting the grounds of the erstwhile
Nikolaus barracks on Nikolausring to be a suitable location.2 With the appointment of a large
number of socialists to the city council following the 1902 elections, the project for the munici-
pal public bath was promoted with even more enthusiasm, which led to a new proposal on 3 Feb-
ruary 1904. On 28 September 1904, the city council agreed to adopt a modified version of the
Bauinspektor (municipal building inspector) Fritz Beblo’s scheme presented earlier that year.3
The building was completed between 1905 and 1908 (public bath); a wing specially dedicated
to public health was opened in 1910. Beblo’s plans provided for two separate pools: one for men
(14.4 m × 26.6 m) and one for women (10.3 m × 18.4 m). Arranged perpendicular to each other,
the two swimming baths were characterized by their tall half-hipped roofs. Situated in between
were the Turkish baths, complete with a steam bath and a so-called Wildbad (cold water bath).
In addition, there were 54 shower and bathtub rooms, which were not purposed for swimming
but for practicing personal hygiene. They were arranged across three levels according to class
and divided by gender, respectively. They were situated in the two wings on either side of the
oval-shaped foyer that projected out from the facade.

The building in Strasbourg was developed as a reinforced concrete skeleton structure.4 Several
reasons may have led to this decision. The Strasbourg building authorities were very open to
innovation and thus endorsed the use of new construction types, such as reinforced concrete.
Eduard Züblin, one of the top reinforced concrete design engineers at the time, even relocated

to Strasbourg to establish his construction company in 1898,5 further evidence of the city’s
experimental construction climate at the turn of the 20th century. Züblin’s endeavors to establish
the French Hennebique construction system in Strasbourg were not least crowned with success
because the municipal building authorities had been restructured in line with the German admin-
istrative model. That the city building department played a key role in the authorization of the
reinforced concrete structure for the public bath manifests itself in the fact that its officials were
responsible for evaluating the building application.

The load-bearing analysis enables us to identify the edifice as a reinforced concrete skeleton
structure in line with the Hennebique system, even if it is not explicitly named. From the point
of view of structural engineering, the barrel-shaped ceiling vaults spanning the two swimming
baths for women and men, respectively, presented a highly innovative solution. As had been
the case in Munich’s Müllersches Volksbad, built using the construction system of Monier, and
with the public bath in Colmar,6 which Züblin built in 1904, a decision was made in favor of
reinforced concrete structures over an iron structure, which was popular above all in England.
The reason was twofold: first, reinforced concrete vaults were less susceptible to corrosion,
and second, they offered additional heating advantages. Moreover, the swimming bath roofs
in Strasbourg were conceived as a double shell in order to prevent condensation. In order to be
able to realize vaults spanning 16.90 m over the male pool and 17 m over the female pool, gird-
ers (so-called sickle-shaped binders) (Fig. 1) were positioned at 5.25 m (male pool) and 4.4 m
(female pool) intervals. Longitudinal ribs were placed between the main beams to support two
“covers”—one at the top and one at the bottom, leading to the space-defining shells of a maxi-
mum thickness of 80 mm. In addition, the plans show that the lower shells were hung from the
upper longitudinal ribs between these girders.3 Accordingly, the structure cannot be regarded as
a “shell” in the strict sense of the technical term as this shell does not itself perform any support-
ing action from a load-bearing perspective and, therefore, has not been designed as such. The
reinforced concrete roof structures are visible neither in the female nor in the male pools, just as

Fig. 1: Eduard Züblin’s sickle-shaped binders over the swimming pool in the bath5 (AVCUS 843
W 488)

(a) (b)

Fig. 2: (a) Fritz Beblo’s municipal public bath at Strasbourg. (b) Barrel vaults over the swim-
ming baths

they are invisible in the foyer, which boasts a dome likewise made of reinforced concrete. The
ceilings are embellished with stucco work and colored borders (Fig. 2).

The second notable innovation by Eduard Züblin concerns the design of the pool. In 1900, the
company built the first public bath featuring a pool made of “béton armé” (reinforced concrete)
in the Alsatian town of Gebweiler/Guebwiller.7 Until then, swimming pools had been made of
tamped concrete or unreinforced concrete. This kind of construction was prone to stress crack-
ing in the case of fluctuations in temperature, which led to leakages in the pool. Here, Züblin
was able to capitalize on his engineering experience in designing silos.8 For Gebweiler and for
the public bath at Strasbourg six years later, he came up with a structure with a thickness of
just 150 mm, which was reinforced on the outer wall by ribs made of reinforced concrete. The
pool is mounted freely on supports and separated from the structure by an expansion joint to
prevent tension caused by temperature-related changes in lengths. Züblin’s innovation was duly
noted in the technical press; Fritz von Emperger published a paper on the pool in Gebweiler in
1902 in his journal “Neuere Bauweisen und Bauwerke aus Beton und Eisen” (State-of-the-art
construction methods and buildings made of concrete and steel).7 The builders of the Strasbourg
public bath were not given much time to see their creation in action. After fewer than ten years
in operation, Alsace once again became French, or perhaps Alsatian to be more accurate; the
“old Germans” were hounded out, among them the pool architect Fritz Beblo, who moved back
to Munich where he took up the post of head of city architecture. The Swiss engineer Eduard
Züblin emigrated to Zurich, where he died in 1916.

The municipal bathhouse of Strasbourg is exceptional in France; it represents the only example
of a public bath dating from the period of German rule in Alsace (1871–1918) that is still func-
tioning in its original form today. It is also the only public pool located directly in Strasbourg
city center, and its place in the collective consciousness is, if possible, even more prevalent.
Since 1908, the imposing building that houses the pool for men and women has dominated the
residential area on the former Nikolausring, today known as the Boulevard de la Victoire. The
location lies on the border between the medieval city center and the “Neustadt” or “German New
town” that began to develop from the 1880s to the northeast of the city, and which is now being
integrated in the UNESCO preserved area of the city of Strasbourg. In 1910, this impressive
building was extended to incorporate a medical wing, which housed sports facilities and medical

baths, as well as a treatment center for tooth and gum infections. The uneven trapezium form of
the plot still determines the facet of the baths today. The whole front length of the elevated high
baroque façade is visible from the street.1 A preservation report, recently commissioned by the
city of Strasbourg, lists only minor damage to the plaster, caused by humidity.9 To the rear of
the building, there is an area that backs onto Rue Prechter and is hardly used today. Feasibility
studies, hotly debated at the time in Strasbourg on a political level, have proposed this area for a
possible extension. A preservation report by the Atelier Oziol-de Micheli, dated February 2016,
is the first attempt after the building was declared a listed monument on 10 October 2000.2

Today, the bathing facilities, including the sauna area of the municipal baths, are still retained
from the original. Only minor details, like the closing mechanisms in the changing rooms for
example, were redeveloped in the interwar years. The room datasheet also documents moisture
damage on the wood floor in the changing cabins.9 The glass windows are also from the origi-
nal. In the same way, the slightly more playful women’s bath (18 m long) has also retained all
of its original features.2 Both the men’s and women’s pools are equipped with fully functioning
showers with fittings from the original period. Also dating from the pre-World War I period
is the ventilation system, in a slightly altered state but still functioning today.9 The reinforced
concrete constructions of the pools were well-known for long-lasting stability, with no need for
any repairs. The preservation report mentioned only corrosion of some reinforcing bars on the
expansion joint of the female pool. The reinforced concrete roof structures, also designed by
Züblin, were of the highest standard technically achievable at that time and have been main-
tained with particular care. These very complicated features call for a sophisticated restoration
scheme, which the authorities were able to present only six years after the first preparation
meetings in 2010. The main problem impeding rapid decision making was the question of a
partial privatization of the bath, which—according to earlier plans—would comprise at least the
steam bath and the sauna. However, bowing to public pressure, at a meeting on 24 June 2016,
Strasbourg’s mayor Roland Ries, accompanied by the department heads of finance, health and
sport, as well as the staff of the municipal building institutions, presented a plan that will keep
the bath in the public domain. The estimated renovation cost of 30 million euros will be partly
financed by earnings on the maintenance cost and from the rent of the former engine building to
a public health company. The former health building, to the left of the structure, is not included
in the renovation plan. As it has the greatest financial value due to its strategic position on the
avenue, speculations are ongoing about its future destination.

[1] Kostka A. La genèse transnationale des Bains municipaux de Strasbourg. In Strasbourg
De la Grande-île à la Neustadt: un patrimoine urbain exceptionnel, Cassaz D, Eberhardt S
(eds) Lieux Dits: Lyon, 2013; 113–120.
[2] Fritsch F. Bains Municipaux de la Ville de Strasbourg. Strasbourg: Direction Regionale
des Affaires Culturelles. Unpublished document, 2011.
[3] Weber Ch. Les Bains Municipaux at Strasbourg (1905–1908): an example of cultural and
technical transfer between France and Germany. In Proceedings of the Second Conference
of the Construction History Society Queens’ College, Campbell JWP, James-Chakraborty
K (eds), Cambridge, 20–21 March 2015, 2015; 199–208.

[4] Weber Ch. Der Ingenieur Eduard Züblin in Straßburg: Erste Überlegungen zu bautechnis-
chen Transferphänomenen. METACULT. No. 2, 2014; 39–44.
[5] Everts-Grigat S, & Fuchs K. Züblin: 100 Jahre Bautechnik 1898–1998 Eduard Züblin AG:
Stuttgart, 1998.
[6] Triboux P. Strasbourg, Colmar, Mulhouse: les programmes de bains municipaux en Al-
sace au début du XXe siècle. Une politique volontaire d’hygiène publique. Livraisons de
l’histoire de l’architecture 2007; 14: 81–93.
[7] Emperger F. Neuere Bauweisen und Bauwerke aus Beton und Eisen, vol. 2 Lehmann &
Wentzel: Wien, 1902.
[8] Kostka A, Weber Ch. Municipal baths at the beginning of the twentieth century: an exam-
ple of Franco-German cultural and technical transfer in the Upper Rhine Region. Proceed-
ings of the Fourth International Meeting of the European Architectural History Network,
Dublin/Ireland, 2–4 June 2016, 2016; 146–161.
[9] Atelier Oziol-de Micheli. Les Bains Municipaux Boulevard de la Victoire Strasbourg:
Diagnostic Patrimonial, Strasbourg. Unpublished document, 2016.
[10] Franck E. Städtisches Schwimm- und Medicinalbad in Straßburg i. E. Bauzeitung für
Württemberg, Baden, Hessen, Elsass-Lothringen, vol. 5, no. 41, 1908; 325–328.
[11] Grandvionnet Ph. Les bains municipaux de Strasbourg (1894–1911) dans l’architecture
européenne des bains publics. METACULT. No. 4, 2015; 20–27.


History and Rehabilitation of
Reinforced Brick Ceiling

Michael Fischer, Prof., PhD, Struct. Eng., Dr. Lorenz & Co. Bauingenieure GmbH,
Berlin, Germany

This paper provides a current assessment of the fire resistance of a real-world example of a
Kleinesche Decke. Knowing the slab’s construction history is essential for its assessment. In
Germany, Johann Friedrich Kleine obtained the patent for the first and still well-known rein-
forced brick floor (Fig. 1) in 1892. By 1910, more than 60 different slab systems of that kind
came up here. For many years, this diversity outpaced that of reinforced concrete slabs devel-
oped at about the same time. However, above all, reinforced brick floor systems were also more
widely used. Easy to build and inexpensive, reinforced brick floors shaped multistory construc-
tion worldwide since the early 20th century.1

Current Structural Condition

The former Albert Schweitzer School in Hannover was constructed in 1896–1897. The build-
ing is to be converted into a residential building. Although all floor slabs are designed as a

Fig. 1: Exemplary illustration of a Kleinesche Decke1


Kleinesche Decke, they vary slightly

from room to room. This fact entails,
to some extent, different fire resist-
ance ratings. Generally, there are two
types of slabs: slabs in the housing
space and those in the corridors. The
wood strip flooring in the housing
space (Fig. 2) is covered with differ-
ent floor coverings, such as magnesite
flooring or linoleum. Furthermore, the
smaller rooms were constructed using
floor girders with reduced dimensions
due to shorter spans. However, these
Fig. 2: Schematic diagram: slab structure of the hous-
variations are not relevant to the fire
ing space
resistance classification.

All slabs above the corridors,

excluding the top slab, com-
ply with the schematic diagram
variant 1 (Fig. 3). Variant 2 for the top
slab (Fig. 4) differs with respect to the
unconfined floor girder’s top flange,
entailing a reduced fire resistance

Fire Resistance
Fig. 3: Schematic diagram: slab structure of the Generally, it is not easy to deter-
corridor—variant 1 mine the fire resistance of the present
slabs as the Kleinesche Decke type
is not listed in the German national
standard DIN 4102—Brandverhalten
von Baustoffen und Bauteilen (DIN
4102—behavior under fire).2 The
following classification is based on
comparisons with similar construc-
tions listed in the DIN as well as on
evaluations of about 30 load tests on
historical reinforced brick floors car-
ried out in recent years.1 The slabs are
assigned different fire resistance clas-
sifications due to their differences in
structure. There are two different clas-
Fig. 4: Schematic diagram: slab structure of the cor- sification criteria: slab bay and floor
ridor—variant 2 (top slab) girder (Table 1, Fig. 5).

Slab bay Floor girder

Room Min. in Max. in Recommended In accordance Recommended
accordance accordance fire resistance with DIN fire resistance
with DIN with DIN classification 4102-4 classification
4102-4 4102-4
Housing F 30 Classification
F 90 F 30 F 30*
space not possible
See table 27, See
row 1.2 table 29
With a With a
brick course brick
d=100 mm course
(Fig. 5) d = 100 mm
Corridor F 60 (Fig. 5) F 60 F 30 F 30
variant 1 According
to table 27, See table 29
row 2.1
Centre Bottom flange
distance plastered with
u = 15 mm 20–30 mm lime
from 30 mm cement (MGII)
reinforcing bar
to lower edge
(Fig. 5)
Corridor F 30 F 30 Classification F 30*
variant 2 See table 27, not possible
(top slab) row 1.2
With a
brick course
d = 100 mm
(Fig. 5)

* Load tests on reinforced brick floors of different construction types.

Table 1: Fire resistance classifications with reference to DIN 4102 as well as to the
evaluations of load tests

Load tests (experimental load-bearing capacity tests) on reinforced brick floors of different con-
struction types (particularly on the slab type Kleinesche Decke) were conducted by different
institutions in the last few years. They showed that these slabs can normally carry considerably
heavier loads than the loads principally calculated (Table 1). There are different reasons. For
example, for the calculations, slab bays are usually assumed to be simply supported between
the girders, whereas in practice, slab bays are often partially fixed. However, above all, the slab

bays carry a considerable part of the load due to

the arch action. Additionally, this observation is
supported by the fact that failure due to bending
occurred almost always simultaneously with the
crushing of the compression area in the present
case. As the slab bays enhance the structural per-
formance of steel girders due to their composite
structure, part of the brick ceiling can be consid-
ered to be an effective width of the floor girder.1
Therefore, it can be assumed that in the current
case, where the top flange (housing space and cor-
ridor variant 2) is merely inadequately confined,
Fig. 5: Core sample taken from the slab
there is no immediate danger of a floor collapse,
examined showing 100 mm bricks and
even in case of fire loading on the girder’s com-
30 mm reinforcing bar
pression area. There are several historical reports
of fire tests conducted on reinforced brick floors of the Kleinesche Decke. As a result, these slabs
and girders, which were mainly plastered with lime cement, were assigned a “high fire resist-
ance” at temperatures of at least 1000°C. Through load tests carried out after the fire tests, it was
also proved that the load-bearing capacity of the slabs remained unaffected.3 One of these reports
describes a fire test that was discontinued after 70  min while the slab remained undamaged.4
Unfortunately, these reports show the tendency for the Kleinesche Decke’s behavior only under
fire conditions. However, these reports cannot be used within the scope of current standards.

Summary and Conclusion

It is not possible to determine unambiguously the fire classification of the present Kleinesche
Decke built in 1897 by using the current German national standard DIN 4102—Brandverh-
alten von Baustoffen und Bauteilen (DIN 4102—behavior under fire).2 By comparing the given
example with the slab types documented in the DIN 4102, as well as by evaluating numerous
experimental load-bearing capacity tests on reinforced brick floors, it can be recommended that
the present slab bays should be categorized into class F 60 for parts of the corridor and into class
F 30 for parts of the housing space. Moreover, it is recommended that the present steel girders
should be assigned class F 30. Therefore, as a precaution, it is recommended that class F 30
should be applied for the entire system. The partial fixity of slab bays along the bearing area,
proven by experimental load-bearing capacity tests and, in particular, the arch action of the slab
bays, grant additional safety benefits with respect to a fire load [refs 5 to 9].

[1] Fischer M. Steineisendecken im Deutschen Reich 1892–1925, Cottbus, Dissertation, 2008.
[2] DIN 4102-4. Brandverhalten von Baustoffen und Bauteilen – Teil 4: Zusammenstellung und
Anwendung klassifizierter Baustoffe, Bauteile und Sonderbauteile Beuth: Berlin, 1994.
[3] Lange W. Konstruktive Neuerungen aus dem Gebiet des Hochbauwesens. Winter: Bremen,
[4] Kolbe E. Die wichtigsten Decken und Wände der Gegenwart. Kühne: Oberhausen, 1905.

[5] Ahnert R, & Krause KH. Typische Baukonstruktionen von 1860 bis 1960, zur Beurteilung
der vorhandenen Bausubstanz. Bauwesen: Berlin, 2001.
[6] Mittmann T, Bermes B. “Brandprüfungen an historischen Stahlsteindecken”. Sonderdruck
FeuerTrutz Magazin 5, 2012.
[7] Pötke W. Tragfähigkeitsnachweise an Stahlsteindecken für die Umnutzung alter Gebäude.
Selbstverlag: Berlin, 1998.
[8] Promat. Bauen im Bestand, Bautechnischer Brandschutz für Massivdecken. Eigenverlag:
Ratingen, 2012.
[9] Steller F, Fiedler, L.-D., & Quade J. Nachweis der Tragfähigkeit von Stahlsteindecken am
Beispiel der Leipziger Pianofortfabrik. Bautechnik 1996; 73: 8–14.


Reconstruction of the Neues
Museum in Berlin

Gerhard Eisele, Prof. Dipl.-Ing., Professor, University of Applied Sciences, Potsdam, Germany
Josef Seiler, Dipl.-Ing., Partner, Ingenieurgruppe Bauen, Karlsruhe, Germany

The Neues Museum in Berlin, Germany, was built between 1841 and 1859 under the manage-
ment of Friedrich August Stüler, a pupil of Karl-Friedrich Schinkel. Although designed with a
solid appearance to the outside observer, the building’s interiors were a compelling combination
of the latest cast-iron technology and ancient lightweight construction techniques. However,
shortly after its opening, the museum building began to show settlement damage. This settle-
ment activity persisted and reached a depth of 400 mm at the most unfavorable location. In 1943,
during World War II, the central staircase hall was bombed. The northwest wing, the Egyptian
courtyard and the southeast dome suffered the same fate in 1945. Some parts of the building
were left fully exposed to the elements for over 40 years. The significance of the remaining fab-
ric was recognized in the mid-1980s, prompting a decision to reconstruct the Neues Museum;
however, significant portions of the damaged building sections had to be demolished. Some
historic structures are unsuitable for a formal evaluation solely based on generally accepted
verification methods or modern codes and standards. For this reason, experimental methods had
to be considered at an early stage in order to support the structural stability analysis required for
the historic elements.1 The verification methodology was defined at the concept planning stage
and implemented experimentally in close cooperation with Prof. Steffens, Ingenieurgesellschaft
mbH (PSI), Bremen.

Aspect of Structural Design during Reconstruction

Following the early use of iron in buildings in the 1790s in England, the Neues Museum was
one of the first prestigious buildings in Berlin in which the use of iron became a distinct fea-
ture in both structural and architectural terms. Because of the issues that arose from the unsta-
ble ground, dead loads had to be minimized, and masonry ceilings built on “clay pots” were

thus used. Given the history of the building, most parts of the surviving original ceilings were
found to be in astonishingly good structural condition. However, real problems arose in some
of the floor bays where weather exposure had damaged the fabric or where significant sag-
ging occurred in central sections due to the removal of lateral supporting members. The client
requested that all ceilings should be upgraded to fulfill current requirements so that they could
withstand service loads of up to 5 kN/m2 in accordance with DIN 1055. Estimates and informa-
tion given in the original building documentation give rise to the assumption that load-bearing
capacities of 200 kg/m2 (2 kN/m2) were considered sufficient at the time of construction.2 The
structural design aimed to preserve the historic structures as a technical monument and to return
them to their original purpose, as undisturbed as possible, while also responding to the client’s
requests. The remaining original structures took priority so that the exhibition layout could be
adjusted to the specific situation as and when required. Structurally supporting members, such
as large sections of the iron structures, were evaluated using modern verification concepts. The
investigations revealed that some of these structures provided significant load-bearing reserves
for their original purpose.

Assessment of the Load-Bearing Capacity of Clay Pot

The concave design of the ceiling areas was based on an ancient technique that used cylindri-
cal, completely closed, hollow clay “bricks” laid
in gypsum mortar, which were referred to as clay
pots in the project documents. This method pro-
vides an extremely low weight-per-unit area at
remarkably high load-bearing capacities (Figs. 1
and 2).

The clay pots were manufactured with only

7–10 mm thick walls and lids. They were used as
bricks for various purposes, again held together
with a gypsum mortar. Among other features,
Fig. 1: Condition of the Calotte ceiling this construction method was used for the calotte
consisting of clay pots prior to reconstruc- ceilings consisting of transverse arches and
tion (courtesy: Ingenieurgruppe Bauen) brickwork pendentives interspaced with spheri-
cal domes of around 4.5 m in diam-
eter, barrel vaults with 5–6 m spans
(Fig. 3) and vaulted ceilings between
steel beams.

Structural Analysis
Clay pots h ~ 16 cm, ø 12 cm

Clay pots h ~ 20 cm, ø 13 cm In the initial step, the loads acting

on the existing clay pot ceiling sys-
Clay pots h ~ 23 cm, ø 14 cm
tems (calotte ceilings, barrel vaults
2–3 Solid clay bricks h = 26 cm
and vaulted ceilings) were estimated
Fig. 2: General structure of a clay pot vault based on preliminary calculations.

For this purpose, assumptions had to be made

regarding material characteristics, such as the
modulus of elasticity, load-bearing capacity and
stiffness of the system. All reference documents
found in the literature were used. The analyses of
a historical load test performed on a single-bay
vault3 were compared with the results of a cal-
culation using a three-dimensional (3D) finite-
element (FE) model. Parametric studies carried
out for structurally determinate subsystems were
used to vary the derived assumptions for the
material characteristics and to analyze the results Fig. 3: Soffit of new clay pot barrel vault
of the calculations. after removal of falsework (both illustra-
The basic findings from this work were as follows: tions courtesy: Ingenieurgruppe Bauen)

• The highest loads acting on the clay pot structures are in the barrel vaults.
• Due to the relatively low dead loads, the effects of loads that vary from bay to bay (“kin-
ematics”) must also be considered for the barrel vaults because they form multibay systems.
• The design can be interpreted as a combination of two load-bearing effects: a diaphragm
effect in the plane of the pot lids and a more flexible honeycomb structure in the cross-
• The service loads that the ceilings may potentially withstand cannot be determined without
supplementary experimental tests.

Experimental Tests
The findings of the structural analysis were used to develop the following program to estimate
the load-bearing capacity of the clay pot ceilings:

• Preliminary load test on an existing four-bay barrel vault to identify the system.
• Basic tests on small specimens with associated numerical analyses.
• Load test on a newly built sample vault.
• Development of an appropriate computation model and structural verifications for all clay
pot systems.
• Confirmation of the computation model, and thus of the load-bearing capacity, using load
tests in selected locations.
• Inclusion of the plaster floor forming part of the historic ceilings in the load-bearing capac-
ity analysis and evaluation of its contribution to the ceiling areas to be reconstructed.

Preliminary Load Test on an Existing Four-bay Barrel Vault

A four-bay barrel vault with spans of approximately 6 m was chosen for the preliminary load
tests. Two bays were alternately loaded in order to analyze the stiffness of the overall system,
deformation behavior, the modulus of elasticity of the clay pot mortar matrix, torsion of the

transverse arches, continuity effect and biaxial structural effect. Mobile load frames were used
to generate service loads that varied from bay to bay. Online measuring equipment and an asso-
ciated acoustic emission analysis (AEA) were used to ensure that the testing was nondestruc-
tive. The system for subsequent FE calculation was derived from the measured deformation and
deflection values.

Basic Tests on Small Specimens

In addition to conserving and repairing the existing ceilings, two of the original seven bays of
a barrel vault system were to be complemented by five new clay pot floor bays replicating the
original design. This meant that the clay pots and the gypsum mortar mixes had to be selected
in order to ensure that their mechanical properties and appearance largely corresponded to the
original materials. To achieve this goal, laboratory tests were conducted on about 30 small
specimens of various “pots” and mortars in different combinations and load cases. An AEA
of these small specimens was performed for calibration purposes. The materials ultimately
used for the reconstruction work were selected based on these tests. The tests and associ-
ated calculations with a refined model based on an elastoplastic material law were used to
prove the existence of a redundant structural system. After failure of the rigid outer shell (i.e.,
pot lids breaking away), the load shifts to the honeycomb structure formed by the pot walls
and the mortar within the cross-section; this is associated with a corresponding increase in

Sample Vault
Because the original vault could not be used for limit load tests, a new full-scale two-bay sample
vault was constructed within the building. This structure also provided an opportunity to test
specific techniques and work sequences. In the first test series, the effect of a partially load-
bearing plaster layer was investigated. A supplementary load-bearing plaster floor was included
in subsequent testing. The structure was loaded to failure in the course of the second series of
measurements and included the structural effect of the plaster. The results complemented the
values calculated for the existing and newly designed floor bays and ultimately confirmed that
the load-bearing capacities were sufficient for museum use.

Reinforcement of Cast-Iron Girders with CFRP Sheets

In keeping with the trend of the time, the original building was constructed with many cast-iron
components. This is unproblematic wherever such structural members are loaded mainly in com-
pression. However, cast-iron sections were also used as joists, between which the infill brick-
work could be inserted very easily to form vaulted ceilings. These joists are subjected to bending
with a marked tensile bending zone, which is why they also had to be verified for the service
loads required by modern use patterns. The tensile strength of the original cast iron amounts to
only 30–40% of its compressive strength. Inhomogeneities caused by the manufacturing pro-
cess (shrinkage cavities) weaken the tension zones, thus leading to a dangerous susceptibility
to brittle failure. Approximate calculations using design methods specified in the literature3
resulted in theoretically permissible service loads between 1.0 and 2.0 kN/m2 assuming undam-
aged structural components—quite optimistic given the history of the building. These values

are insufficient if the building is to be used as a

museum. Historic cast iron cannot be welded, and
bolted steel fish plates were not an option owing
to the shape of the elements. One possible solu-
tion was to reinforce the iron with carbon fiber-
reinforced plastic (CFRP) sheets in the tensile zone
(Fig. 4). This method is widely known in concrete
construction and complies with conservation
requirements because mechanical interventions
in the member to be reinforced are not necessary.
In addition, installation of the reinforcement is
reversible. No prior experience existed regarding
reinforcement measures for this combination of
materials. For this reason, a verification concept
was developed jointly with the Institute of Con-
crete Construction and Fire Protection at Braun-
schweig University of Technology. This concept
is based on the method used for unreinforced
beams developed by Frey and Käpplein.3

The first step was to obtain basic information on

Fig. 4: Section of ceiling above the ves- the behavior of cast-iron beams. New beams were
tibule consisting of clay pots between cast for this purpose. In our high-tech society, it
cast-iron girders (300 mm deep). Below: is not at all easy to produce grey cast iron to the
detailed section showing CFRP sheet rein- lower quality standard equivalent to the original
forcement underneath the historic soffit condition. The beams were partially damaged
lining made of zinc (courtesy: Ingenieur- prior to testing and subjected to experiments to
gruppe Bauen) determine their load-bearing capacities in the
reinforced and nonreinforced condition. Based on
the test results, a realistic design method was developed that proved a sufficient degree of postre-
inforcement safety, even for previously damaged cast-iron beams. The reinforcing CFRP sheets
were selected and dimensioned on this basis (Fig. 4): they are 1.4 mm thick and 50 mm high.
For verification purposes, on-site confirmation tests were performed in selected trial bays after
installation of the reinforcement. These tests provided impressive proof of its effectiveness. The
specified service load of 5 kN/m2 was thus confirmed for all historic ceilings. References [4, 5]
give a more detailed account of this section.

Increasingly, construction engineers need to strike a balance between historic structural designs
and modern use requirements. Although these challenges are usually mastered (virtually any-
thing is possible from a purely technical point of view), thorough understanding of historic
structures and conservation objectives is not widespread. However, such an understanding is
key to the successful treatment of buildings with high conservation significance. In the case
reported in this paper, the most important responsibility of the engineers was to recognize, at an
early stage, that purely theoretical approaches would not deliver the intended result and that all
parties involved in the project—such as the client, user, architect and required experts—needed

to engage in the discussions, which also had to be facilitated. Key services that needed to be
provided included the development of verification concepts, the planning and design of the
required tests in a cross-disciplinary approach and coordination of all the activities with the
supervising authority.

[1] Steffens K. Experimentelle Tragsicherheitsbewertung von Bauwerken. Grundlagen und An-
wendungsbeispiele. Ernst & Sohn: Berlin, 2001 (in German).
[2] Hoffmann CW. Die feuerfesten Decken des Neuen Museums hierselbst. In Notizblatt des
Architektenvereins Berlin, 1846 (in German).
[3] Frey A, & Käpplein R. Beitrag zum rechnerischen Nachweis der Tragfähigkeit alter
Biegeträger aus Gusseisen. Stahlbau 1993; 62(8): 221–230 (in German).
[4] Eisele G, Gutermann M, Seiler J, & Steffens K. Wiederaufbau des Neuen Museums in
Berlin 2 Tragwerksplanung pro Baudenkmalpflege. Bautechnik 2004; 81(6): 407–422 (in
[5] Eisele G. Tragfähigkeitsbewertung an Natursteinsäulen am Neuen Museum in Berlin.
In Natursteinsanierung Stuttgart 2006. Neue Natursteinrestaurierungsergebnisse und
messtechnische Erfassungen, Grassegger G, Patitz G (eds) Fraunhofer IRB: Stuttgart, 2006;
19–28 (in German).
[6] Lorenz W. Das Neue Museum Berlin (Historische Wahrzeichen der Ingenieurbaukunst in
Deutschland, 15) Bundesingenieurkammer: Berlin, 2014 (in German).


The Necessity for Construction
History to Assess Historic
Bridge Bearings

VolkerWetzk, BTU Cottbus-Senftenberg, Chair of Construction History and Structural

Preservation, Cottbus, Germany

With existing bridge structures having to be maintained or refurbished, we find that historic steel
bridge bearings are reentering the engineers’ perception. Due to their sturdiness, they are mostly
found to be in sound condition despite having served for more than a century, often negligently
maintained and carrying increasing loads. If still in use, however, we have very little reliable infor-
mation as to their structural assessment. This uncertainty often leads to the costly replacement of
bearings that look intact. Such premature exchanges contribute to a permanent loss of historic fabric.

Historic Bridge Bearings

With hindsight, we can recognize three generations of bridge bearings, which are distinguishable
with regard to their mode of action and to their material: wood, iron/steel and synthetic materi-
als. This article deals with the second generation, that is, those made of iron and steel. They were
introduced in the same period that saw iron being widely used for bridges, which allowed for new
dimensions in bridge building that were hardly dreamt of only few years before. Rollers, pins
and pivots came to characterize this generation of bearings in a very short time; these prominent
features, together with the optimized bearing geometry, give this generation its specific aesthetic
value. Initially, the bearings were used to provide movability in a longitudinal direction, and
subsequently in the lateral direction too. Occasionally, we do not find any fixed bearings in these
bridges at all.1 This second generation is characterized by two main materials—cast iron and steel,
the latter produced by casting or by forging. The compact bearing plates with complex geometri-
cal designs were initially made of cast iron. However, bearing technology was an early instance
of appropriating steel castings. Beginning in the 1870s, for both tilting and bearing plates, steel
castings became a serious competitor for the last domain of iron castings in bridge building. By
1880, steel castings were ranked as excellent, albeit expensive, substitutes for cast iron. From
1920 onward, the former had fully replaced cast iron. Historic bearings still in use tend to be

(a) (b)

Fig. 1: (a) Moveable bearing of the second Dirschau Bridge (1889–1891). The pivots and stilts
with a simple geometric design for absorbing heavy loads were forged from expensive crucible
steel ingots, whereas the other main parts of the bearings were already Martin steel castings. (b)
Fixed bearing at Hamburg Brooks Bridge (1887) with stiffened ribs ordinarily used2

made of steel castings. Establishing steel casting also resulted in a specific geometry for the cast-
bearing parts. Initially, the molds developed for iron casting were used, but soon, foundry practice
required castings with an almost equal thickness. Plates with stiffened ribs became characteristic
for bearings cast with steel (Fig. 1). This measure was to prevent large shrinkage holes. Neverthe-
less, massive bearing castings without ribs were used well into the 20th century. If the casting was
performed incorrectly, such bearings might contain significant shrinkage holes.2

Structural Assessment of Historic Bearings

In compliance with the practice in structural engineering, assessing historic bearings requires
knowledge to evaluate their load-bearing capacity and serviceability. Because of the particular
task of moveable bearings as part of the entire structure, however, the issue of serviceability is
as important as that of the load-bearing capacity: A single nonfunctional bearing—regardless of
the individual cause—has to count as a failure of the entire bearing, and this in turn may entail
serious consequences for the level of safety in the substructure and/or superstructure. According
to the ICOMOS-Charter of Victoria Falls,3 the knowledge of the fabrics’ history is extraordi-
narily relevant to the structural assessment of historic fabric. Such an approach might be time-
consuming; nevertheless, it is essential for producing the relevant qualitative and quantitative
knowledge. This holds true for the assessment of historic bearings as well. Most quantitative
statements will be based on material testing performed on the bearing on-site (Fig. 2). For both
choosing the correct methodology and evaluating the results, knowledge of the historic practices
with regard to technology and engineering procedures provides indispensable guidance.

Assessment of Load-Bearing Capacity

Dependable statements concerning the load-bearing capacity of bridge bearings require informa-
tion about their mechanical properties as well as the textural properties of the bearing material.
Knowledge about the material’s texture can be obtained on site without any problems. Experi-
ence shows that surface flaws are very rare, whereas indicators for the existence of volumetric

Fig. 2: Material and textural investigation at Teltow-Kanal Bridge, Berlin; from left to right:
Hardness measurement, magnetic crack testing, ultrasonography investigation4

flaws are found much more often (Fig. 3). Adopting a historical perspective may help understand
the problem behind such flaws. Early 20th-century literature—when bearings made from steel
castings were most popular—addressing technological issues reveals that equalizing shrinkage
to avoid holes during production was very complicated.5 However, if compared with foundry
products required at that time for applications in mechanical engineering, the shape casting of
steel-bearing elements was not a big challenge. This may have led to something of a “routine
attitude” among foundry men, when it came to casting bridge bearings that led to imprecisions
during production, which in turn could cause significant volumetric flaws.2

This means that a responsible structural assessment of volumetric flaws is particularly signifi-
cant. Experience yields a first and comforting observation: even bearings with a considerable
shrinkage hole have performed reliably and coped with stresses for many decades. In addition,
it turns out that a shrinkage hole mostly occurs in the core region of a cast part during produc-
tion. In most cases, they are small and thus insignificant with regard to the overall dimension
of the bearing part and the distribution of stresses. For example, in the case of the shrinkage
hole shown in Fig. 3, the flaw was only found to exist in the middle fifth of the plate’s longi-
tudinal direction. Knowledge about the material’s mechanics can only be gathered in situ by
means of nondestructive testing procedures; tests concerning hardness measurements revealed

(a) (b)

Fig. 3: (a) Moveable bearing cast c.1939. (b) Slice taken from bearing plate (dashed in left fig-
ure). In spite of the shrinkage hole, plastic deformations occurred in the contact region between
plate and roller (dashed in right-side figure)— a clear indicator of the very high level of overall
stiffness of the bearing plate4

(a) (b)

Fig. 4: (a) Laboratory microstructural investigation in the peripheral area of a historic bearing
made of steel casting; the decarburization close to the surface can be seen as well as the outer
layer of tinder.8 (b) Surface of a bearing plate made of steel casting prepared on site for exami-
nation of material. The polished layer of tinder is visible (middle) as well as the removed layer
close to the periphery (right) (Photograph courtesy: Ref. [4])

the potential of the Leeb method for assessing the tensile strength of historic bearings material.6
The extraction of a small sample for further tests in the laboratory is only possible in single
cases. Successful nondestructive access to the bearings presupposes detailed knowledge of how
steel castings were produced in different periods. For normalizing the brittle cast texture, all
casting blanks were subjected to an extended heat treatment. For a complete textural transfor-
mation, unalloyed cast steel had to be annealed at a temperature depending on the carbon share
of the material. The entire process of the thermal treatment, however, was only scientifically
proven as late as 1915. This means that in the case of early steel castings with low carbon share,
a complete textural transformation as part of the thermal treatment at production is unlikely.7 A
negative side effect of this heat treatment was the decarburization of the peripheral area, which
can be distinguished in Fig. 4a by the increased share of ferrite toward the edge. Furthermore,
the annealing yielded a thin, but very hard, layer of tinder enveloping the comparably soft ferri-
tic margin layer (Fig. 4a).7 Grinding down this peripheral area is a precondition for successfully
applying nondestructive tests. Further treatment may be required depending on the examination
procedure selected. For example, if hardness is to be measured, the surface has to be sanded in
conformity with the applicable standards (Fig. 4b).

Assessment of Serviceability
Any assessment of the serviceability will mostly concern moveable bearings. The main focus is
on the contact zone between the rolling elements and the bearing plates as any plastic deforma-
tions of these areas determine the rolling resistance of the bearing. Since the introduction of this
bearing generation, a great body of research has addressed this issue. It has also been central
to a recent research collaboration between BTU Cottbus and BAM Berlin, yielding a number
of publications, which also address the historical development of the contact issue.4,9,10 History
provides useful insights when it comes to assessing the contact issue and thus the serviceability.
Suffice to say that the rolling elements in the 19th century were more or less dimensioned based
on empirical procedures and provide load-bearing capacity reserves if compared with roller
sizes arrived at via current design approaches (Fig. 5). Furthermore, history reveals that the
careless attitude toward maintaining bridge bearings, often complained about today, dates back

Cast Iron
80 Cast Steel
Required Geometry of Rollers

70 Wrought Iron and Steel

60 High-Strength Steel

Werder, 1857

Schwedler, 1861
Köpcke, 1869
Fränkel, 1869

Baldermann, 1871

Winkler, 1875
Baentsch, 1876
Laissle, 1876
Steiner, 1882

Haeseler, 1888

Reuleaux, 1889
Tetmajer, 1889
Bach, 1889
Deslandres, 1893
Weyrauch, 1894

Kübler, 1900
Tetmajer, 1905
Vianello, 1905
Schaper, 1908
Bernhard, 1911

Kollmar, 1914
Schaper, 1922
B.E., 1922
Minister, 1923
B.E., 1925

DIN 1050, 1937

DIN 18800-1, 1990
Fig. 5: Development of the rollers’ radii according to the calculations in Germany4

to the early period of employing iron and steel bearings. Apparently, owners of bridges have not
been aware of the machine-like nature of bearings that require regular monitoring and at least
some cleaning and maintenance.11 Without these, the loss of serviceability may lead to critical
levels of safety in the sub- and/or superstructure.

Value of Construction History

Knowledge about the history of bearings as well as about the general conditions of technology
and engineering of specific historic periods is essential for assessing the structural quality of
bridge bearings made of steel. It provides the base for a successful approach for both the analy-
sis of materials and textures as well as their appropriate evaluation by engineers.

[1] Wetzk V. Brückenlager. 1850–1950. Dissertation Thesis BTU Cottbus. https://opus4.
kobv.de/opus4-btu/frontdoor/index/index/docId/2006, 2010.
[2] Wetzk V. The use of steel castings in mechanical and civil engineering—Germany.
1850–1950. In Proceedings of the Fifth International Congress on Construction History,
Friedman D, Bowen B, Leslie T, Ochsendorf J (eds), Chicago, 3–7 June 2015, vol. 3,
2015; 585–592.
[3] ICOMOS. ICOMOS-Charter—Principles for the Analysis, Conservation and Structural
Restoration of Architectural Heritage. Charter of Victoria Falls. Victoria Falls, Zimbabwe,
[4] Wetzk V, Eisenkolb T, & Mehdianpour M. Bewertung historischer Brückenlager – Das
Kontaktproblem – Teil 1. Stahlbau 2011; 80(6): 404–412.
[5] EMPA. Schweizer Verband für die Materialprüfungen der Technik (Ed.): Der Stahlguss
als Baustoff, Bericht Nr. 36, Zürich, 1929.

[6] Wetzk V, Schüle P. Assessing historic bridge bearings—the potential of hardness measure-
ments for characterizing cast steel. In Proceedings of the Tenth International Conference
on Structural Analysis of Historical Constructions, Balen, K v, Verstrynge, E (eds), Leu-
ven, 13–16 September 2016, vol. 3, 2016; 828–831.
[7] Wetzk V. Historic bridge bearings—material research on cast steel. In Proceedings of
the Fourth International Congress on Construction History, Campbell J, Pan Y, Baker N,
Driver M, Heaton M, Tutton M, Yeomans D, Rosoman T (eds), Paris, 3–7 July 2012, vol.
3, 2012; 243–252.
[8] BAM. Microstructural Investigation of Bearings made of Steel Castings. Without report,
[9] Eisenkolb T, & Mehdianpour M. Bewertung historischer Brückenlager – Das Kontakt-
problem – Teil 2. Stahlbau 2012; 81(7): 510–518.
[10] Eisenkolb T, & Mehdianpour M. Bewertung historischer Brückenlager – Experimentelle
Untersuchungen zur Rollreibung. Stahlbau 2013; 82(12): 890–897.
[11] Wetzk V. Editorial to themed issue on bridge bearings. Proceedings ICE—Engineering
History and Heritage, vol. 168, EH2, 2015; 53–54.


Refurbishing of the Nibelungen
Bridge in Worms, Germany

Eberhard Pelke, Head of Structural Engineering Department, Hessen Mobil, Wiesbaden, Germany

State of the Structure and Historical Assessment

The original bridge was opened to traffic in 1900. It consisted of three, two-centered steel arches,
with the carriageway superimposed, which were supported on pillars with caisson foundations.
The shore bridges are solid, three-centered arches made of tamped concrete faced with natural
stone. The present Nibelungen Bridge in Worms forms part of federal highway 47 and links the
municipalities of Worms and Bürstadt. The bridge has a total length of 745 m and is divided into
three part structures of differing construction consisting of the same 109 m long shore bridge
on the left side of the Rhine in Rhineland-Palatinate, the 351.8 m long center part of the bridge
crossing the Rhine (Fig. 1) and the shore bridge on the Hesse side of a total length of 295.5 m.
The current traffic volume amounts to some 23 000 vehicles every 24 h, with a moderate heavy
vehicle share of 8.7%.

In the case of the shore bridges, the 12 three-centered arches made of tamped concrete are
still in existence. Solid sandstone faces the front side of the 11.4 m-wide arches and the areas
directly in front of the hinges. The thickness of the arch lies between 0.67 m at the apex and
c. 1 m at the quarter points of the arches. The arches are designed as compound curves with a
radius of between 33 m in the case of the largest arch and 26 m in the case of the smallest one.
The transverse prestressed carriageway plate made of B 450 quality concrete, incorporated in
1953, is supported over the whole surface in the crown area and above the masonry segments of
the arches in the other areas. It is connected three times above the crown and impost hinges in
each arch. Traffic used to drive directly on the carriageway plates, as was quite usual for the first
prestressed concrete bridges. Between 1972 and 1974, the bridge was completely sealed using
a mastic seal, and in the 42 transverse joints, single-seal expansion joints were provided. In the
area of the shore bridges, the expansion joints had to be extensively touched up in 1981. Other
preservation measures are in keeping with normal maintenance practice. In his lecture at the
general meeting of the German Concrete Association (DBV) in Berlin in May 1952, Finsterwal-
der1 himself presented the fundamental design idea of cantilevering in prestressed concrete that

Fig. 2: Finsterwalder’s design idea

from 19312

dated back to the year 1930. On the occasion

of the competition for the construction of a
new Dreirosen Bridge in Basle, his company,
Dyckerhoff & Widmann, had proposed two
Fig. 1: Nibelungen Bridge, Worms, 1953 half arches connected rigidly above the pillar,
the crowns of which were to be short-circuited
via a tension member made of 60 mm thick wire ropes without connection. The prestressing was
generated independently here by the dead weight (Fig. 2).

The further historical development of prestressed concrete bridges and the outstanding position
of the Nibelungen Bridge are acknowledged in Ref. [3]. Around 1948, Finsterwalder4 found the
right material for his single prestressing bar in the self-hardening St 90 steel, known as St 60/90
shortly afterward, from the Krupp Rheinhausen steel works. With nearly double the tensile
yield strength, the steel was unsusceptible to brittle fracture and stress corrosion. Finsterwalder
succeeded in rolling a screw thread onto the approximately 26  mm thick rods, thus making
them suitable for anchoring and connections, which he joined ingeniously simply using plates,
sleeves and screws. Furthermore, Finsterwalder’s single bars had the advantage that there was
nearly no slippage when the prestressing jacks were removed, which was the prerequisite for
cantilevering in short sections.5 Tied to the position of the caisson foundations of the destroyed
steel bridge, the Nibelungen Bridge crosses the Rhine at Worms in three large spans with effec-
tive widths of 101.65, 114.20 and 104.20  m. The cantilever girders protrude like monoliths
from the reinforced concrete hollow pier, following Finsterwalder’s idea formulated in 1931
of a “cantilever whose vault thrust is cancelled out by tensioned, straight cables”.2 Two hol-
low boxes connected through the carriageway plate with structural depths of between 6.5  m
at the pier cut and 2.5  m in mid-span form the 13.5  m wide cantilever girder cross-section
(Fig. 3). A vertically prestressed joint, transmitting transverse forces, with stilted cast steel roller
bearings avoid any reciprocal displace-
ment of the cantilever girder ends.

While the superstructure grows out of the

river piers simultaneously on both sides in
sections of 3 m toward the middle of the
span, thus keeping the moments of differ-
ence to be absorbed for the subsoil small,
the end piers had to be ballasted to control
the overturning moment from the grow-
ing superstructures or were hung back
Fig. 3: Cross-section of Finsterwalder’s longitu- via the prestressed structural components.
dinal and transverse prestressing This was helped by the existing founda-

tions that had absorbed the horizontal thrust of the destroyed steel arch bridge. “The key to the
construction” Finsterwalder1 observed. Finsterwalder used his well-tried St 90 single bars for
prestressing. He graded the prestressed reinforcement for the superstructures cantilevering out
up to 57.1 m in accordance with the reduction of the bending moment toward mid-span. The
haunch helped him keep the shearing stresses constant over the length of the structure. Under
full load, with the most unfavorable live load conditions, Finsterwalder allowed for a tension
wedge that was later called limited prestressing. In total, 486 bars with a diameter of 26 mm
were necessary in order to be able to resist the fixed end moment of the full load at center piers
(Fig. 3). At the end of each section, on average, 24 bars were stressed and anchored.1 One verti-
cal reinforcement tendon per web directly alongside the cantilever joints absorbed the tensile
splitting forces from the anchored tendons. The concrete grade of B 450 was not only a technical
but also a logistical challenge for the first postwar years. After 22 months of construction, the
bridge was opened to traffic on 30 April 1953.

In the opinion of the early prestressed concrete engineers, no tensile stresses should be permitted
in prestressed concrete components as far as possible. This led to prestressed stirrups in order to
avoid principal tensile stresses. Creeping and shrinking were known; however, engineers were
still uncertain about how to determine prestressing losses. Thus, assumed global reductions of
120–150 MPa were usual until the end of the 50s, which went back to the Frankfurt Trials by
the Neue Baugesellschaft Wayss & Freytag in 1936. As there was no danger from salt used for
thawing ice, friction and not protection from corrosion, was the central topic for consideration
during construction with subsequent bonding. Limiting the concrete tensile stresses in a trans-
verse direction by an appropriate choice of system or transverse pretensioning led the engineers
of that time to the conviction that the concrete would remain crack free and not require any seal-
ing. Following Freyssinet in this tradition, the Nibelungen Bridge is prestressed three dimen-
sionally; however, as a result of his lower high tensile strength prestressing bars, Finsterwalder
did not achieve full prestressing for every load position. The design of the bridge was based on
the seventh draft of DIN 4227 but also included the discussion process still in progress. With the
exception of the more precise proof of shear reinforcement in the case of exceeding the permis-
sible main tensile stresses, the seventh draft was essentially like the definitive first German pre-
stressed concrete standard. A minimum shear reinforcement was introduced with the additional
provisions for DIN 4227 by the federal authorities 12 years later, in 1966. With the haunched
cross-section design, the principal tensile stresses could be maintained nearly constant and were
limited to omit any reinforcement. Nevertheless, Finsterwalder ordered a low shear mesh rein-
forcement and mild longitudinal and transverse reinforcement all over the whole cross-section.

Steps to Refurbishment
The planning for the refurbishment of the Nibelungen Bridge was subdivided into five sections:

1. Fundamental expertise on the state of knowledge and possibilities of application of the

prestressed concrete method of construction at the time of execution.
2. Technical building material examinations as well as examinations of the tendons and hinge
prestressing of the center part of the bridge.
3. Checking of the bridge with the findings from and preparation of variants for strengthening
the Nibelungen Bridge into German bridge class 60/30, satisfying present day traffic.
4. Refurbishment design.

5. Follow-on special tests and examinations during building work to evaluate design in 2.4.

Assessment of the State

The state of the center part of the bridge is outlined as follows6:

• Favorable crack situation without shear cracks

• Good concrete quality, however, without subsequent hardness
• Usual contamination with chloride and degrees of carbonating
• Flaking of concrete only locally in the area of the crown elements as well as humidity pen-
etration on the wearing parts of the expansion joints
• Locally inadequate degree of grouting of the tendons due to the then-too small ducts and the
too narrow construction of the tendon sleeve coupling
• Tendons of the tieback to compensate for the cantilever moment on the land pier on the
Hesse side inadequately grouted and corroded
• Severe corrosion and wear phenomena on the crown hinges coupling the individual cantile-
ver systems and their tie rods endanger traffic safety
• Unsuitable drainage system (open gutter inside the hollow box along with free fall system)

The durability of the shore parts of the bridge is reduced to a greater extent through the defective
expansion joints fitted later and the defective drainage system. The jointed carriageway plates
are no longer a match for today’s aggressive environmental influences.

Refurbishing Measures
The refurbishing of the center part of the Nibelungen Bridge is based on an economic viability
check and on the finding that the bridge’s structural stability is not endangered. This enables a
structural strengthening without reinforcing any shear deficits to avoid any structural damage to
the longitudinal existing prestressing bars. Its extent can be outlined as follows:

• Cleaning of the dirty parts (pier and crown hinges soiled by bird droppings)
• Demolition of the railings, crash barriers, lighting, curbs and carriageway surfacing
• Check of the defective building material quantities after demolition of the sealing and be-
fore awarding the contracts for further works
• Examination of the tendons, including the regrouting
• Reinforcement of the superstructure by means of straight, external tendons (Figs. 4 and 5)
• Substitution of the existing stilt bearing to spherical bearing, including bracing the hinge
• Renovation of the existing expansion joints in the crown hinge area, as well as in the transi-
tion to the separating piers
• Renovation of the existing drainage system
• Repair of the concrete in the damaged areas
• Renovation of sealing and surfacing, curbs, crash barriers and lighting.

Fig. 4: Reinforcing the cantilever beams with external longitudinal tendons6

Fig. 5: Detail of the anchorage of the external additional prestressing tendons: (a) cross-section
of a hollow box at end of the cantilever; (b) detail of the anchorage block with transverse pre-
stressing bar St 950/1050

With the choice of lightweight concrete, as a result of the continuous carriageway plate of the
shore part of the bridge, the restraints are clearly diminished as both the coefficient of thermal
expansion and the dead weight of the carriageway plate are reduced, and thus, the deformation
of the plate and the horizontal force are decreased as a result of the friction. Expansion joints
are only envisaged at the plate ends. The carriageway plate is given sliding supports in the outer
spans by the arrangement of fibrated concrete slabs between the superstructure and reinforced
concrete head beams on the masonry. The renovation of the curbs, seal, surfacing and vehicle
restraint systems; the repair of concrete and masonry surfaces; and the fitting of a closed drain-
age system complete the measures for the shore parts of the bridge. In the land pier on the right
bank of the Rhine, the cantilever moment is dissipated into a force couple via cross girders
through connecting compression columns and tiebacks. The tiebacks consist of two edge ten-
sion members with 14 single prestressing bars, each with a diameter of 26 mm, and a middle
tension bar made up of 34 single prestressing bars. The tendons were detected with the highest
degree of corrosion and grouting error damage. Monostrands had to be added and anchored by
means of core drill holes in the foundation and the upper tieback transverse girders. Recesses
in the middle of the tieback accommodate the tensioning wedge for prestressing the additional

tendons. Further monostrand tendons serve to strengthen the tieback transverse girder. Further
measures for both land piers, such as repairing the natural stones and their joints, improvement
in accessibility or renovating damaged manhole covers, are routine tasks. In conclusion, the land
pier on the Hesse side will be given a new, external staircase tower.

[1] Finsterwalder U. Bau der Strassenbrücke über den Rhein in Worms. Beton- und Stahlbeton-
bau 1953; 48(1): 1–5.
[2] Dischinger F. Elimination of bending tensile stresses in R.C. Bridges. IABSE, Second Con-
gress Munich, Pre-Report. IVb2, 1936; 775–798.
[3] Pelke E. Pre-Stressing of Bridges in Germany up to 1965 Part I and Part II. Proc. Inst. Civil
Eng., Eng. History Heritage 2011; 164(2/4): 99/211–108/218.
[4] Finsterwalder U. Dywidag-Spannbeton. Bauingenieur 1952; 27(5): 142–158.
[5] Kern G. Das DYWIDAG-Spannverfahren. In Dyckerhoff & Widmann A-G. München
(Herausg.) Festschrift Ulrich Finsterwalder 50 Jahre für DYWIDAG G. Braun: Karlsruhe,
[6] Pelke E, & Zichner T. Ertüchtigung der Nibelungenbrücke Worms. Beton- und Stahlbeton-
bau. 2015; 110(2): 113–130.


Early Prestressed Steel–Concrete
Composite Continuous Bridges in

Y. Rammer; B. Espion; Professor, Université Libre de Bruxelles, Brussels, Belgium

Two Belgian post-tensioned composite steel–concrete continuous bridges built in the 1960s
are currently undergoing heavy repair due to corrosion of the main steel girders and severe
degradation of the concrete slabs due to deicing. The bridges are located on important European
highways. The authors were involved in the assessment of the residual bearing capacity of the
cantilever sidewalks of one of these bridges during temporary deviation of the traffic on the
bridge. This assessment was undertaken before the renovation started, and the authors discov-
ered that the highway administration implemented the compression stress values that were com-
puted in the original calculation notes. This puzzled the authors and led to the need for further
historical research about this original, but now obsolete, continuous bridge construction method
that actually induces high prestressing losses.

Only recently was the historical development of composite structures taking into account the
connection of the steel members with concrete parts fully addressed:1 during the postwar period
(1950–1975), referred to as “the classical phase” by those authors, standardized testing and
calculation methods resulted in numerous bridges and buildings. Continuous and cantilever
composite bridges are affected by hogging bending moment regions in the vicinity of interme-
diate supports, causing tensile stresses in the deck concrete slab and affecting the durability of
the deck due to flexural cracking.2 Initial ideas for avoiding these tensile stresses involved the
introduction of some kind of “precompression” in the concrete deck slab, either by (a) jacking
the supports or (b) prestressing the concrete slab by tendons (c) before or after connecting the
steel and concrete parts of the structure. Creep and shrinkage effects were, already at that time,
known to induce redistribution of stresses between the steel girders and the concrete deck slab.
Regarding continuous bridges, theoretical development seems to have started after World War
II,3 and presumably, the first full-scale bridge of that type was constructed across the Ruhr River
in Herdecke.4 This road bridge has three spans (48.2 m + 64.2 m + 48.2 m = 160.6 m) with an
overall uniform depth of 2.4 m and a deck width of 17.8 m (five steel I girders of 2.0 m depth

and a concrete slab of about 400 mm). The sequence of construction was: (a) erecting the steel
structure on temporary supports; (b) lowering the supports in order to introduce some “prebend-
ing” in the steel; (c) installing the deck slab, placing the reinforcement and tendons; (d) lowering
the supports in order to introduce compression in the concrete slab; and (e) post-tensioning of
the STUP-Freyssinet 12ϕ5 tendons (the contractor was Wayss und Freytag, and the 12ϕ7 were
not available at that time). The bridge was opened in March 1951. The fact that this bridge was
equipped with STUP tendons gave the authors the motivation to look further into the French
literature and to try to discover when the original idea of prestressing continuous composite
bridges arose. The first occurrence was found in a report quoting tests conducted by the SNCF
(the French National Railway Company) on composite bridges : “Des tabliers continus en béton
armé, associés à des poutres métalliques peuvent être construits, à condition d’annuler, par
précontrainte du béton, les tensions aux droits des appuis. Cette précontrainte peut être obtenue
simplement par dénivellation d’appui”.5 At the end of World War II, Ridet, referring to a con-
ference held in May 1944, stated that STUP-Freyssinet presented an alternative proposal for a
two-span bridge (two spans of 35.9 m) at Donchery (Fig. 1). Longitudinal tendons were to be
placed on the top and bottom of the slab.7

This 1944 proposal was developed and further explained in an article on continuous composite
bridges,6 using post-tensioning to introduce compression stress in the concrete deck slab (before
live loads) of 9 MPa (Fig. 2).

Herstal and La Louvière Bridges

In Belgium, post-World War II economic reconstruction was fast as the public and industrial
infrastructure was hardly damaged. However, after the Universal Exposition in 1958, the country

Fig. 1: Donchery Bridge elevation6

Fig. 2: Donchery Bridge cross-section6


entered a period of economic, social and political unrest. Major infrastructure programs were
delayed. However, some projects were pursued, including the two bridges at Herstal (Fig. 3) and
La Louvière (Fig. 4), which are currently undergoing heavy rehabilitation (Table 1).

The Herstal Bridge (no. OA27) overpassing the Meuse River has a similar, but shorter, adjacent
bridge crossing the Albert Canal (no. OA26). Both are, in fact, twin decks (one in each direc-
tion). The structure is statically determinate. The reasons for this are possible mining subsidence
(Fig. 5). The sequence of construction was: (a) erecting the steel structure on temporary supports
(Fig. 3); (b) concreting the cantilever central span; (c) lowering the cantilever span temporary

Fig. 3: Herstal Bridge during

construction8 Fig. 4: La Louvière Bridge during construction9

Herstal (Bridge OA27) La Louvière (Bridge nr 34)

Spans 3 (65 m + 110 m + 65 m = 240 m) 3 (50 m + 109 m + 50 m = 209 m)
Type of span Continuous (40 m cantilever span) Continuous
Calculation Statically determinate Statically indeterminate
Geometry Variable depth (2–5.5 m) Variable depth (2–5.5 m)
Width (of each parallel 12.75 m 18.20 m
twin bridges)
No. of steel girders 5 6
Depth of slab 230 mm 190 mm
Formwork for slab Wood on scaffolding Precast concrete planks
Method of erection Temporary supports Incremental launching
Longitudinal tendons 224 STUP-Freyssinet 12ϕ7 27 STUP-Freyssinet 12ϕ7
Transversal tendons STUP-Freyssinet 12ϕ7 per N/A
600 mm
Position of tendons Inside slab Under slab with deviators
Shear connection Friction by transversal post- By shear connectors
Jacking/prebending Yes Yes
Opening to traffic 1964 1967
Designer Alexandre Birguer Antoine Van Der Vloet
Steel contractor Ateliers de Jambes-Namur Ateliers de Jambes-Namur
Table 1. Main bridge data

Fig. 5: Herstal Bridge longitudinal section and elevation; note the 40 m suspended span (points
C and D are hinges)10

Fig. 6: Herstal Bridge transversal section (original drawing from Birguer office, Administration

Fig. 7: La Louvière Bridge : transversal section11

supports in order to introduce a “prebending” 100 ton force in the steel; (d) installing the deck
slab, including placing the reinforcement and tendons; (e) transversal post-tensioning (Fig. 4)
for shear connecting; and (f) longitudinal post-tensioning of the STUP-Freyssinet 12ϕ7 ten-
dons. Anchor heads are situated under the deck slab for longitudinal tendons and embedded in
the sidewalks for the transversal tendons (Fig. 6).

The La Louvière road bridge also has three spans with a similar shape, but the bridge is strictly
continuous and statically indeterminate. The sequence of construction was: (a) erecting the steel
structure with a cantilever construction (Fig. 4), (b) lowering the end supports in order to intro-
duce some prebending in the steel, (c) installing the deck slab using precast concrete planks
and placing the reinforcement, (d) installing the tendons under the deck, (e) jacking up the

end supports in order to introduce some compression in the concrete slab (f) and, finally, post-
tensioning of the STUP-Freyssinet 12ϕ7 external tendons protected by plastic ducts (Fig. 7).
Tendons are slightly deviated under the deck by means of steel profiles connected to the main
steel girders. Anchor heads and blocks are situated under the deck slab.

The examples studied suggest that the idea of prestressing continuous composite beams should
probably be attributed to the STUP Company during World War II, not very long after the devel-
opment of Freyssinet’s 12ϕ5 anchorage and tendon. However, the construction method of the
four bridges mentioned in this paper (Donchery, Herdecke, Herstal and La Louvière) is quite
complex due to the mixed use of supports jacking and post-tensioning sequences. Designers at
that time were not aware of the prestressing losses over a long period or of the time-dependent
effects during construction sequences on the redistribution of stresses between steel and con-
crete.12 When investigating such bridges with the view of extending their lifetime, engineers
should have sufficient knowledge of “old” calculation methods and standards in order to identify
discrepancies in actual knowledge of the time-dependent structural behavior. Further research is
necessary to try to understand how those complex structures (i.e., the different phases, includ-
ing partial strutting, jacking, lowering and tensioning) were calculated and, furthermore, what
would be the results of modern calculations of prestressing losses and redistribution. Today’s
techniques of expertise however leave one essential issue unresolved: how can the actual resid-
ual prestressing in the structure be assessed?13

The authors acknowledge the assistance of ir.Pierre Gilles and ir.Françoise Taquet from the Wal-
loon Administration for allowing convenient access to their archives for these bridges.

[1] Pelke E, Kurrer K-E. On the evolution of steel-concrete composite construction. Proceed-
ings of the Fifth International Congress on Construction History, Chicago, 2015.
[2] Ernens M, Cremer J-M, Dotreppe J-C. Cracking and durability of concrete slabs of compos-
ite bridges. IABSE Report, No. 999, 1997; 187–192.
[3] Dischinger F. Stalbrücken in Verbund mit Stahlbetondruckplatten bei gleichzeitiger Vors-
pannung durch hochwertige Seile. Der Bauwingenieur 1949; 24(11): 321–376.
[4] Homberg H, & Köhling E. Die neue Ruhrbrücke Herdecke, eine durchlaufende Verbun-
dträgerbrücke mit Vorspannung. Der Bauingenieur 1951; 26(5): 129–176.
[5] Ridet J. La construction mixte acier-béton armé dans les ouvrages d’art. Mémoires de
l’AIPC 1947; 8: 171–194.
[6] Guérin J , Pigeau H. La construction mixte fer-béton dans les ouvrages d’art à travées con-
tinues. Annales de l’Institut Technique du Bâtiment et des Travaux Publics, No. 157, No-
vember 1950, 1950.
[7] Ridet J. La construction mixte fer-béton dans les ouvrages d’art. Annales de l’Institut Tech-
nique du Bâtiment et des Travaux Publics. Circulaire Série 1, No. 48, May 1947, 1945.

[8] Birguer A. Constructions composites et précontraintes en acier-béton. Annales de l’Institut

Technique du Bâtiment et des Travaux Publics, No. 221, May 1966, 1966.
[9] Laviolette M. Pont n°34 sur l’autoroute de Wallonie. Acier-Stahl-Steel, No. 6, Juin 1968,
1968; 290–295.
[10] Fougnies R, Mahieu L. Evolution récente des ponts en béton précontraint en Belgique:
le Pont de Herstal. Annales des Travaux Publics de Belgique No 3, June 1969, 1969;
[11] Fougnies R, Mahieu L. Evolution récente des ponts en béton précontraint en Belgique: le
Pont de La Louvière. Annales des Travaux Publics de Belgique No 3, June 1969, 1969;
[12] Ducret J-M. Étude du comportement réel des ponts mixtes et modélisation pour le dimen-
sionnement. PhD Thesis Nr 1738. École Polytechnique Fédérale de Lausanne, 1997.
[13] Rammer Y, & Espion B. A case study of replacement of external post-tensioning cables.
Bridge Eng. 2015; 167(BE3): 183–192.


Making Rennie’s 1796 Lune
Aqueduct Watertight Againi

Leslie Clarke, CEng, FICE. Principal Engineer, Canal & River Trust, Leeds, UK

Introduction and History

The Lune aqueduct carries the Lancaster Canal across the River Lune just to the north of the city
of Lancaster. It is a magnificent, monumental, five-span, masonry structure constructed between
1793 and autumn 1796 under the auspices of John Rennie. The aqueduct is now a grade I listed
structure and forms part of the estate of the Canal & River Trust (Fig. 1). Leakage from the
aqueduct canal trough has been a continuing problem since construction, as can be seen from
the buildup of calcite deposits and the staining on the structure. Calcite deposition by means of
precipitation is a slow process that takes many years. The process requires slightly acidic water
as an initiator and carrier—just the type of conditions brought about by the Industrial Revolu-
tion of the 1700s and industrial production of the 1800s. On the north abutment of the Lune
aqueduct, the calcite has formed a surface glaze or flowstone over the masonry. Over that time,
quantities of lime have gradually been leached out of the mortar, not only of the canal trough
structure but also the aqueduct structure itself. Leakage has occurred from beneath the arches on
the upstream and downstream sides, from the pier faces, from the spandrel walls, from the wing
walls and from below the ledges at towpath level.

Aqueduct Structure
The masonry of the aqueduct is sandstone of the millstone grit series that was sourced from
several quarries local to the site. The stone has some variation in grain size and coloration con-
sistent with different sources. The masonry piers and abutments are founded on piled timber
rafts. Although the structure above the piers looks very solid and bulky, it is in fact hollow. The
arch supports a series of vaults springing from two longitudinal walls built from the top of the
arch (extrados) within the structure. The canal is contained within a masonry trough 2.3 m deep

This essay is a reduced version of a paper published in Engineering History and Heritage. Please see the original for
full details in Ref. [1].

Fig. 1: The Lune aqueduct, 1797. Engineer: John Rennie. The aqueduct carries the Lancaster
Canal across the River Lune, near Lancaster, UK

and 6.2 m wide. The trough is a flattened, widened U-shape formed from two ellipses, and the
greater axis is the width between the outside retaining walls and the lesser depth of the canal.
The trough comprises two layers of masonry, an outer rubble course (229 mm thick) and an
inner-facing course of ashlar blocks (305 mm thick), all underlain by a 914 mm minimum thick-
ness of puddle clay, which is contained within the masonry spandrel walls on either side, each
spandrel wall being 1.524 m thick. The gap between the spandrel walls and the back of the canal
through walls is filled with puddle clay. The base of the puddle clay is supported on the top of a
series of internal stone vaults within the aqueduct structure. These vaults are formed from two
914 mm thick internal walls, springing off the extrados of the arch, and these support three inter-
nal Gothic arches. The inside of the aqueduct is thus partly hollow, thereby reducing the loads on
the piers and abutments and thus the foundations. Each pier contains a series of secured longitu-
dinal wrought-iron chain bars between the upstream and downstream ends, one at the level of the
stone course below low water (of the river) and the other under the level of the arch springing.
In addition, each arch contains four secured wrought-iron bars running from the upstream side
to the downstream side. The ironwork was intended to assist in holding the structure together.

Problems Caused by Leakage

Quantities of water leaking through an aqueduct can have considerable deleterious effects on
the fabric of the structure over time. Lime can be leached from the joints between the masonry
blocks, causing gradual small movements in position. Freeze–thaw cycles can cause delamina-
tion of layers of stone and cracking within individual blocks, as well as erosion of lime mortar
joints. Freezing water within the masonry fabric can cause major movement of elements of
the structure. An increase in the moisture content of contained fill can lead to major structural
movement at stress concentration points. All of these effects are visible on the Lune aqueduct
in various places. Over time, these effects can cause the structure to deteriorate gradually, both
in appearance and in structural performance. In the northeast arch at river water level, drips of
water from the canal are visible. On the next two piers and buttresses, damp patches can be
observed. Damp is visible on the spandrel wall. Calcite deposits are clearly visible generally.
Vegetation can be seen growing on the structure, a clear sign that water is present. Figure 2
NOVEMBER 2008 151

shows the southeast abutment with the start of the arch spring-
ing, showing leakage, calcite deposits from leached lime binder,
cracked masonry blocks and delaminated blocks due to freeze–
thaw cycles on wet stone.

On 15 September 1798, there was an embankment breach failure

at the southeast approach to the aqueduct.2 The embankment was
reconstructed and sealed. On 31 October 1798, the canal trough
rent (fissured) at the end of the south abutment, with a similar
situation occurring at the north end.2 The back of the abutment
was excavated and repuddled with clay. On 23 November 1798,
Rennie recommended that the copings on top of the canal trough
Fig. 2: Damage to the south- should be taken up, and the puddles trenched over on both sides
east arch springing of the aqueduct.3 This would only have been conducted as a con-
sequence of leakage. Between 1800 and 2012, there was general
leakage of varying quantities from the canal trough through the aqueduct, with water wetting
parts of the face of the aqueduct, dripping into the river from the arches and onto the river
towpath from the arch on the southeast. After 1900, localized repointing with cementitious mor-
tar was carried out (precise dates unknown). In April 2006, there was major leakage from the
embankment, arch and east wing wall at the south end of the aqueduct. In November 2008, there
was further major leakage at the south end of the aqueduct in the same area as in April 2006.

Recent Leak Sealing Works

Maintaining canal structures that are over 200 years old in a robust, functioning and reasonably
watertight condition, such as the Lune aqueduct, is a difficult task when funds are limited and
with a network that extends across England and Wales. A risk-based approach is adopted with
regard to this asset management issue, with priorities being allocated to individual problemati-
cal assets. With respect to the Lune aqueduct, there has been a significant number of recent,
major leakage problems dating from 2006, described below.

April 2006
Severe leakage appeared from the embankment in the area of the southeast wing wall. A hole of
100 mm diameter was discovered in the center of the canal bed very close to the embankment
and canal trough junction. There was an additional hole under the adjacent towpath approach
wall on the southwest. As a quick, cost-effective, temporary measure, the holes were filled with
general-purpose grout under gravity; the existing clay bed of the canal was removed over a 10 m
length from the end of the aqueduct trough, and new puddle clay was placed over the approach
to the aqueduct. The clay was taken partially up the masonry approach walls and over the start
of the aqueduct trough.

November 2008
Leakage from the south abutment was first noted in November and progressively worsened over
a number of days. In early December, the leak became significant, and canal water also began to

flow from the base of the southeast embankment

close to the interface with the masonry wing wall
of the aqueduct, in the same area as the 2006 prob-
lem. This became a cause of increasing concern,
and the canal was dewatered. Once dewatered,
the canal was inspected. No significant holes or
fissures were noted in the clay of the canal bed.
On further investigation, voiding was discovered
below and behind the towpath masonry approach
wall at the aqueduct transition. Leakage water
was exiting into the void, flowing under the canal
Fig. 3: Southeast end leak sealing, Novem- bed to the opposite side through the embankment
ber 2008 to the southeast and emerging near the base of the
wing wall. A more permanent repair was made,
consisting of a short transition length of water-retaining channel from the end of the aqueduct
trough onto the embankment (Fig. 3).

January 2011–2012
For leak sealing of the aqueduct trough, a series of grouting works was planned. This choice
of solution was arrived at partly due to the classification of the aqueduct as a grade I listed
structure, a structure of great heritage value and importance, and additionally following lengthy
discussions and consultations with the Canal & River Trust heritage advisor, Lancaster City
Council heritage officer and English Heritage. The result of these consultations was that the use
of an applied, internal, watertight sealing system on the internal face of the trough masonry was
precluded, even though this would have produced a canal trough that would have been virtually
watertight. Reconstruction of the trough was also considered an option, but this would have
been too expensive and time-consuming.

Grouting Works, First Stage: February/March 2011

The first stage of grouting works, the purpose of which was to seal the joints in the canal trough
masonry, was undertaken with the canal out of water. Following the stage 1 grouting, the overall
amount of seepage from below the aqueduct had been reduced by approximately 90–95% and
had been confined to certain areas. However, it was felt that the top two to three courses of wall
masonry had not been dealt with sufficiently, mainly due to the method adopted. Therefore, a
second stage of grouting works was proposed.

Grouting Works, Second Stage: June 2011

The purpose of this grouting was to fill the upper two to three courses of masonry in the off-
side wall, where there were still thought to be leakage problems that had not been completely
addressed by the internal trough grouting works, and as had previously been deduced by moni-
toring variations in canal water operating levels and leakage rates. The same type of grout as
used in the first stage was injected from the top of the offside wall along the whole wall length

through the rear face of the coping stones. Following the grouting, a further slight reduction in
leakage was observed, to around 95% of the original.

Grouting Works, Third Stage: June 2012

In the course of the major improvement works previously completed in April 2012, damaged
coping stones along the offside edge of the canal trough had been removed for replacement.
During this process, it was possible to inspect the top of the canal trough walls. The inspection
showed that the joint between the two masonry layers was in poor condition, with an almost
complete loss of lime binder, leaving the mortar effectively as a loose saturated sand containing
silt inclusions from canal water. Following the stage 3 grouting, the residual seepage, compared
with that after stage 2, was reduced by approximately 40%.

Observations: Grouting and Leak Sealing

From the works carried out and the observations made during the aqueduct restoration works, it
is apparent that the joint between the ashlar and rubble masonry courses of the canal trough is a
potential seepage path for water. This can occur through the smallest of shrinkage or separation
cracks between mortar/grout and the ashlar stone blocks, crack widths that are not visible to
the naked eye. As an alternative, the application of a waterproof coat on the inside of the canal
channel trough, rather than further grouting, would give more certainty of outcome in terms
of eliminating the residual seepage. However, this rests on the assumption that listed building
consent could be obtained for this approach, which, judging by the discussions with heritage
advisers from the Canal & River Trust, Lancaster City Council and English Heritage, would not
be supported or granted for a structure of this importance.

There had been leakage between the stiff structure of the aqueduct and the soft embankment
approaches since the canal opened in 1797. It is hoped that incorporating water-retaining rein-
forced-concrete channels on either side of the original masonry trough of the aqueduct will
permanently eliminate this leakage. The grouting operations to the canal trough along the length
of the aqueduct have been very successful. Leakage has been reduced by 95–98% of the levels
observed before restoration began. Although the canal trough does still exhibit residual seepage,
the quantity is very small and will not lead to further deterioration of the masonry structure.

Lessons for Consideration

The matters discussed in this paper relate to a well-known, monumental, historic structure.
However, lessons can still be learned that may apply to aspects of other projects, not only to
repair or renovation work but also to new work. These lessons are as follows:

• understand the history and nature of the structure

• understand the context of the site
• identify the effects of problems

• investigate and understand the causes of the problems

• when the causes are known, only then define the solutions
• solutions should be sympathetic and not have unintended consequences
• be flexible in approach, do not stick rigidly to predetermined solutions in light of new or
additional evidence
• observe, monitor and review the results and use this information to confirm or revise the
remaining works
• disseminate the information for others to benefit from.

If historic structures and assets are to be valued, treasured and retained for the nation, as the
Canal & River Trust believes the canal system should be, then they require regular, thorough and
systematized inspection and risk management processes so that problems can be identified at an
early stage, and measures can be put in place and undertaken to remedy them at an appropriate
time. Understanding and expertise are required in dealing with these problems so that appropri-
ate solutions are put into effect. This expertise needs to be passed on to future generations in
order to sustain this approach.

[1] Clarke L. Making Rennie’s 1796 Lune aqueduct watertight again. Proc. Inst. Civil Eng.
Eng. History Heritage 2014; 166(EH4): 198–206.
[2] Lancaster Canal Navigation Company. Minutes of the Committee Meetings (1798–1813).
RAIL 844/231, The National Archives, (1798).
[3] Rennie J. Specification for the Mason Work for the Lune Aqueduct. Lancaster Maritime
Museum, LANLM. 1985.76.1, 1793.


Restoration of Robert Maillart’s
Reinforced Concrete Bridges in

Eugen Brühwiler, Professor, EPFL – Swiss Federal Institute of Technology, Lausanne,


Bridges of high cultural value and aesthetic quality deserve respectful treatment and, conse-
quently, construction interventions must balance these assets with the severe requirements of
utilization. This is particularly relevant to structural engineers and bridge owners involved in
rehabilitation or modification interventions. This contribution presents, by means of restora-
tion of bridges designed by the famous Robert Maillart, how interventions can be performed
with adequate respect for cultural value while complying with the demands of modern traffic
use. It demonstrates how noninvasive and thus low-cost interventions can be performed on
“old” concrete structures with adequate respect for cultural and aesthetic values in order to
improve them in view of a second service duration. This goal of noninvasive interventions
can only be achieved by using advanced engineering methods (as allowed in standards valid
for existing structures) and intervention technologies specifically suited for existing concrete

Robert Maillart
Robert Maillart (1872–1940) was a Swiss civil engineer who strongly influ-
enced design and construction using reinforced concrete. Robert Maillart
was born in Berne, Switzerland, and studied at the Swiss Federal Institute of
Technology (ETH) in Zurich. His main mentor was Professor Wilhelm Rit-
ter from whom he learned to develop structural shapes that could be easily
handled by structural analysis. Maillart could rarely test his original designs
prior to construction, and thus former EMPA director Mirko Ros who
performed load testing on many of Maillart’s bridges, held an important
role in validating Maillart’s original design approach. Maillart’s main inventions are the three-
hinged arch and the deck-stiffened arch designs for bridges, as well as the beamless floor slab for

industrial buildings, supported by columns with mushroom-like heads for continuous force trans-
mission. His design approach and principles as well as his slender bridges changed the aesthetics
and engineering of bridge construction significantly and are influencing engineers until today.

Examination and Noninvasive Intervention

to Improve Durability
Based on the standards valid in Switzerland for existing structures,1 examination of the structural
performance of Maillart’s bridges presented here was conducted using updated values for action
effects (dead load, road traffic) and resistance of structural members. The results showed that the
requirements for structural safety are fulfilled in view of today’s and future traffic demands. The
overall condition of the bridges was generally “satisfactory to good” despite the fact that almost
no maintenance has been conducted. As was to be expected from more than 80-year-old exposed
concrete, carbonation depth was significantly beyond the first layer of steel rebars (rebar cover of
about 20 mm). Consequently, there was likelihood of steel rebar corrosion if concrete humidity
was sufficiently high (i.e., higher than 80%). In addition, few local zones showed visible rebar cor-
rosion, for example, along edges where it was difficult to correctly position the rebar and compact
the concrete. Given this condition of the exposed concrete, a noninvasive concept was proposed
and executed to improve the durability of reinforced concrete, consisting of two major aspects:

• Restoration of the local zones showing corroding rebars using a mortar composed of sand
and gravel of similar origin, as well as using timber boards for the formwork similar to the
ones used for the initial construction.
• In-depth hydrophobic treatment of rain-exposed concrete surfaces with the objective to pre-
vent ingress of rainwater into the concrete. This treatment allowed the concrete humidity to
be maintained at a rate lower than 75%. It is effective in time as the water-repellent agents
penetrate more than 5 mm into the concrete where they are protected from ultraviolet sun
radiation and leaching effects due to rainwater.

Thorough preliminary tests were conducted to identify the restoration mortar that fits best to the
existing surface texture and color and to validate the success of the in-depth hydrophobic treat-
ment for the given concrete. Products for in-depth hydrophobic treatment appeared more than
ten years back in the construction domain because it was possible to produce water-repellent
nano-size agents, allowing for easy penetration into concrete by capillary suction.

Schwandbach and Rossgraben Bridges

In 1933, Robert Maillart designed and built two bridges located close to each other in a remote
area near Schwarzenburg, south of Berne, in Switzerland. The Schwandbach bridge (Fig. 1) is the
most well-known deck-stiffened arch bridge of Maillart because it is like a spatial structure car-
rying a curved roadway. The Rossgraben bridge (Fig. 2), with its arch span of 82 m, is the second
largest open-box, three-hinge arch structure, and the world-famous Salginatobel bridge is the
most important three-hinged arch bridge. Both bridges have high cultural and aesthetic values.

Both reinforced concrete structures showed satisfactory performance in terms of structural

resistance and durability. Therefore, after more than 70  years of being in service, limited

Fig. 1: Schwandbach bridge, after restoration (2005)

Fig. 2: Rossgraben bridge, after restoration (2005)

intervention was necessary to restore their durability according to the noninvasive intervention
concept described above. Rehabilitation works performed in 2005 included improving the entire
water drainage system, placing a waterproofing membrane on the deck slab, locally repair-
ing zones where steel rebars were damaged by corrosion and protecting the exposed concrete
surfaces with in-depth hydrophobic treatment. This low-impact and cost-effective intervention
brought no visible change in the structures, although it significantly improved the service life
of these bridges. Ten years after the intervention, the bridges’ condition is good, and the surface
protection is still effective.

Schrähbach Bridge
The Schrähbach bridge is also located in a remote area where it is part of an only 3–4 m-wide
agricultural road around an artificial lake created by the construction of a dam. There is only
limited traffic, including single heavy agricultural vehicles. There are no plans to change this
traffic regime in the future. The bridge was designed and built in 1923 by Robert Maillart
(Fig. 3). It is the first realization of the structural system called “deck-stiffened arch” invented
by Maillart for the purpose of reducing structural dimensions and thus construction costs. This
structural system allows building a thin arch on a light scaffold, which is cheaper to build.



Fig. 3: Schrähbach bridge in 1925 after filling of the spandrels (a); original project by
Maillart (b)

The necessary stiffness of the structure is provided by the bridge girder with a U-shaped cross-
section appropriate to carry concentrated vehicle wheel loads (particularly those located at the
quarter span of the arch). The 59 m long and 3.6 m wide bridge structure carries a 3 m wide
single roadway. The bridge comprises three parts—two short approach viaducts and the deck-
stiffened arch structure. The arch has a span of 28.8 m and a rise of 4.0 m (rise-to-span ratio of
1 : 7.2). The deck girder (with a height of 1.3 m) and the arch (of a thickness varying from 0.18 to
0.22 m from rise to abutment) are connected monolithically through 150 mm thin spandrel walls
spaced by 2.43 m. The thickness of the deck slab is 0.17 m and carries a 200 mm thick concrete
overlay pavement added in 1935.

Some months after construction and load testing, the spandrels of the arch bridge and the first
opening of the approach viaducts were filled with bricks that have no structural function (Fig.
3a). As this intervention is not reported, one may suspect that these fillings were executed—most
probably without asking Maillart—for architectural reasons as there was an overall architectural
concept for all engineering works around the lake implying massive construction. In addition,
Maillart’s slender structures were not respected and acclaimed by most of his contemporary
engineers and architects.

Examination and Intervention Concept

In view of its renovation, an examination of the Schrähbach bridge was performed. The
results show that the structural safety for all structural members is verified for unrestricted
road traffic, and limitation to the utilization (such as a limitation of allowable axle loads or
gross vehicle weights) is not necessary. The reinforced concrete of the bridge can be recondi-
tioned and protected according to the concept described above and has been already applied to
the Schwandbach and Rossgraben bridges. The intervention concept also aims to restore the

Fig. 4: Schrähbach bridge: present condition

structure according to Robert Maillart’s original plans (Fig. 3b), namely, to remove the ugly
spandrel filling after 90 years! However, as of today, the Schrähbach bridge has still not been
rehabilitated (Fig. 4), it is actually threatened by demolition and replacement, as decided by the
owner and his consulting engineer. Subsequently, restoration of this bridge unfortunately is in
court and the case is still under way, and even technical and economic arguments do not seem to
make the “law”. This case shows once again the extent to which professionals in charge of civil
works need sensitization and education about the cultural values of their works as well as mod-
ern engineering methods and technologies to preserve cherished structural engineering heritage
structures worldwide, like the ones by Robert Maillart.

The examples of three bridges designed by Robert Maillart show that rehabilitation interven-
tions on bridges of high cultural value can meet the severe requirements of utilization while
respecting historical value and aesthetic quality. The objective of the interventions is to apply
so-called noninvasive or “soft” methods to restore and protect reinforced concrete exposed to
environmental action. There are many “undiscovered” and “ignored” bridges built over the last
60 years that deserve similar recognition as the ones by Maillart. Structural engineers and bridge
owners need to be more aware of these aspects when conducting rehabilitation or modification
interventions. Structural engineering with the ultimate goal of limiting construction interven-
tions to a strict minimum is thus intertwined with the interest of preserving monuments and
limiting costs to bridge owners. It also means giving value to bridges as well as appreciating the
art of structural engineering and the identity of structural engineers.

[1] Brühwiler E, Vogel T, Lang T, & Lüchinger P. Swiss standards for existing structures. Struct.
Eng. Int. 2012; 22(2): 275–280.
[2] Brühwiler E. Les ponts en béton armé de Robert Maillart – intervenir pour pérenniser. In La
Sauvgarde des Grandes Œuvres de l’Ingénierie du XXe Siècle Cahiers du TSAM, PPUR –
Presses Polytechniques et Universitaires Romandes: Lausanne Suisse, ISBN 978-2-88915-
113-4; 2015, 126–141.


Examination of Two Riveted Railway
Bridges over the River Rhine

Eugen Brühwiler, Professor, EPFL – Swiss Federal Institute of Technology Lausanne,


Riveted steel bridges were built over a period of more than 100 years up to the 1950s. Several
of them are considered historical and should be preserved as heritage of structural engineer-
ing. Often, an important future service life may be identified such that, economically, it is not
justified to replace a riveted bridge because of some arbitrary age criterion, as has been often
done in the past. This contribution deals with the examination of two railway bridges in riveted
wrought-iron and early steel construction. Both were built in the 19th century and span over
the River Rhine. They are still in service accommodating the modern-day railway traffic of the
Zurich suburban railway. Both bridges have been examined recently to verify the structural and
fatigue safety for long-term future utilization while preserving the cultural values of the bridges.
The standards valid in Switzerland for existing structures were applied:1 updated railway traffic
load models, updated data regarding the fatigue behavior of riveted joints and information from
in situ deformation measurements were used to verify the structural and fatigue safety based on
precise bridge-specific information.

Railway Bridge over the River Rhine between Koblenz

(Switzerland) and Waldshut (Germany)
Bridge Structure
The investigated bridge (Fig. 1) crosses the Rhine river in northern Switzerland to carry a one-
lane railway line between Koblenz (Switzerland) and Waldshut (Germany). It was designed by
Robert Gerwig and built in 1858–1859 by the German company Benckiser. It comprises riveted

wrought-iron members. The straight

lattice truss structure is one of the last
examples of a construction type that was
typical for the railroad construction boom
in Europe during the third quarter of the
19th century.

The wrought-iron structure was designed

as a continuous girder over three spans of
37.5, 55 and 37.5 m, with a total length of
130 m supported by abutments and piers
of natural stone masonry. The bridge girder
carries a single track by an open deck car-
riageway, that is, the timber sleepers are
Fig. 1: Railway bridge over the River Rhine
directly fixed to the stringers. The bridge
between Koblenz (Switzerland) and Waldshut
was initially designed to carry two tracks;
however, it always only carried one track.

Cultural Values
The Rhine river bridge Koblenz–Waldshut is considered the oldest riveted railway bridge still in
service in Continental Europe. This bridge is generally accepted as an engineering monument of
high value. In 1994, it was given a Brunel Award, which is the most important award for railway
architecture. The cultural values of this bridge may be evaluated as given below.

Historic Value
The multiple lattice girder structure is one of the last representatives of a type of construction
that was typical of the pioneer’s time of railway construction in the third quarter of the 19th
century. The structure, which is still in its original condition, is an important reference in the
work of Robert Gerwig (1820–1885), who was also the builder of the north ramp of the Got-
thard railway and other important railways in Central Europe. Robert Gerwig is considered one
of the most important engineers of railway construction in the 19th century in Switzerland and
southern Germany.

The bridge is an aesthetically pleasing by virtue of the transparency of the lattice girder structure
and its interrelation with the massive river piers and abutments of natural stone masonry (Fig. 2).
The wrought-iron structure is characterized by the riveted construction. It has filigree and bold
appearance despite the strong appearance as a continuous beam of constant height. The massive
river piers and abutments contribute significantly to the impression of stability. The bridge’s
slenderness and the transparency of the lattice girder visualize technical efficiency. The symme-
try and repetition of identical structural elements contribute to uniformity and order. This func-
tional structure set in natural surroundings was planned as a simple structure and appropriately
designed. Any decorative elements were omitted. The green color of the corrosion protection
painting is ideal as it is one of the possible colors of iron.

Fig. 2: Details showing the interrelation between the wrought-iron lattice girder and natural
stone pier and abutment

Relation to the Surroundings

The bridge stands across the Rhine River with great visibility. The sober continuous girder bridge
with its severe lines stands in contrast to the natural river landscape. The piers and abutments
and the subsequent approaching masonry viaduct located on the Swiss side create a harmonious
transition of the riveted structure characterizing a technical object to the natural surroundings.

This bridge stands for the highest engineering efficiency in all matters: short time of construc-
tion, low material use, minimized cost of construction and a sober and flawless appearance.
Its cultural values are assessed as being very high. This bridge belongs undoubtedly to the
most important existing riveted bridges in Europe. It is a monument to the art of structural

Examination of Structural Performance

In 1999, the bridge was upgraded to accommodate the modern railway traffic of the Zurich
metropolitan area. To estimate the fatigue life of the bridge structure, a detailed examination
was performed using realistic past and future traffic models as well as advanced knowledge of
the fatigue behavior of riveted wrought-iron details. It could be proven that the bridge structure
was safe, and its service life could be considerably extended for the foreseen passenger train
traffic.2 More recently, the bridge was again examined in view of a long future service life of
80  years and increasing future passenger train traffic. The examination of the bridge perfor-
mance revealed that criteria of fatigue safety and structural safety are fulfilled, including a future
fatigue life theoretically longer than 80 years. No extraordinary interventions need to be per-
formed to keep the bridge in service. To guarantee the bridge’s durability over the next 80 years,
future rehabilitation works, including rehabilitation of the open railway track, were outlined.
Preservation and further utilization of the more than 150-year-old bridge are significantly more

economical in terms of both costs for maintenance interventions and life-cycle costing, com-
pared to the hypothetical option of bridge replacement.3

Railway Bridge over the River Rhine at Eglisau

Bridge Structure
The railway bridge across the River Rhine at Eglisau in Switzerland was built between 1895 and
1897 for single-lane railway traffic. The bridge was designed by Robert Moser and chief engi-
neer E. Züblin, distinguished railway engineers during the main period of railway construction
in Switzerland in the second half of the 19th century. The central part of the 457 m long bridge
is a riveted steel truss structure made of early mild steel, built by the steel construction company
Buss from Basel, Switzerland. The truss girder has a span of 90 m and a height of 9 m. Multiple
arch approach viaducts in natural stone masonry, with piers up to 50 m in height, follow in the
north and south of the steel truss (Fig. 3). During 1982–1983, the original carriageway (that
consisted of a ballasted track on Zores iron profiles) was replaced by a steel trough with ballast.
Several joints of the truss girder were also strengthened using posttensioned bolts, and the cor-
rosion protection painting of the steel construction was renewed entirely.

Cultural Values
This almost 500 m long bridge that is 50 m high above the river valley is very dominant in the
landscape. The high massive piers of the masonry arch approach viaducts and the powerful
riveted truss beam in between confer a certain boldness to the bridge. The overall appearance
is characterized by the hybrid structure, that is, lightweight steel beam of long span over the
river and massive masonry multiple arches of shorter spans for the approaching viaducts. This

Fig. 3: Railway bridge across the River Rhine at Eglisau, Switzerland. Scaffolding at mid-span
and at the abutment was installed during the monitoring campaign

type of hybrid bridge construction was very frequent for railway lines built in the 19th century
and before the use of reinforced concrete. Typical in the railway domain, this bridge also stands
for the highest engineering efficiency in all matters: short time of construction and low cost
of construction. This bridge is still in its original constitution. It is considered an engineering
monument of high value by the authorities for cultural heritage.

Examination of Structural Performance

Long-term monitoring of the structural elements was conducted over one year4 with the objec-
tive to gain precise information about the railway action effects on structural members. Moni-
tored deformation values were exploited by rain flow analysis and served as the basis for
fatigue safety verification. As the locations of measurements are generally not identical to
the cross-sections of verification, measured strains were translated to the relevant verification
cross-section by means of factors that were determined by detailed structural analysis. Using
these values, all fatigue-relevant structural details were first verified with respect to a fatigue
limit of riveted joints of 51 MPa. Then, damage accumulation calculation according to the
Palmgren–Miner Rule was performed for the few elements where the fatigue safety check,
with respect to the fatigue limit, was not fulfilled. Sufficient fatigue safety could finally be veri-
fied for the entire riveted structure, and additional service duration of more than 50 years for
this riveted structure could be validated. The examination showed that after almost 120 years
of service, there is no notable fatigue damage in the riveted structure. Taking into account even
higher future traffic loading, a second service life may thus be expected. Consequently, the
implemented approach to determine the updated action effects allowed for explicit considera-
tion of data from long-term monitoring. Monitored data were used to determine accurately
fatigue-relevant stresses in fatigue-prone structural elements. Hence, uncertainties in the deter-
mination of updated action effects were reduced. The performed examination based on long-
term monitoring was economic as the cost of the long-term monitoring and accompanying
theoretical studies was only a small fraction of the cost of hypothetical major strengthening or

The two riveted railway bridges are important monuments of structural engineering, and they
will remain in service for modern railway traffic despite their relatively long past service life.
Extending the service life essentially means giving value to bridges as well as appreciating the
art of structural engineering and the identity of structural engineers. Bridges dating to the 19th
century are commonly considered “old”. The present examples show that examination based on
modern structural engineering for existing structures may reveal that riveted bridges can stay in
service for a long future service life. There is no “old” bridge, and structural performance is the
sole relevant and adequate criterion to keep a bridge in service. In the past, riveted bridges were
systematically replaced. Even now they are considered by some railway authorities to be “old”
and prone to fatigue and thus as candidates for bridge replacement. In addition, cultural and
aesthetic values are often disregarded by structural engineers. Consequently, a change in para-
digm is still needed. Structural engineers have to understand that modern topical discipline and
competence is to examine existing structures using advanced methods while respecting cultural
values and financial constraints.

[1] Brühwiler E, Vogel T, Lang T, & Lüchinger P. Swiss standards for existing structures. Struct.
Eng. Int. 2012; 22(2): 275–280.
[2] Keller A, Brühwiler E, Hirt MA. Assessment of a 135-year-old riveted railway bridge. Pro-
ceedings, IABSE Symposium in San Francisco, IABSE Report, vol. 73/2, 1995; 1029–1034.
[3] Brühwiler E. Examination of fatigue safety of a 150-year old riveted railway bridge. Pro-
ceedings, IALCCE’12, Vienna, 3–6 October 2012, 2012.
[4] Brühwiler E, Bosshard M, Steck P, Meyer C, Tschumi M, Haldimann S. Fatigue safety
examination of a riveted railway bridge using data from long term monitoring. Proceed-
ings, IABSE Conference, Rotterdam 2013: Assessment, Upgrading and Refurbishment of
Infrastructures, 2013.


The Rendsburg High Bridge across
the Kiel Canal

Matthias Bartzsch, Struct. Eng., Karsten Geißler, Prof.; GMG Ingenieurgesellschaft

Dresden Germany

Overview of the Structure

The Rendsburg High Bridge is one of the most important technical monuments in Germany and
a landmark of the town Rendsburg. This older than 100-year-old bridge leads the railway line
Hamburg–Flensburg (–Denmark) across the Kiel Canal. This railway line is the main line to
Scandinavia, which lies under a very heavy railway traffic route. The viaduct was built between
1911 and 1913 using a design proposed by Friedrich Voß and facilitates a clear passage height for
shipping of 42 m. Due to the great importance of the bridge for national and international railway
traffic, the bridge is being extensively retrofitted by its owner, the Federal Water and Shipping
Administration (WSV), as well as by the German railway company Deutsche Bahn (DB) (Fig. 1).

The structure was built as a riveted steel construction with an entire length of nearly 2.5 km. It
consists of several parts: the main bridge and the northern and southern transition bridges. The
main bridge (“canal bridge”) is situated in the middle of the structure. It has a length of 294.6 m

Fig. 1: Rendsburg High Bridge with the “canal bridge” (the main bridge, left) and the “loop
bridge” (right)

and a free span of 140 m above the canal. The

main structure of the canal bridge was built
as a framework, while both longitudinal and
transverse girders were built as solid-web
girders. The canal bridge is linked by the
connecting piers to the northern and south-
ern transition bridges. The transition bridges
consist of a chain of piers and superstruc-
tures—51 piers and 105 superstructures, each
a short superstructure (11.50  m) on the pier
and a long superstructure (26.50–28.50  m)
between two piers. The superstructures were
constructed as single-span solid-web gird-
ers. The substructures were built as truss
piers. The northern transition part addition-
ally houses another bridge of 75 m length, the
so-called “loop bridge”. As the railway line
needs to reach the Rendsburg train station at
the north of the viaduct but still has to over-
come a height difference of about 30 m, the
line is routed in a wide loop. The line crosses
Fig. 2: Transporter Bridge itself at the so-called loop point. The loop
bridge, a trussed frame structure, is situated
there. The construction material of the via-
duct is steel from the beginning of the 20th century (“Flusseisen nach 1900” according to the
German standard). Its strength characteristics are similar to today’s steel S235 but with a larger
scatter. The remaining service life of the bridge is estimated to be more than 50 years. A special
feature is the additional transporter bridge at the canal bridge. It consists of a segment of road-
way (a gondola) that is carried by cables, which are fixed to an upper carriage. The upper car-
riage, which is driven electrically, runs on special rails between the pylons of the canal bridge.
Up to four cars, but primarily pedestrians and cyclists, can be transported by the transporter
bridge (Fig. 2).

Development of Traffic Loads

The bridge was originally designed for the so-called “Preußischer Lastenzug A” (LZ A—
Prussian load arrangement A). Additionally, a reserve of 20% referred to LZ A was regarded
with farsightedness for later load increases at the design of the canal bridge, the loop bridge
and the piers. The load arrangement LZ A consists of two locomotives (axle load 17 t and
distributed load 71  kN/m), two tenders (13  t and 65  kN/m) and single-sided wagons (13  t
and 43 kN/m). In the future, the bridge will be used with the following load arrangements:1
a double-tracked traffic with a freight train of load class D2 (axle load 22.5 t and distributed
load 64 kN/m) on one track and a lightweight passenger train (DRZ) on the other track or2
a single-tracked traffic with a freight train of load class D4 (22.5 t and 80 kN/m). In terms
of distributed loads, the vertical forces of the former load arrangement LZ A and those of
today’s loads D2/DRZ (double-tracked) and D4 (single-tracked) are almost the same. Only

the axle loads are higher today. However, a considerable increase concerns the horizontal
(longitudinal) forces caused by braking and acceleration. While the braking forces were
determined by 1/7 of the vertical loads in the past, today they are determined by 1/4 of the
vertical loads—that is, nearly doubling of the braking forces. In addition, an acceleration
force of 1000 kN (transferred on a length of 30 m) has to be considered on the second track.
These high horizontal forces lead to a great load increase for the high and slender bridge. This
applies especially to the acceleration force, whose load length of 30 m nearly matches the
structural length of a truss pier with its associated (longitudinally fixed) long superstructure.

Horizontal Load Transfer

The viaduct is divided into single structures in longitudinal direction. The horizontal (longitudi-
nal) forces may be transmitted between these single structures only by the rails and the friction
bearings of the superstructures. The friction bearings are plated steel on steel with a friction
coefficient of μ = 0.2–0.4, which may transmit horizontal loads under the effect of vertical loads.
That horizontal load transfer by the rails and the friction bearings along the bridge enables a load
distribution of concentrated horizontal loads, and these are primarily the acceleration forces. A
direct application of the high acceleration forces and braking forces to the analysis model would
cause a numerical overloading of the single structures, mainly the truss piers. Thus, a realistic
load transfer/load distribution needs to be considered. The longitudinal load transfer was deter-
mined by a nonlinear analysis model of the whole bridge, including the rails. The analysis model
takes into account load-dependent, nonlinear force-displacement functions between the rails
and the railroad sleepers and the girders, as well as at the friction bearings. The nondeterminis-
tic parameters of the force-displacement functions were varied by numerous calculations. The
results of the analysis are the realistic horizontal forces undertaken by the single structures. For
example, the local induced acceleration force of 1000 kN may generally be reduced to 400 kN
per single structure. The remaining load is distributed to the nearby single structures. The com-
putation was verified by in situ measurements.

Strengthening Concept
The structural analysis yielded some typical cases of overstress at the bridge under today’s traf-
fic loads:

• members that are directly affected by the higher acceleration and braking forces, for exam-
ple, the main diagonal bars of the truss piers and the foundations of the truss piers (over-
turning of the piers),
• members that are affected by the higher axle loads of the trains, for example, the longitudi-
nal and transverse girders of the deck,
• members that participate unintentionally in the primary load-carrying system, for example,
wind bracing and deck elements.

The overstressed members need to be strengthened. Typical strengthening measures at the riv-
eted construction are (Fig. 3):

• strengthening plates on the truss members—the strengthening plates need to be integrated

into the truss joints,

Fig. 3: Strengthening by additional plates (a, b); replacement of wind bracing (c)

• latticing of open cross sections in or-

der to increase the torsional stiffness
(lateral torsional buckling),
• replacement of rivets against fitted
bolts if the shear capacity is exceeded,
• replacement of whole truss mem-
bers—generally only possible for sec-
ondary members , for example, the
wind bracing.

The (well-tried) static system of the bridge

construction should not be changed.
Fig. 4: Working under traffic load The execution of the strengthening meas-
ures at the riveted framework construction
with manifold graded cross sections is technically very challenging. Often, additional connect-
ing and filler plates are necessary. Welding of the old material is not intended. Most of the
strengthening measures are carried out under railway traffic. Until 2014, the bridge had been
used for single-track traffic; since the beginning of 2015, the bridge has been used for double-
track traffic (Fig. 4).

With extensive repair and strengthening measures, the over 100-year-old Rendsburg High
Bridge is being retrofitted for many years to come. Today’s railway traffic loads, especially
the acceleration and braking forces, are significantly higher than the original design loads.
Sophisticated computations and dynamic measurements make it possible to identify and to
verify remaining load capacity. Only using that can the bridge be retrofitted in a technically
and economically feasible way. The measures are usually carried out under railway traffic,
which makes high demands of their design, planning and execution. The Rendsburg High
Bridge is a masterpiece of civil engineering. Preserving it is an important contribution to
building culture.

[1] Geißler K , Bartzsch M. The Rendsburg High Bridge across the Kiel Canal. Proceedings of
the IABSE Conference – Structural Engineering: Providing Solutions to Global Challenges,
Geneva, Switzerland, September 2015, 2015; 466–473.
[2] Bartzsch M, Geißler K, & Schmachtenberg R. Die Ertüchtigung der Rendsburger Eisenbah-
nhochbrücke über den Nord-Ostsee-Kanal. Stahlbau 2015; 84(3): 171–181.


Steel Viaduct Refurbishment Inspired by the
Original Structure and Its History—The Best
Solution with Regard to Structural,
Economical and Heritage Requirements

Werner Lorenz, Professor, Chair of Construction History and Structural Preservation,

BTU Cottbus-Senftenberg, Cottbus, Germany

The preservation and strengthening of bridges protected by conservation orders is a structural
engineering field that is likely to cause conflicts. The cultural significance of these edifices
imposes an obligation to fulfill heritage requirements. However, load-bearing capacity, fatigue
resistance and cost effectiveness still need to be ensured. The extensive refurbishment of the
elevated viaduct for Berlin’s metro line No. 2, built from 1909 onward, is an impressive example
of the efficiency of the principle of “intervention close to the original”. This approach philoso-
phy is inspired by the original structure and develops the structural design of an intervention
from a comprehensive understanding of the history of the structure and aims for a solution that
systematically picks up on the strengths of the original fabric.

History and Significance

The 1.7 km long viaduct carrying today’s metro line No. 2 in the Prenzlauer Berg district of
Berlin was built in two phases, the first beginning in 1909. Trains started running on this section
in 1913 (Fig. 1), and an extension to Pankow, the end of the line at that time, followed in 1927–
1930. The concept, design and layout of the viaduct and stations were the result of an intensive
collaboration between engineer and architect, which, even then, was hailed as exemplary in con-
temporary publications. For example, “a form in which the technical has become aesthetic …,
a simplicity that has bearing ...” was how Karl Scheffler described it in 1914,2 and in 1922, Paul
Wittig wrote “engineered structures with a modern expression ... in which it is hardly possible to
tell where the activities of the engineer end and those of the architect begin”.1 The chief engineer
was Johannes Bousset (1865–1945), and the architect was Alfred Grenander (1863–1931) from
Sweden, who was responsible for the architecture of Berlin’s railways above and below ground
for 30 years and who knew how to give them dignity and significance. Protected by a preserva-
tion order since 1978, today, the viaduct is an artery in and an emblem of Prenzlauer Berg.

Structure, Details and

Distinct structural and constructional differ-
ences between the two construction phases
can be seen in the primary structure. The rea-
son for these differences can be found in the
dynamic development of structural steelwork
after World War I. In 1909, the primary sys-
tem was still made up of an alternating series
of two-pin frames with cantilevers supporting
Fig. 1: View of the viaduct, first phase, 1909– intermediate beams. In 1928, the designers
1913, photograph pre-1920 (see Ref. [1]) selected fixed-base hollow-box columns car-
rying beams in the form of balanced “rocker
arms” on pinned supports, with longer spans
of 28.50 m. On the other hand, in both phases,
the permanent way, consisting of track, sleep-
ers and ballast, was laid on a deck of curved
plates, the narrow ends of which changed
to bent plates in double curvature (Fig. 2).
As they were stressed like membranes, thin
plates were possible; contrary to recommen-
dations at the time, 7 mm was chosen instead
of 8 mm. Owing to the different plan shapes
(rectangular and trapezoidal) and the various
dimensions, the approximately 2200 plates
Fig. 2: Removal of historic deck plates (cour- that had to be fabricated were divided into
tesy: Lorenz & Co. Structural Engineers, more than 400 different types. The 547(!)
2009) bridge bearings, with 12 different basic types,
are worth a special mention. The reasons for
the high number and many different types included the short spans and the differing boundary
conditions of the load-bearing structure. Most of the original bearings are still in place. Types of
steel common at the time of construction were used (1909–1913: mild steel; 1927–1930: grade
St 37 steel). The properties of mild steel were nearly the same as the grade St 37 used later.
Rivets were used for most connections.

Diagnosis and Safety Evaluation

The development of the refurbishment concept followed detailed inspections, carried out from
1999 onward, to ascertain the condition of the structure. Calculations and loading tests, some
performed by the Federal Institute for Materials Research & Testing in Berlin, enabled the
compilation of a highly differentiated record of the state of damage and fatigue. Overall, despite
intensive use over a period of up to 100 years plus the higher axle loads of modern trains, the
results showed no reason to consider replacing the viaduct; the robust historic form of con-
struction had stood the test of time. Therefore, construction technology and heritage conser-
vation saw themselves as partners with essentially similar interests: preservation instead of

rebuilding! Nevertheless, various steel members needed considerable repairs, a number of load-
bearing components had to be replaced, and above all, the corrosion protection needed complete
renewal. The most alarming thing, however, was the condition of the deck. Corrosion had led to
some plates having rusted through almost completely, and numerous cracks near the supports
definitely pointed to a fatigue problem (see below).

Refurbishment Concept
The refurbishment concept had to take into account the fact that the viaduct is important in
different ways. The task was to ensure that the viaduct remained a safe part of the transport
infrastructure. At the same time, however, the prominent cultural significance of this historical
legacy had to be taken seriously in every respect. The engineers in particular had a great respon-
sibility: planning the engineering works was no longer just a technical assignment but a heritage
conservation one as well! With conflicting priorities, it was vital to agree on the fundamental
heritage preservation goals and the elements constituting the heritage assets themselves as early
as possible: What is it exactly that makes it “historic”? What are the particularly valuable “hot-
spots”? What has to be tackled especially carefully? Besides the structure, form, silhouette and
color of the viaduct, the list included its aesthetic significance, the multitude of different bridge
bearings and the characteristic form of the deck—precisely those elements that needed atten-
tion! As a result, the heritage conservation planning parameters were summarized in a few prin-
ciples: As little as possible, as much as necessary! Minimally invasive interventions! Repairs by
way of “homoeopathically” improved replicas while preserving the appearance! Otherwise, the
plans proceeded along the lines of the three-phase method of the ICOMOS Charter “Principles
for the Analysis, Conservation and Structural Restoration of Architectural Heritage”3,4—dating
from 2003—and the strict approach derived from that: diagnosis–safety evaluation–design of

An Example—Renewing the Deck

A more accurate survey of the cracks in the deck plates revealed a number of obvious defects
already inherent at the time of building. For example, the transition from the curve of the bal-
last plate to its horizontal bearing strip often did not lie directly on the supporting flanges of
the longitudinal or transverse beams as intended but protruded instead by up to 35 mm owing
to excessive construction tolerances. In reality, instead of acting as a membrane, the thin plate
was being loaded as a cantilever; the additional bending stresses that occurred every time a train
passed made fatigue cracks here inevitable. Above all else, it was this small detail defect that, in
the end, forced the replacement of the entire historic deck! As a replacement, the client initially
favored abandoning the ballast track and swapping to a slab track; in doing so, the old plates that
would be released from their structural function could, however, remain in place. Two options
were discussed: the “concrete on steel” solution, in which slabs with twin-block sleepers cast on
trapezoidal profile sheeting would be supported on the existing transverse beams of the viaduct
(Fig. 3), and an alternative “steel on steel” system, with longitudinal steel beams guaranteeing
that the loads would be transferred from the rails in their highly elastic mountings to the existing
transverse beams. However, there were considerable difficulties with both solutions. Those dif-
ficulties were due to the departure from the flexible permanent way with its ballast bed and, in
particular, incompatibilities between the different movement joint requirements of the rails, the

Fig. 4: Option 3: Replacement of deck plates,

Fig. 3: Option 1: Slab track “concrete on changed in detail “homoeopathically” (cour-
steel” (courtesy: Lorenz & Co. Structural tesy: Lorenz & Co. Structural Engineers,
Engineers, 2011) 2011)

slab track and the historic steel load-bearing structure plus the associated load paths for loads
due to braking and accelerating. Therefore, this author developed a third option that essentially
replicated the historic deck plates but avoided the flaws diagnosed in the damage survey through
precise optimization of the details of the historic model (Fig. 4). This required the development
of a solution that could handle more than 400 different plate geometries with one tolerance
standard that would prevent the unacceptable cantilevering of the original plates and the ensuing
fatigue problems.

The key to success was, on the one hand, the compilation of a detailed plate database in which
the geometry of every one of the roughly 2200 existing plates was recorded. This database ena-
bled sensible clusters to be formed for the replicas by specifying a maximum cantilever of just
10 mm. By increasing the plate thickness as well, from 7 to 9 mm, it was thus possible to verify
the durability of the relevant stress cycles with a utilization of merely 88% even under the most
unfavorable conditions. On the other hand, a method that permitted economic fabrication of the
number of different plate clusters required to meet this specification had to be found. The answer
came from the steelwork contractor RW Sollinger Hütte GmbH, who proposed fabricating the
double-curvature plates by using hydraulic cold forming to bend an initially flat plate. In the
meantime, this method could be easily adapted to the different geometric boundary conditions
and could thus comply with the necessary erection tolerances.

Heritage and Construction History and the Structural

Engineer’s Responsibility
Complete refurbishment of the viaduct for the U2 line, including the two high-level stations,
was completed on schedule in the spring of 2011 after more than ten years of preparations,
planning and building work (Figs. 5 and 6)—and without exceeding the budget of approx. €78
million. The project emphasized the fact that construction technology and heritage conservation
do not necessarily have to contradict each other, even with a transport infrastructure project such
as this one. Instead, constructive dialogue between the partners can lead to new solutions that are
interesting both technically and economically. It also confirms the thesis that the closer refur-
bishment concepts adhere to the specification of the historic structure itself, the better they are.

Fig. 5: Refurbished viaduct with renewed deck Fig. 6: Refurbished bearing supports (cour-
plates (courtesy: Lorenz & Co. Structural tesy: Lorenz & Co. Structural Engineers,
Engineers, 2011) 2011)

This project also impressively demonstrates that when developing an intelligent and truly sus-
tainable refurbishment concept, the design engineers need both engineering science and con-
struction history skills during all phases of the design process in order to:

• recognize the construction history significance of the historic structure and use this as a
basis for respectful treatment,
• locate the specific design forms in their contemporary context and use these to understand
load-bearing structure, details and load paths,
• identify the typical weaknesses and flaws in the forms of construction of the period and thus
devise upgrading strategies based on the historic structure and
• decide on the “hotspots” that require especially careful heritage conservation treatment in
such an engineered structure, determined primarily by its historic design.

Historic bridge structures such as the viaduct in Prenzlauer Berg are not simply complex evi-
dence of feats of human achievement. They are also, and principally, authentic testimony to the
history of the art of engineering, the history of the engineers themselves. Dealing with these
structures respectfully requires engineers to take on much more responsibility than the heritage
conservationists. Without knowledge and awareness of construction history, however, it is not
possible to assume that responsibility.

[1] Wittig P. Die Architektur der Hoch- und Untergrundbahn in Berlin. Zirkel Architektur-
Verlag: Berlin, 1922.
[2] Scheffler K. Gute und schlechte Arbeiten im Schnellbahngewerbe. In: Der Verkehr. Jahr-
buch des Deutschen Werkbundes, Jena, 1914; 42–47.
[3] International Council of Monuments and Sites (ICOMOS). Principles for the Analysis,
Conservation and Structural Restoration of Architectural Heritage. Charter of Victoria
Falls, 2003.

[4] International Council of Monuments and Sites (ICOMOS). Recommendations for the Anal-
ysis, Conservation and Structural Restoration of Architectural Heritage. Guidelines on the
Charter of Victoria Falls, 2003.
[5] Roca P. The study and restoration of historical structures: from principles to practice. In:
Proceedings of the Fifth International Conference on Structural Analysis of Historical Con-
structions, Lourenço PB, Roca P, Modena C, Agrawal S (eds), New Delhi, 6–8 November
2006, 2006; 9–24.
[6] Fischer M, & Lorenz W. Stahlbau unter Denkmalschutz – Grundinstandsetzung von Vi-
adukt und Bahnhöfen der Hochbahnlinie U2 in Berlin-Prenzlauer Berg. Stahlbau 2011;
80(6): 419–427 (incl. further references).


The Main Avenue Bridge, Cleveland,
Ohio, USA

Dario A. Gasparini, Professor Emeritus of Civil Engineering, Case Western Reserve

University, Cleveland, Ohio, USA
William Vermes, Professional Engineer, Jones Stuckey, A Division of Pennoni, Cleveland,
Ohio, USA

The Main Avenue Bridge in Cleveland is a remarkable engineering work in many ways. It is a
large-scale, complex design and has an interesting social history. It embodies many structural
innovations, and it was carefully detailed with consistent aesthetic principles. It was built with
great efficiency in 17 months, and its performance over 77 years has been excellent. The bridge
has a total length of approximately 1800 m (or 2440 m with approaches) with at least five hori-
zontal curves and several vertical curves. The bridge was built to provide an east–west transpor-
tation link that did not require motorists to go through downtown Cleveland streets. It replaced
the Main Avenue swing bridge over the Cuyahoga river, which was a traffic bottleneck, and it
relieved traffic from its southern neighbor, the Detroit-Superior Bridge.

Planning and Design

Planning for the bridge began in 1928, but the Great Depression and failure of a county bond
levy in 1930 halted the project. However, in September 1937, it became one of the initial pro-
jects funded by the Federal Emergency Administration for Public Works, whose charge was to
fund meritorious public works with the intent to “provide stimulus” to the American economy.
Engineering staffing began in August 1937 and was completed in December. The Cuyahoga
County Engineer’s Office received over 130 applications, many from graduates of leading US
engineering schools. About 35 engineers were hired to design and supervise construction at a
typical salary of $400 per month. At this time, John O. McWilliams was the Cuyahoga county
engineer, W. E. Blaser was county bridge engineer, and Wilbur J. Watson was the project con-
sulting engineer. The person who led all aspects of the design was the chief design engineer
Fred L. Plummer. Prior to joining the County Engineer’s Office, Plummer had been an associate

professor of structural engineering at the Case School of Applied Sciences in Cleveland. He was
a member of the American Welding Society and a strong advocate of welding. He authored two
books, Soil Mechanics and Foundations and Fundamentals of Indeterminate Structures. Given
the focus of his structural book, it is understandable that Plummer set the structural theme for
the project: continuity of structural forms for improved structural efficiency. This theme evoked
aesthetic principles that were consistently followed while detailing the bridge. Construction of
the bridge began on 12 May 1938, and the bridge was dedicated 17 months later, on 6 October
1939. Its total cost, including $1.6M for land acquisition, was $7.2M.

Structural Forms
Plummer and his design engineers used a variety of structural forms, including:
• Plate girders, continuous over four spans, for the lakefront ramp length ≈ 262 m
• Eight “continuous cantilever” riveted rigid frames for the eastern approach length ≈ 366 m
from West 3rd street to West 9th street
• Continuous beams on rigid frame bents plus one simply supported truss for length ≈ 213 m
the eastern approach from West 9th street to “Pier 10”
• Ten “continuous cantilever” trusses length ≈ 768 m
• Four spans of continuous reinforced concrete beams on concrete rigid frames length ≈ 55 m
• Riveted plate girder for the West 25th street overpass length ≈ 27 m
• Continuous riveted rigid frames for a “subway ramp” overpass length ≈ 69 m
• Continuous barrel-type concrete rigid frames for the West 28th street subway length ≈ 17 m
ramp overpass
• Three-span welded rigid frame bridges for the West 28th street overpasses. length ≈ 36.6 m
In this brief paper, it is not possible to provide a complete description of all these structural
forms. The following are brief highlights of some of the innovative aspects.

Continuous Plate Girders

Reference [1] described the design of the lakefront ramp of the eastern approach. The ramp has
a 4.15% grade and accommodates a horizontal curve. The 16.7 m-wide roadway is supported
by three plate girders, continuous over four
spans. Figure 1 shows the 82.6 m span plate
girder, which at that time was the longest plate
girder span in the Americas. Its depth over the
piers is 4.78 m.

The ramp is at a 57° skew with respect to

its piers, which are aligned with the railroad
tracks below. The transverse joists were also
made continuous over the three girders. Steel-
grid decking, later filled with concrete, was
welded directly to the joists and provided
lateral bracing to the top flanges of the gird-
Fig. 1: Plate girders of 82.6 m span on Main ers. Stress resultants in the continuous girders
Avenue Bridge lakefront ramp were determined using Hardy Cross’ moment

distribution method of analysis, published in 1933, and later checked using the slope-deflection
method. The girders were fabricated by Bethlehem Steel in Bethlehem, Pennsylvania, and were
shipped by train to the site in sections that were up to 42.7 m long.

Continuous Cantilever Trusses

After the lakeside ramp, the bridge continues with a set of eight spans of continuous riveted rigid
frames and continuous beams on rigid frame bents. The structural form changes to a set of ten
truss spans near the eastern bank of the Cuyahoga river, as shown in Fig. 2.

The trusses are of “continuous cantilever” construction; that is, alternate spans have two cantile-
ver arms that support “drop-in” spans. The longest truss span, over the Cuyahoga river, is 122 m.
The trusses have a deviation in their alignment to accommodate a horizontal curve. At the loca-
tion of the deviation, special details were developed to allow for radial thermal movements.2
Several design decisions were made based on aesthetics, including the ratios of the truss depths
at midspan to those over the piers. To provide a “modern” appearance, no laced sections were
used, and to minimize changes in cross-sectional dimensions, higher-strength “copper bearing
silicon steel” was used where required by greater element forces.

Welded Steel Rigid Frame Bridges

Two separate bridges were built to pass over West 28th street. As shown in Fig. 3, each bridge is
formed by two rigid frames, with a total length of 36.6 m and a main span of 21.3 m. Although
rolled sections were used for both the beams and the columns, the beam-to-column connections,
the splices and the joist and deck connections were all of welded construction; no riveting was
used (except for the fascia plates). Reference [3] provided a full discussion of the design and
construction processes.

Welding processes were, of course, developed much earlier in the 20th century, and applications
to shipbuilding and aircraft fabrication were pioneered in Europe. In the USA, Ref. [4] noted
that “serious commercial development of structural welding commenced in 1926”. By that time,

Fig. 2: Continuous cantilever trusses of the

Main Avenue Bridge (courtesy: Cleveland Fig. 3: Welded rigid frame bridge over West
Memory Project) 28th street

processes for the economic production of heavily covered electrodes had been perfected, and
interested companies, such as Westinghouse Electric, General Electric, and Lincoln Electric,
heavily promoted welding technologies. Reference [4] described an all-welded steel plate girder
railway bridge and an all-welded truss bridge built in the USA in 1928 as well as welded highway
bridges completed in Poland and Czechoslovakia in 1928 and 1931, respectively. Reference [5]
described international developments on welded bridges, especially the famous Belgian Vier-
endeel frames. Reference [6] discussed German all-welded highway bridges; by 1939, “150
railroad bridges and 500 highway bridges” had been welded in Germany.7 Many were plate gird-
ers, but other forms were also built. Although a few welded plate girders and truss bridges were
built in the USA in the early 1930s, US engineers, in general, were slower to accept all-welded
highway bridges, primarily because of concerns over fabrication-induced stresses, fatigue and
brittle fracture. The American Welding Society (AWS), which was founded in 1919 and which
has published The Welding Journal continuously since 1922, published their first bridge welding
code in 1936.8 The James F. Lincoln Arc Welding Foundation was also established in Cleveland
in 1936 to promote and disseminate information on welding technologies. The American Asso-
ciation of State Highway Officials (AASHO) also had separate provisions on welding bridges by
the end of 1935, but these were not included in their 2nd Edition of the Bridge Code, published
in 1935. AASHO and AWS collaborated on revisions, which were incorporated into their 1941
revised codes. The design of the West 28th street bridges must have followed the AASHO load
provisions and either the AWS or the AASHO welding provisions. Results of a survey of state
highway officials on welding practices were reported in the March 1939 issue of The Welding
Journal. Although most states were using welding for repair, strengthening and attaching “sec-
ondary members”, of the 41 states that responded, not one reported using welding for “primary
members”. Therefore, the West 28th street overpasses of the Main Avenue Bridge, completed in
1939, may be the oldest extant welded rigid frame bridges in the USA.

Piers and Foundations

The piers and foundations for the bridge were challenging because of the restrictions imposed
by the railroads, the presence of “construction fill” near the Lake Erie shore and the potential
instability in the riverbanks under the loads
imposed by the trusses. The piers for the lake-
front ramp, visible in Fig. 1, were especially
innovative. Because of size limitations within
the railroad right-of-way, the designers used
composite sections; that is, steel outer shells
with appropriate shear transfer attachments
were filled with reinforced concrete. To trans-
fer moments in the piers to the foundation,
two layers of steel grillages were fastened at
the bases of the steel pier shells. A rectangular
sheet-piling wall was then driven, and the gril-
lages were embedded in concrete. The piers
Fig. 4: Hyperbolic paraboloid surfaces on for the trusses, visible in Fig. 2, were hollow
concrete piers for trusses (Cleveland Memory sections with a wall thickness of 686 mm,
Project) supported by pile foundations. Curiously, one

of the faces of each pier is a hyperbolic paraboloid surface, as shown in Fig. 4. Although increas-
ing cross-sectional dimensions from top to bottom is structurally sensible, affecting this by a
warped surface seems to have primarily been an architectural flourish.

The foundations, piers and superstructures of the Main Avenue Bridge were carefully designed
and have performed exceptionally well. However, two design decisions, coupled with uneven
maintenance and liberal use of deicing salt in the severe Cleveland winters, caused problems.
The first decision was to use a concrete-filled, steel-grid deck, welded directly to the stringers.
The second was to provide roadway drainage by allowing water to flow through open finger
joints in the deck to a collection system below. Over time, the drainage system failed at many
locations, allowing salt-laden water to flow over structural members, causing corrosion. The
water also caused corrosion in the steel-grid deck, which in turn led to buckling of the pave-
ment and closure of expansion joints. This distress necessitated a series of nighttime closures
and emergency repairs. In 1991, the bridge was closed completely, and an18-month rehabili-
tation was performed. The entire 25 m-wide, steel-grid deck and the supporting stringers (to
which the deck was welded) were replaced. The bridge now has a reinforced concrete deck over
new stringers. Superstructure components were repaired or replaced as required, including the
replacement of 40% of the floor beam cantilever brackets. Sealed expansion joints with a new
drainage system were also installed. The total cost of this rehabilitation was $65 million. For the
past 24 years, the Main Avenue Bridge has served the public well. In 2007, a complete repaint-
ing of the truss spans revealed significant section loss in two truss lower chord box members
requiring emergency strengthening. Since 2014, a series of other rehabilitation projects have
been performed, including strengthening of select gusset plates following a recent structural
load rating and revised section loss measurements. During the 77-year service of the Main Ave-
nue Bridge, no repairs have been performed on corrosion-resistant silicon steel truss members
or on the welded frame superstructures of the West 28th street overpasses.

[1] Plummer FL. Girder span record boosted to 271 ft. Eng. News-Record 1941; March 27:
[2] McWilliams JO, Blaser WE, & Plummer FL. Features of Cleveland’s Main Avenue Bridge.
Roads and Streets 1939; December.
[3] Plummer FL. Welded rigid frames, European style. Eng. News-Record 1940; July 18: 55–
[4] Fish GD. Arc-Welded Steel Frame Structures McGraw-Hill Book Co.: New York, 1933.
[5] Grover LM. Foreign countries lead U.S. in welded bridges. Eng. News-Record 1936; May
14: 703–709.
[6] Schaper G. New German bridges. Welding J. 1937; September: 32–39.
[7] Kommerell O. Bridge welding in Germany. Welding J. 1939; August: 472–474.
[8] American Welding Society. Specifications for Design, Construction, Alteration and Repair
of Highway and Railway Bridges by Fusion Welding AWS: New York, 1936.


Renovation of a Historic Railway
Lift Bridge

Jurgen Voermans, Struc. Eng., Royal Haskoning DHV, Rotterdam, The Netherlands,
Jaco Reusink, Struct. Eng., Engineering Department, Municipality of Rotterdam, Rotterdam,
The Netherlands

The Koningshaven bridge in Rotterdam is a vertical lift-type, movable railway bridge. The
bridge was completed in 1927 and was replaced by a tunnel in 1993. The bridge was designated
as a national monument in 2000. To preserve the bridge for future generations, an extensive
renovation project is under execution. This paper elaborates on the history, the design, the struc-
tural assessment and major challenges of the structural retrofit.

Brief History
The construction of the Koningshaven bridge was part of a major project extending the connec-
tion of the Amsterdam–Rotterdam railway line to the Moerdijk–Antwerpen railway line. This
part of the project involved the accomplishment of a double-track railway through the densely
populated Rotterdam inner city and the crossing of the Nieuwe Maas river. The Koningshaven
bridge was completed in 1877 connecting the southern part of the river and canal crossing
between the northern shore of the Nieuwe Maas river through Noordereiland and the southern
shore of the Koningshaven (Fig. 1). The Koningshaven bridge consisted of a steel arch with a
span of 80 m on both sides of a symmetrical swing bridge with a total length of 54.5 m. Open-
ings of 20 m allowed vessels to pass on either side of the central pivot pier (Fig. 2). As time
progressed, the swing bridge could no longer meet the requirements of the busy navigation
and railway traffic. The width of the bridge openings was too narrow, and the bridge had to
be opened frequently. Both navigation and railway traffic were seriously obstructed. Several
collisions occurred as a result. The collision on 2 November 1918 of the German steamship
Kandenfels was decisive in the discussion to replace the swing bridge with a vertical lift bridge.

Fig. 1: Map of Rotterdam in 1888. The loca-

tion of the bridge is indicated by the red arrow Fig. 2: Swing bridge crossing the Koningshaven

In 1927, the swing bridge was replaced with the current vertical lift bridge designed by ir. P.
Joosting (1867–1942), chief engineer of the Bridge Construction Department of the Dutch Rail-
ways. In 1993, a 2796 m long four-track railway tunnel was opened, and the railway bridges
spanning the Maas River became superfluous. After fierce protests against the proposed demo-
lition, the Koningshaven bridge was designated as a national monument. The municipality of
Rotterdam obtained the ownership of the bridge from Dutch Railways. At that time, the bridge
had already suffered from a decade’s lack of proper maintenance. Today, the bridge is highly
admired as a memory to the Industrial Era, with a dense concentration of harbor activities at the
inner city river south bank. The bridge is regarded by the public as a national icon.

Historical Design
Until 1927, the approach spans of the swing bridge remained in place. The new lift bridge consists
of two towers with a movable bridge span in between, which is moved up and down by means of
ropes. The lift span is balanced by counterweights and covers a span of 53.5 m; the distance between
the central planes of the main trusses is 8.8 m. Towers rise 60 m from the base to the top. The vertical
clearance at mean high water level provided under the soffit of the lift span is 45 m. The complete
steel superstructure consists of built-up steel members made of plates, angles and channels riveted
together. The drive system applied is innovative. Instead of placing the drive machinery on top of
the lift span in accordance with a Waddell-type span drive vertical lift bridge that was commonly
applied at the time, it was placed in a machine room at the southern tower. This resulted in a signifi-
cant reduction of structural weight of the lifting part. The lift span is suspended by 48 counterweight
ropes with a diameter of 40 mm, 12 ropes at each corner. The ropes pass over eight cast-iron sheaves
at the top of the towers and continue down from the other side of the sheaves to the counterweights.
The sheaves have a diameter of 3.60 m and contain six sheave grooves (Fig. 3). Lever-type equal-
izers are used at the connections to the lift span to ensure equal loading of the ropes (Fig. 4).

To pull the span up and down, eight operating ropes with a diameter of 26 mm are used (Fig. 5).
The ends of these ropes are attached to geared operating drums in the machinery room. The four
uphaul ropes extend from the drums upward and run around intermediate sheaves to sheaves at
the top of the southern tower and then to the lift span. At both ends of the lift span, they make 90°
turns around deflector sheaves upward to the top of the northern tower where they are attached.

Fig. 3: Cast-iron sheave supported by a spher-

ical roller bearing Fig. 4: Lever-type equalizer

Fig. 5: Course of counterweight and operating

ropes: (a) downhaul rope; (b) uphaul rope; (c) Fig. 6: (a) Counterweight ropes; (b) course of
counterweight rope1 leveling ropes1

The four downhaul ropes are wound in the opposite direction on the drums and run around inter-
mediate sheaves at the base of the southern tower to the lift span. At both ends of the lift span,
they make 90° turns around deflector sheaves downward to the base of the northern tower where
they are attached. Leveling ropes are applied in order to maintain the two ends of the span at the
same elevation during operation to prevent unsynchronized movement. These ropes consist of
two pairs of ropes with a diameter of 26 mm and a Z-shaped course. The length of these ropes
does not change during raising and lowering of the span (Fig. 6). Two ropes run from the base of

the southern tower around deflector sheaves at both ends of the lift span to the top of the northern
tower; the other two ropes run from the base of the northern tower to the top of the southern tower.

In the machinery room, one direct current motor of 200 HP drives through a reduction gearing
and a shaft of two pinions. These mesh with the ring gears of both operating drums. At both
points, cast in situ drums with brakes with a diameter of 2 m are attached. The machinery can
move the lift span at a speed of 0.9 m/s.

In 2016, the bridge was subjected to an extensive renovation. The purpose of the renovation
was technical and durability lifetime extension of minimum 15 years. During the renovation,
minimal hindrance of the navigation traffic
was accepted. The scope of the renovation
was determined by technical inspections and
both structural and risk assessments (techni-
cal and personal safety), which were required
for movable bridges in the Netherlands. The
structural assessment was carried out accord-
ing to the Eurocodes (2011+NA), in com-
bination with the Dutch NEN 8700 code
covering the structural reliability assessment
of existing structures in case of retrofit and
disapproval. A reliability level of β = 3.3 for
Fig. 7: Removing of the lift span renovation was applied. The structural assess-
ment focused mainly on the towers as the
load-bearing capacity of the approach spans
and lift span proved not to be critical. Distinction has been made between normal operational
use and the temporary renovation period. Only the latter is discussed here. During renovation,
the old lead-based paints were fully removed by blasting and were replaced by a new corrosion
protection system. To facilitate proper conservation application circumstances and to prevent
contamination of the environment due to paint, the steel structure had to be fully enclosed and
conditioned. To achieve this, a scaffold was erected to allow the structure to be wrapped in
plastic foil. Lowering the lift span and consequently raising the counterweights in case of high
wind speeds would conflict with the scaffold. Therefore, a decision was made to remove the lift
span during renovation (Fig. 7). The counterweights were fastened above to the towers by a steel
frame. This procedure has multiple advantages:

• Renovation of the lift span in a conditioned environment enhancing the quality of the con-
• Enhancing safe working (not at height)
• Reduction of wind load effects on the towers
• No obstruction to navigation traffic by the lift span that needs to be lowered halfway at high
wind speeds
In comparison with the situation during normal use, the area exposed to wind was increased,
thus increasing the wind load on the towers as the scaffold is horizontally supported by the

towers. Although the lift span was absent, the total wind load on the towers increased signifi-
cantly. The high wind pressure factors of the truss towers caused higher wind forces than in the
wrapped situation and it was decided that for large truss structures, the Eurocode wind loads
are generally conservative. Minor uplift may occur at SLS (Serviceability Limit State) during
extreme wind gusts, because of the absence of the dead load of the lift span that would act
favorably to fix the towers at their supports. Therefore, work was not permitted on the scaffold
at a wind speed ≥ 9 Beaufort. It has been recognized from previous projects that, after blasting,
the scope of steel renovation may be extended to other parts. In the contract, a procedure was
included to cope with unforeseen situations.

The following conclusions can be drawn:

• Extending the lifetime of historical movable bridges that are designated as monuments is
difficult as the structural modifications due to the technical and personal safety require-
ments (e.g., safety stairs or elevator) are often contradictory to the appearance of the pro-
tected monument bridge.
• The wind load on large truss structures according to the Eurocode is generally conservative.
• The driving gear only requires minor worn-based replacements, such as cables and sheaves.
• Special attention and nonstandard procedures are required for the sensitive high-quality
application of the full new corrosion protection system on historic riveted structures; some
of the locations are difficult to reach for quality blasting and painting, and the steel surface
can be rough due to corrosion.
• Elaborate technical inspections and recalculation proved essential in describing the neces-
sary renovation scope.

[1] Archive Nederlandse Bruggenstichting: historic photographic material.
[2] De Boode A, & Van Oudheusden P. De ‘Hef’ – biografie van een spoorbrug. Uitgeverij De
Hef: Rotterdam, 1985.
[3] Joosting P. De in aanbouw zijnde hefbrug over de Koningshaven te Rotterdam. De Ing-
enieur. 1927; 42(6): 89–103.
[4] Oosterhoff J, Coelman BH, & De Wagt WA. Bruggen in Nederland 1800 – 1940 Beweeg-
bare bruggen. Matrijs: Utrecht, 1999.
[5] Voermans J, & Reusink J. Renovation of a historic railway lift bridge. In IABSE Conference
Geneva, 2015 – Structural Engineering: Providing Solutions to Global Challenges. IABSE:
Zurich, 2015; 474–481.


An (Almost) Extinct Engineering
Heritage Asset—The Case of the
Reichsautobahn Bridges

Roland May, Chair of Construction History and Structural Preservation,

BTU Cottbus-Senftenberg, Cottbus, Germany

In early May 2015, a motorway bridge dating from the 1930s was demolished near Pfungstadt.
Refurbishment had been ruled out, and the bridge had to make way for a replacement. A routine
matter hardly worth a mention, replacing old bridges along motorways and major roads has been
an everyday task for many decades. The headline to an article in the local newspaper announc-
ing the impending demolition work was therefore correspondingly laconic: “Facelift for South
Hesse’s motorway bridges”.1 The demolition, which went almost unnoticed by experts in this
field, however, was a milestone in the story of the treatment of one of the most important herit-
age assets in the history of 20th-century transport. This unspectacular structure was indeed the
last remaining historic bridge on the oldest section of the German motorway network.

The Reichsautobahn and Its Bridges

At the time, the Reichsautobahn was probably the most ambitious traffic infrastructure project in
the world. Work on this motorway network began in the autumn of 1933—only a few months after
the National Socialists had seized power. Although there was plenty of experience from previous
motorway plans to fall back on, these roads were declared to be “roads of the Führer” and were
instrumentalized as symbols of a new order. Nevertheless, even wary foreign countries followed
the rapid construction of an unprecedented road network almost 4000 km long, purely for motor-
ized traffic, with increasingly unconcealed enthusiasm. Today, the Reichsautobahn is quite rightly
regarded as the defining blueprint for modern motorways. Initially, the motorways project fol-
lowed primarily economic goals. Soon, however, cultural aspects increasingly became the focus of
attention, with the motorway network supposedly demonstrating the reconciliation between nature
and technology. However, exploiting this for propaganda purposes proved to be quite complicated
at first; after all, although the motorways were “thousands of kilometres long, they were also rather
flat” (Ernst Bloch). Thus, besides the very ambitious integration into the landscape, the count-

less bridges, as the most significant landmarks,

became central elements in the propagandistic
exploitation (Fig. 1).

Unfortunately, the first bridges, including the

aforementioned bridge near Pfungstadt, were
an utter fiasco in terms of their appearance.
Therefore, the architect Paul Bonatz was
appointed as artistic adviser; he had already
enjoyed great successes in cooperating in the
design of engineering structures. Together
Fig. 1: Landscaping and bridge building in with the engineer Karl Schaechterle and the
1938: motorway in the Wisenta valley near architect Friedrich Tamms, Bonatz supplied
Schleiz crucial ideas for developing a truly high-class
bridge design philosophy for the motorway
network.2 In contrast to almost all other building projects carried out by the National Socialists,
these motorways were widely acclaimed internationally. One remarkable feature was the diver-
sity of the bridges. Modeled on different principles, the bridge designs embodied conflicting
ideas, ranging from pure creative engineering in steel and reinforced concrete to “vernacular”
and even monumental structures in stone. Over a period of not quite ten years, some 6000
bridges and culverts were completed along these motorways, with another 2000 or so at vari-
ous stages of completion. The catchphrase in the propaganda of the National Socialists at that
time was therefore the “land of bridge-building”.3 And indeed, the motorways offered a unique
insight into the various possibilities of bridge building in the second quarter of the 20th century.

A Story of Continuous Losses

Toward the end of World War II, more than just a few motorway bridges were destroyed by the
Wehrmacht in order to hamper the advance of Allied troops. Nevertheless, much of the damage
had been already repaired by the mid-1950s. One real problem for the retention of the Reichsau-
tobahn as a unique infrastructure heritage asset in the coming years, however, was the growing
volume of traffic, particularly in former West Germany. Wider carriageways and increasing traffic
loads frequently led to bridges having to be demolished and replaced by new structures. As the
term heritage began to be applied more widely, engineering structures and facilities started to
attract the attention of the conservationists. The first motorway bridges were declared monuments
in the early 1980s, and the first scientific monograph on the Reichsautobahn appeared at the same
time4—since followed by many other publications. However, these measures could not stop the
ongoing loss of cultural assets, especially after this trend was given new impetus by the unifica-
tion of East and West Germany. A gigantic avalanche of investments was now underway under the
heading of “German Unity Transport Projects”, the intention of which was to upgrade the ailing
transport infrastructure of the former GDR. This resulted in a remarkable paradox. Although the
“untouched state” of the motorways in the east of the country was praised with enthusiasm,5 this
did not change the fact that, very soon, their appearance matched that of motorways in the west.
Countless bridges were demolished in the course of modernization measures, including many
outstanding examples of German engineering, but their potential value as heritage assets was
not seriously considered. Only the intended demolition of the bridge over the Teufelstal (Devil’s
valley) provoked some opposition. This bridge, formerly Germany’s largest arch bridge in rein-

Fig. 2: Examples of the loss of outstanding motorway bridges. Top, left to right: River Elbe, Dres-
den (1934–1935, demolished 1995); Schkeuditz interchange (1935–1936, demolished 2000);
Elbe floodplain, Hohenwarthe (1934–1937, demolished 1994). Bottom, left to right: Lauter val-
ley near Kaiserslautern (1934–1937, demolished 2015–2016); Devil’s valley near Stadtroda
(1936–1938, demolished 1999–2000); Danube, Leipheim (1934–1937, demolished 1999)

forced concrete, had been protected since 1993 and had even been named a potential UNESCO
World Heritage candidate in 1997.6 However, although prominent experts testified that the bridge
could be refurbished, even this icon of the Reichsautobahn was demolished at the turn of the 21st
century. At the same time, demolition continued unabated in the west of the country. Especially
bitter here was the loss of the Danube bridge at Leipheim, the most important example of the
contemporary adoption of Robert Maillart’s concept of three-pin arch bridges (Fig. 2).

The Difficulty of Preserving Historic Transport

Infrastructure Assets
The problem was definitely not restricted to former motorway bridges and certainly did not go
unnoticed. For example, in 1999, a publication by the German National Committee for Heritage
Preservation drew attention to the frightening scale of the destruction of cultural assets in the
realm of transport infrastructure.7 However, to date, we have seen no substantial change in pol-
icy. One reason for this state of affairs can be found in the administrative structures, which make
the preservation of heritage assets such as motorways extremely difficult in Germany. Here, the
fundamental responsibility for issues regarding motorway bridges lies with the highways and
heritage conservation authorities, organized on the level of the separate federal states. Heritage
conservation authorities are normally involved in building measures, even smaller projects, but
even today, they only rarely consider construction history aspects. The reasons for this are, on
the one hand, that some heritage conservation authorities show no particular interest in becom-
ing involved in such matters. On the other hand, their work is hampered by a lack of strategic
preservation plans. Thus, decisions regarding demolition or retention are frequently left solely
to the engineers in the highways authorities. There is no doubt that they have become aware of
the early motorway bridges’ significance and that they certainly base their decisions on the best
of their knowledge and belief. However, an unfortunate mix of a lack of construction history
knowledge, overcautious safety concerns regarding the “theoretical service life” and a prefer-

ence for new structures over refurbishment in the public financing system leads them to favor
the demolition of a historic structure in most instances. As far as this author is aware, Germany
does not yet have an official register of all heritage assets from the time of the Reichsautobahn.
So, it is not surprising that, so far, the group of protected motorway bridges seems arbitrary.
Conspicuous here is the dominance of large stone arch viaducts. Apparently, from the stand-
points of both engineering (good ability to carry higher loads) and heritage protection (espe-
cially vivid depiction of National Socialist monumentality), these bridges seem to represent the
“ideal” heritage asset when it comes to motorway bridge building. The clear preference for a
certain type of bridge is, however, highly problematic because this means that one vital element
of Reichsautobahn bridge building, namely, the surprising diversity, can no longer be seen in an
adequate number of examples.

The situation regarding underpasses and overpasses is particularly frustrating. Seen from the
driver’s viewpoint, although the latter are much more important than large bridges carrying the
motorway, they were and still are being poorly treated. The fact that such bridges normally con-
stitute a series along motorway segments has only been acknowledged once so far, with “route
46” being protected since 2003.8 This motorway segment was never completed, and this fact
undoubtedly simplified the granting of such comprehensive protection. However, the fragments
spread across remote forest areas over a length of about 30 km lack one essential element: the
“experience” of a connecting road. An example from North Rhine-Westphalia shows just how
difficult it is to preserve overpasses on roads that are in use. The “Weg Hesseler” overpass at
Beckum, built in 1938, was one of the first bridges in the world using the Freyssinet prestress-
ing system. A preservation order for this bridge was granted in 1991, and so, it survived the
upgrading measures carried out shortly before the turn of the century, whereas an overpass
built at the same time just a few kilometers away using Ulrich Finsterwalder’s rival system
quietly disappeared in 1996. After considerable doubts arose regarding the structural safety of
the Beckum bridge, a decision was made to relocate the superstructure in 2012. Owing to the
considerable costs, this measure, no doubt carried out with the best of intentions, led to a storm

Fig. 3: Attempts to deal with the motorway bridge-building inheritance: the “Weg Hesseler”
prestressed concrete overpass (1938) following its relocation to the Vellern motorway rest area;
a preserved segment of welded superstructure from the Mühlenfliess Bridge (1937/1938)—now
an exhibit in the motorway history collection at Erkner Motorway Maintenance Depot; the
extension (2003–2005) to the Saale valley bridge at Jena (1938–1941) with an “appropriate”
design language

of public protest. Not totally blameless here was undoubtedly the presentation of the “torso”
on a nearby motorway rest area in a manner that can hardly do justice to the heritage idea. This
is just another instance in a whole series of seemingly almost desperate attempts to preserve at
least some of the motorway bridge-building inheritance (Fig. 3).

An example that demonstrates that a different approach is possible, despite considerable traffic,
is Merritt Parkway in Connecticut, which is protected in its entirety—a length of over 60 km.
However, in the meantime, it would be nearly impossible to find a comparable segment of the
Reichsautobahn that has not been extensively reshaped. Most of the historic sections have been
upgraded or enclosed between noise barriers or given new structures. These days, gaining some
kind of authentic impression of the earlier “adventure of the motorway”9 is virtually only pos-
sible on routes outside the country, in former German territories in Eastern Europe. However,
it is probably only a question of time before these examples also vanish. It would appear that
preserving the last original parts of the early motorway network (and compiling a meaningful
record of structures to be demolished) is only possible when a number of fundamental ideas
recently proposed in a remarkable dissertation10 are quickly put into practice. The most important
element here would be the compilation of a national register containing facts on the general state
of preservation in addition to substantiated information on the historical, construction history
and cultural relevance of individual motorway segments. With such a basis, it would then be
possible to develop targeted preservation plans. However, in Germany, a fundamental improve-
ment to the situation regarding the retention of historic transport infrastructure can only succeed
when the players involved join forces in some kind of association, such as the USA’s Historic
American Engineering Record (HAER), which has been around since 1969. We can already see
a first glimmer of hope: the series Wahrzeichen der Ingenieurbaukunst (landmarks in the art of
engineering) published by the Federal Chamber of Engineers has already had a lasting impact on
German construction engineers’ awareness of their engineering heritage. The recently founded
Gesellschaft für Bautechnikgeschichte e.V. (German Construction History Society) is emerging
as a competent partner for authorities when it comes to construction history issues. There are
also many voluntary groups performing valuable documentary work. Critical, however, is the
fact that German structural engineers must see themselves as advocates of their own history to
a greater extent. The (almost complete) disappearance of the motorway heritage asset vividly
demonstrates the consequences of a lack of awareness of history among this profession. It is true
that the early German motorways are cumbersome in historic and technical terms, but only when
engineers fully appreciate the value of construction history for their current everyday engineering
will they perhaps succeed in preserving a symbolic segment of the Reichsautobahn for the future.

[1] “reh”. Frischekur für Südhessens Autobahnbrücken. Darmstädter Echo 2015; 10 April.
[2] May R. Pontifex maximus. Der Architekt Paul Bonatz und die Brücken. MV-Wissenschaft:
Münster, 2011.
[3] Gruber E, & Schütz E. ‘A land of bridges’. On the conception and presentation of bridges
for the Reichsautobahn in the Third Reich. Daidalos 1995; 57: 20–33.
[4] Stommer R (Ed.). Reichsautobahn. Pyramiden des Dritten Reichs Jonas: Marburg, 1982.

[5] Lucka W. Autobahnbrücken aus der Zeit des Reichsautobahnbaus im Wesergebirge. Ber-
ichte zur Denkmalpflege in Niedersachsen. 1989; 9(3): 130–135.
[6] DeLony E. Context for World Heritage Bridges. ICOMOS: Paris, 1997.
[7] Gympel J. Schrittmacher des Fortschritts – Opfer des Fortschritts? Bauten und Anlagen
des Verkehrs. Deutsches Nationalkomitee für Denkmalschutz: Bonn, 1999.
[8] Stockmann D. Strecke 46. Die vergessene Autobahn, 3rd edn. Stockmann: Veitshöchheim,
[9] Harz H, & Menzel H. Das Erlebnis der Reichsautobahn. Ein Bildwerk Callwey: Munich,
[10] Kriest M. Die Reichsautobahn. Konzeption, räumliche Struktur und Denkmaleigenschaft
eines historischen Verkehrsnetzes. Petersberg: Imhof, 2016.


Construction Technology of Chinese
Woven Timber Arch Bridges

Yan Yang, PhD ; Baochun Chen, Prof.; Shozo Nakamura, PhD ; College of Civil
Engineering, Fuzhou University, China

According to the structural types of the existing timber arch bridges, they can be classified
into three main forms, that is, rib arch, truss arch, woven arch.1–3 The woven arch is a unique
structure and is only built in China, so it is also called Chinese woven timber arch.3 Accord-
ing to their present situation, location and structural details, Chinese timber arch bridges can
be further divided into two types: one is the nonextant ancient Bianhe rainbow bridge and the
other is the extant Min-Zhe timber arch bridge.4,5 The historical records show that the first
ancient Bianhe rainbow bridge was built in China between 1032 and 1033.4 The Chinese timber
arch bridges were widespread in the 11th and 12th centuries over the Bian and Fen rivers, but
none of them survived, and no details of their design and construction technologies have been

(a) (b)

Fig. 1: Two branches of Chinese timber arch bridges: (a) Bianhe rainbow bridge; (b) Min-Zhe
timber arch bridge

recorded. The structures can only be seen from a bridge in the famous painting “Chhing-Ming
Shang Ho Thu” (Festival of Pure Brightness on the River), as shown in Fig. 1a, by Zhang
Zeduan who was an artist who lived during the Northern Song Dynasty (1119–1125).6,7 Fortu-
nately, more than 100 Min (short for Fujian Province)-Zhe (short for Zhejiang Province) timber
arch bridges still exist in mountainous areas in the northeast of Fujian Province and in the south-
east of Zhejiang Province, as shown in Fig. 1b.8

A Chinese woven timber arch bridge achieves large spans by weaving straight logs together
in a special way without nails and ropes, in which the main arch consists of two different
longitudinal polygonal arch systems connected by transverse systems, and the construction
is convenient because the structural members are light and need little processing.8 Due to its
long history and ingenious conception, the Chinese woven timber arch bridge was listed in the
world heritage tentative list by China in 2012. Chinese timber arch bridges are designed and
built by bridge craft workers. The technology has been handed down from masters to their
apprentices, many being father and son, and has thus formed some timber arch bridge fami-
lies with stable characteristics in their construction technologies. However, today, there are
only a few old masters as fewer young people prefer to learn this skill because it is difficult to
find a job. This construction technology is at risk of being lost, and hence, it was listed in the
Urgent Safeguarding List of Intangible Cultural Heritage by UNESCO in 2009.9 It is urgently
important to record the construction techniques, which is the purpose of this paper. Because no
ancient Bianhe rainbow bridge survived, and no detailed records on the construction technol-
ogy have been found, this paper will introduce only the construction technology of the Min-
Zhe timber arch bridges.

Structure of Chinese Woven Timber Arch Bridges

Typical single structures of the two kinds of Chinese timber arch bridges are shown in Fig. 2.
The main structures are similar with small difference, while the Min-Zhe timber arch bridge
has a deck system and covering house.10 The bearing structure of a Min-Zhe timber arch bridge
(Fig. 2b) is constructed using a woven arch consisting of longitudinal and transverse straight
logs assembled in a special manner.

There are two longitudinal systems—one is a three-line polygonal arch with three groups of lon-
gitudinal straight logs of the same length connected with two transverse beams, and the other is a

(a) (b)
Covered house

60 Transverse beam of 2nd system
60 Transverse beam of 1st system
Spandrel structure
Longitudinal beam
30 Spandrel protection plates
21 × 45 = 945 X-bracing
Transverse beam of 2nd system
2nd system of arch ring
1st system of arch ring
The first system The second system

Fig. 2: Sketch of Chinese woven timber arch bridges: (a) Bianhe rainbow bridge; (b) Min-Zhe
timber arch bridge

Upper transverse beams of 2nd system

Level arch ring of 2nd system Straight tenon joint Swallow Tail tenon joint

2nd system
Longitudinal beams
Upper slant rib of 2nd system
General column Level arch rib of 1st system Swallow Tail tenon joint
Straight tenon joint
Swallow Tail tenon joint
Transverse beams of 1st system Straight tenon joint
Below slant rib of 2nd system 1st system
Slant arch rib of 1st system

Fig. 3: The mortise and tenon node

five-line polygonal arch with five groups of shorter longitudinal straight logs connected with four
transverse beams. The longitudinal members are mainly subjected to compressive forces and
take full advantage of the compressive strength of timbers parallel to their fiber. The transverse
members connect the two longitudinal systems into an integral structure. All the components are
connected with the mortises and tenon nodes without nails and ropes, and two kinds of mortise
and tenon nodes are used, including swallow tail tenon and straight tenon, as shown in Fig. 3.

Construction Technology of Chinese Timber Arch Bridge

In the traditional Chinese society, similar to other construction engineering activities, timber arch
bridge construction also has some important religious rituals and folk belief behavior.11 However,
this section concentrates on the construction technology, including the following five main steps.9

Selection of Bridge Location

Selection of a bridge location generally follows Fengshui (wind and water) to meet a quest for
an auspicious location in ancient China. The bridges are located at the site called Shuikou (the
mouth of water), the downstream of a river of a village. It is a belief in Fengshui that this can
cause peaceful lives and prospects for the local people. Many of China’s surviving timber arch
bridges can be seen as proof of the reasonability of the traditional site selection method. Some
rules of Fengshui reflect the consideration of the hydrological and geological conditions of the
bridge site. The bridge site selection also relies on the rich experience, high-level techniques and
wisdom of the masters. In some bridges,
the arches are directly erected on a natural
cliff and crag with shallow carves without
an abutment, as shown in Fig. 4.

Construction of Abutments
and Piers
Most abutments and piers of the Chinese
timber arch bridges are built with big
gravels or block stones on a spread foun-
dation. These structures were all built by
manual methods. There was no advanced
Fig. 4: The abutment equipment for elevation measurement in

the past. A half-section bamboo pipe (the knots were cut off) filled with water was used. If the
bamboo is not long enough for the bridge span, several bamboos may be used with temporary
supports. The bamboos are connected through overlapping and sealed with yellow wet clay in
the joint bottom to prevent the water from seeping out. The level can be reached by adjusting the
height of the bamboos in the supports.

Selection of Materials and Treatment of Structural Members

Chinese fir is a kind of tree popular in the Min-Zhe areas. The fir tree grows fast, and its wood
is easily processed and is not easily subjected to decay or easily eaten by moths. Moreover, its
mechanical performance is more stable than other types of wood. Consequently, it becomes the
first choice for the main construction material by craft masters. Trees with suitable sizes in the
local area are strictly selected, accompanied by some important religious rituals and folk beliefs
to show their best wishes. The selected trees are cut down and transported to the bridge site. Then,
they are constructed according to the design of the craft master. Only simple processes are needed
on the raw materials. All the steps are carried out by manual methods with traditional tools such
as the Luban rulers, carpenter’s ink markers, wooden fork horse, axes, chisels, planers and saws.

Erection of Arch Ring

Erection of the arch ring is the most important step in the construction. It consists of four steps
for a single span bridge (without piers), as illustrated in Fig. 5. The four steps are as follows: (1)
erection of the first system (three-line polygonal arches), (2) erection of vertical columns (also
called general column in Chinese Folk) on the abutment as elevation scales, (3) erection of the
second system (five-line polygonal arches) and (4) installation of X-bracings.

Because no modern lifting equipment was available in ancient times, primitive wood winches
were used for erection, and the wooden brackets were used to support the springing member

1 2

3 4

Fig. 5: The process of erection of the arch ring


(a) (b)

(c) (d)

Fig. 6: Erection of the arch ring: (a) the bracket, (b) wood winch, (c) lifting arch ring members
and (d) installing transversal beam

and arch rings during construction. Two brackets made of two main columns, a cross beam and
several diagonal strut members, and located near the two knee positions of the first system are
built, as shown in Fig. 6a. The members in the first system are erected by wood winches (Fig.
6b) and supported by two brackets (Fig. 6c). After all the slat springing members in the first
system are erected in position, two transverse beams are hoisted up (Fig. 6d), and tenon joints
of the members are inserted into straight mortises of the transverse beam. All the members on
one side are joined together and form two frames laid on the brackets. Then, the crown members
are inserted into the swallow tail mortise of the transverse beams to connect the two springing
frames together and form three-line polygonal arches.

The second system of the five-line polygonal arch ring is easier to build after the completion
of the first system. Its longitudinal members are erected on gaps of the first system from the
springings. When all the crown members of the second system are installed to close the arch, the
main arch construction is completed, as shown in step 3 in Fig. 5. In order to improve the spatial
stability of the main arch, two X-bracings are employed, with one being inserted into transverse
beams with the swallow tail tenon and the other being inserted into vertical columns in the abut-
ment with a straight tenon. Finally, wood blocks are inserted between springing members to

enhance the integrality of the arch ring. It should be pointed out that, in order to prevent serious
damage of the wood members during construction, all the tools, including punner and hammers,
are made of wood instead of iron or steel.

Construction of Spandrel Structures and Covering House

The spandrel structure of the Min-Zhe timber arch bridges, also called horse-leg in Chinese tra-
dition, consists of a pair of inclined members standing on the springing and two or three vertical
or inclined members standing on a quarter transverse beam in the second system. The members
are connected with mortise and tenon joints. The deck system consists of deck transverse beams,
longitudinal beams and deck slabs. Generally speaking, there are six deck transverse beams
(three for each side). One of them is close to the abutment and is also used as the transverse
beam of the general column, while the other is located in quarter span and also serves as the
transverse beam of the horse-leg. The transverse beam near the crown usually utilizes the upper
transverse beam of the second arch ring system. A covering house is similar to a local general
house and is built from the central part to both sides. Names of craft masters and their chorogra-
phy and pedigrees are written on the ridge of the covering house for recognition of their partici-
pation and contribution. During the construction of the covering house, raising of the ridgepole
is a high point that follows a ritual as in building houses and temples in Chinese traditional folk
customs. When the bridge is completed, a ceremony of completion will be held on an auspicious
day. After that, the bridge is open to the public.

Concluding Remarks
Chinese woven timber arch bridges are not only an essence of architecture but also an important
heritage structure in China and even in the world. The traditional construction technology is a
precious intangible cultural heritage handed down from ancient generations. Acquiring the con-
struction technology of Chinese timber arch bridge is the key issue in maintaining the existing
bridges and building of new ones. Recording the construction technology may not only bring the
benefit of learning the technology and attracting more young people to learn, thus preventing its
disappearance in our generation, but it could also be a reference to create innovative construc-
tion techniques for the construction of modern arch structures.

[1] Fernandez Troyano, L., 2003. “Bridge Engineering—A Global Perspective”. Thomas
[2] Chen B, Yang Y. Introduce foreign timber arch bridge. Proceedings of the Third China In-
ternational Symposium on the Covered house Bridge of Timber Arch Structure in Pingnan,
China, October 2009, 2009; 237–241 (in Chinese).
[3] Yang Y, Chen B. Comparative study on the timber arch bridge in China and abroad. Pro-
ceedings of the Third Chinese Colloquium on Ancient Bridge, Nanjing, China, 2010; 103–
110 (in Chinese).
[4] Huangcheng Tang. 2000. History of Science Technology in China, Bridge Volume Science
Press. Beijing, China. (in Chinese).

[5] Huangcheng Tang. 2010. Chinese Timber Arch Bridge. China Building Industry Press (in
[6] Ceraldi C, Ermolli ER. Timber arch bridges: a design by Leonardo. Arch Bridges IV—
Advances in Assessment Structural Design and Construction, Barcelona, Spain, 2004,
2004; 69–78.
[7] Yang Y, Chen B, Gao J. Timber arch bridges in China. Proceedings of the Fifth Interna-
tional Conference on Arch Bridge, Madeira, Portugal, September 2007, 2007; 171–178.
[8] Yan Yang, Shozo Nakamura, Baochun Chen and Takafumi Nishikawa: 2012 A survey on
existing China timber arch bridges, J. Civil Constr. Mater. (Jpn. Soc. Civil Eng.), Vol. 28,
[9] Fengfang Zhou,Zeqi Lu,Xudong Su, 2011. Traditional Construction Technology of China
Timber Arch Bridges, Zhejiang People’s Publishing Press. (in Chinese).
[10] Yan Yang, Shozo Nakamura, Baochun Chen and Takafumi Nishikawa. 2014. The origin of
timber arch bridges in China, J. Jpn. Soc. Civil Eng., 2, 54-61.
[11] Ronald G. Knapp, Peter Bol, A. Chester Ong. 2008. Chinese Bridges: Living Architecture
from China’s Past, Tuttle Publishing, America.
Afterword: Learning from the Past
to Build the Future

The present SED is a compilation of contributions devoted to the vast topic of history of struc-
tural engineering as well as interventions on heritage structures and structures of high cultural
values. Various, sometimes opposed, viewpoints and approaches are expressed and presented.

IABSE Working Group 9 “Construction History” is aware of the rather heterogeneous and con-
troversial nature of the content of this SED. However, this shall stimulate and provoke lively
discussions within the structural engineering community who needs to increase the awareness
of historical and cultural aspects of structures and structural engineering. Current structural
engineering methods and practice are only at the very beginning of effective engineering inte-
grating historical and cultural aspects in the assessment of existing structures and in intervention
projects to adapt or modify structures of cultural values for future demands.

Current structural engineering is predominately driven by a spirit to design and build new struc-
tures “out in the green”. Today’s structural engineers’ vocation still is to design and build, even
when dealing with existing structures ! For many structural engineers, the opinion still prevails
that an existing structure has a finite service life of 80–100 years and then needs to be replaced
by a new structure. While this spirit was maybe rational 50 years ago, it is nowadays far away
from modern society’s demands calling for a focused approach on existing structures, in par-
ticular those of high cultural values. Existing structures are an asset and wealth of a society, and
structural engineers are called upon to maintain and enhance the existing structures effectively,
within the availability of limited (public) funds … instead of replacing existing structures by
new construction.

The main problem is that most structural engineers have little or no education in the engineering
of existing structures, including history of structures. They are not even aware of the opportu-
nities available for the effective modern engineering of existing structures. This problematic
situation is due to traditional civil engineering curricula at most of the technical universities
still focusing largely on the design of new structures in reinforced concrete, steel and may be
timber following provisions of current codes and standards. Considering this rather bureaucratic
and uninspiring design education of structural engineers, it is not surprising that most design
engineers are nowadays considered and treated as “code checkers”. They often have a limited
understanding of the broader context of their design solutions and insufficient skills to collabo-
rate with other professionals like architects and environmental engineers.

sed_Afterword.indd 205 12/4/2017 9:11:47 PM


IABSE Working Group 9 “Construction History” largely discussed these professional issues
during its meetings and excursions, and concluded that structural engineers definitely need to
learn from the past to build the future !

However, this discussion is not specifically reflected in the present SED, and as a conclusion,
WG 9 decided to highlight this issue in this afterword. Two topical issues should be developed
further in the near future within IABSE and its Working Groups:

• Education of structural engineers needs a fundamental change. Modern curricula should be

based explicitly on the needs of “Engineering of existing structures”. Design of new struc-
tures should be relegated. In addition, “History of structures and structural engineering”
needs to become a mandatory fundamental engineering discipline. Greater emphasis should
be given to principles comprising the essentials of all construction materials in both existing
and new structures. Obviously, new technologies like monitoring of structures, advanced
computational models for structural analysis as well as novel high-performance materials
and structural systems are part of a modern curriculum.
• Information and data regarding ideas and solutions of structural engineering in the past
should be scientifically analyzed and exploited. This “mining” process shall produce and
enhance novel knowledge and know-how to design innovative interventions on existing
structures and to create new structures. In the past, several innovative ideas could not be
realized because of lack of appropriate means (materials, methods and tools) available in
former times. Current and future technologies, in particular the computer-based ones, may
help to implement and enhance former structural systems using high-performance materi-
als. Understanding of the past is a rich source of inspiration for structural engineers.

Knowing the past is indispensable for modern structural engineering ! With the present SED,
the IABSE Working Group “Construction History” wanted to make a contribution to this impor-
tant goal.

Eugen Brühwiler, Professor at the Swiss Federal Institute of Technology (EPFL),

Lausanne, Switzerland, Vice-Chairman of IABSE Working Group 9
“Construction History”
August, 2017

sed_Afterword.indd 206 12/4/2017 9:11:47 PM

List of SEDs Available

Following SEDs can be purchased at the IABSE Onlineshop:

SED 15 - Engineering History and Heritage Structures -

Viewpoints and Approaches
Authors: Eberhard Pelke, Eugen Brühwiler, 2017

SED 14 - Sustainable Structural Engineering

Editors: J. Anderson, C. Bucher, B. Briseghella, X. Ruan,
T. Zordan, 2015

SED 13 - Use of Timber in Tall Multi-Storey Buildings

Authors: Ian Smith, Andrea Frangi, 2014

SED 12 - Case Studies of Rehabilitation, Repair, Retrofitting, and Strengthening

Editors: M.M. Bakhoum, Juan A. Sobrino, 2010

SED 11 - Design for Robustness

Authors: Franz Knoll, Thomas Vogel, 2009

SED 10 - Structural Use of Glass

Authors: Matthias Haldimann, Andreas Luible, Mauro Overend, 2008

SED 9 - Cable Vibrations in Cable-Stayed Bridges

Author: Elsa de Sa Caetano, 2007

SED 8 - Use and Application of High-Performance Steels for Steel Structures

Author: Hans-Peter Günther (Editor), 2005

SED 7 - Use of Fibre Reinforced Polymers in Bridge Construction

Author: Thomas Keller, 2003

SED 6 - Structural Bearings and Expansion Joints for Bridges

Author: Günter Ramberger, 2002

SED 5 - Introduction to Safety and Reliability of Structures

Authors: Jörg Schneider and Ton Vrouwenvelder
(3rd reviewed and extended edition, 2017)

SED 4 - Ship Collision with Bridges - The Interaction between Vessel Traffic and Bridge Structures
Author: Ole Damgaard Larsen, 1993

SED 3 - Vibrations in Structures - Induced by Man and Machines

Authors: Hugo Bachmann, Walter Ammann, 1987

SED 2 - Dynamic Response of Reinforced Concrete Buildings

Authors: Hajime Umemura, Haruo Takizawa, 1982

SED 1 - Concrete Box-Girder Bridges

Authors: Jörg Schlaich, Hartmut Scheef, 1982

IABSE Members can download SEDs in ePDF format for free in the Members Area at:
SED_15-1TOC.indd 2 04/12/17 8:50 PM
SED_15-1TOC.indd 3 04/12/17 8:50 PM
About the Authors
To provide in-depth information to practicing stuctural
engineers in reports of high scientific and technical standards
Eberhard Pelke earned his civil engineer diploma
on a wide range of structural engineering topics.
from Darmstadt University. He is Head of the
Department of Bridge Engineering at Hessen Mobil
IABSE Bulletin Board:
Road and Traffic Management in Wiesbaden,
H.Subbarao, (Chair), D. Laefer, (Vice Chair), M. Bakhoum,
Germany. He is a Fellow of IABSE and was from C. Bob, M.W. Braestrup, N.P. Hoej, H.H. Snijder, R.von Wölfel,
2013-17 Chair of IABSE WG9 Construction R. Mor, M.G. Bruschi, I. Payá-Zaforteza, S. Kite, M. Garlock.
The International Association for Bridge and Structural
Engineering (IABSE) operates on a worldwide basis,
with interests of all type of structures, in all materials. Its
members represent structural engineers, employed in design,
Eugen Brühwiler is professor of structural engi- academe, construction, regulation and renewal. IABSE
organises conferences and publishes the quarterly journal
neering at the EPFL—Swiss Federal Institute Structural Engineering International (SEI), as well as reports
of Technology in Lausanne, Switzerland. His and monographs, including the SED series, and presents
activities include modern methods to examine annual awards for achievements in structural engineering.
and enhance structures of high cultural value. With a membership of some 4,000 individuals in more than
He is a Fellow of IABSE and was from 2013-17 100 countries, IABSE is the international organisation for
structural engineering.
Vice-Chair of IABSE WG9 Construction History.
The International Association for Bridge and Structural
Engineering (IABSE) was founded as a non-profit scientific
association in 1929. Today it has more than 300 members iQ
over 0 countries.IABSE’s mission is to promote the exchange
of knowledge and to advance the practice of structural
engineering worldwide. IABSE organizes conferences
and publishes the quarterly journal Structural Engineering
International, as well as conference reports and other
monographs, including the SED series. IABSE also presents
With Contributions From:
annual awards for achievements in structural engineering.
E. Pelke, T.F. Peters, M. Traykova, R. Vergoossen. A. Bögle, W. Lorenz, E. Brühwiler,
N. Janberg, J. Romo, B. Addis, M.J. Beiersdorf, J. Steiner, E. Vianen, R. Spaan,
For further Information:
J.F. Duntemann, B.R. Greve, A. Traykov, D. Partov, N. Winterbottom, B. Heres,
T. Chardakova, D. Wendland, P. van Bogaert, R. Barthel, J. Tutsch, J. Jordan, C. Weber, IABSE
c/o ETH Zürich
A. Kostka, M. Fischer, G. Eisele, J. Seiler, V. Wetzk, Y. Rammer, B. Espion, L. Clarke,
CH-8049 Zürich, Switzerland
M. Bartzsch, K. Geißler, D. Gasparini, W. Vermes, J. Voermans, R. May, Y. Yang, Phone: Int. + 41-44-633 2647
B. Chen, S. Nakamura. Fax: Int. + 41-44-633 1241
E-mail: secretariat@iabse.org
Web: www.iabse.org

SED15_Cover.indd 2 05/12/17 2:44 PM

Structures – Viewpoints and Approaches
Engineering History and Heritage Structural Engineering Documents
Structures – Viewpoints and Approaches

Engineering History and Heritage

The present Structural Engineering Document (SED) is a
compilation of contributions devoted to the vast topic of history
of structural engineering as well as interventions on heritage
structures and structures of high cultural values. Various, some-
times opposed, viewpoints and approaches are expressed
and presented. The rather heterogeneous and controversial

Engineering History
nature of the content of this SED shall stimulate lively discus-
sions within the structural engineering community who needs
to increase the awareness of historical and cultural aspects

and Heritage
of structures and structural engineering. Current structural
engineering methods and practice are only at the very begin-
ning of effective engineering, really integrating historical and
cultural aspects in the assessment of existing structures and
in intervention projects to adapt or modify structures of cultural
values for future demands. Knowing the past is indispensable
Structures – Viewpoints

for modern structural engineering !
and Approaches

Eberhard Pelke

Structural Engineering Documents

Eugen Brühwiler

International Association for Bridge and Structural Engineering (IABSE)

SED15_Cover.indd 1 05/12/17 2:44 PM

Похожие интересы