Structural Concrete 01-2012 sample copy by Ernst & Sohn

1 Volume 13 March 2012 ISSN 1464-4177

- Survey of earthquake damage to RC building faĂ§ades - Seismic response of concrete bridges in New Zealand - Determining the in situ strength of existing concrete - Using the MC 2010 LoA approach for punching shear - Ultimate strength of curved strand tendons - Formwork pressures with highly flowable concretes â&#x20AC;&#x201C; part 2

A NEW STAR FOR THE WHOLE OF EUROPE! HALFEN HTA-CE: Anchor rails with ETA approval, CE-certiﬁed.

Since the beginning of 2010 HALFEN rails have been ETA approved. Just one of the advantages you can depend on. You can now proﬁt from a system that is internationally approved in 30 European countries. Greater economical eﬃciency due to adaptable anchor lengths and spacing, and a dimensioning programme that supports the planner.

Improved reliability in planning HALFEN products; Safety, Quality and international projects. Conﬁdence protection for you and your Company in the CE Marking. Highest quality through meticulous third-party and self-monitoring before, during and after the production process. Comprehensive service with catalogues installation guides, software and one-on-one project advice.

www.halfen.com

anzeige_HTA-CE_ETA_E_181x272_feb11.indd 1

09.02.2011 14:55:40

Contents

The photo on the cover shows a view of the concrete canopy of the “Maison de l’Écriture” at Montricher, Switzerland (structural design: “Muttoni & Fernández, consulting engineers”). It is a complex structure combining ordinary reinforcement, post-tensioning and steel fibres, and refined design methods were used for its analysis and design. For most structures, however, design can be performed on the basis of simpler design methods. The need for codes that offer sufficient flexibility to be used in an efficient manner for design of simple and complex structures can be satisfied by using codes based on a Levels-ofApproximation (LoA) approach. The LoA approach provides simple and physically-grounded design expressions (sufficient for a large variety of cases) but whose mechanical parameters can be refined in successive levels when required (for design or for assessment of complex structures). A more in-depth explanation of this approach is provided on pages 32–41 of this issue. (Photo: Rafael Rojas)

Structural Concrete Vol. 13 / 1

Editorial 1

Articles 3

Andrew Baird, Alessandro Palermo, Stefano Pampanin Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

Alessandro Palermo, Liam Wotherspoon, Lucas Hogan, Mitchel Le Heux, Elena Camnasio Seismic performance of concrete bridges during Canterbury earthquakes

R.D.J.M. Steenbergen, A.H.J.M. Vervuurt Determining the in situ concrete strength of existing structures for assessing their structural safety

Aurelio Muttoni, Miguel Fernández Ruiz The levels-of-approximation approach in MC 2010: application to punching shear provisions

Johann Kollegger, Susanne Gmainer, Klaus Lehner, Josef Simader Ultimate strength of curved strand tendons

Carl-Alexander Graubner, Erik Boska, Christoph Motzko, Tilo Proske, Frank Dehn Formwork pressure induced by highly flowable concretes – design approach and transfer into practice

March 2012 ISSN 1464-4177 (print) ISSN 1751-7648 (online) Wilhelm Ernst & Sohn Verlag für Architektur und technische Wissenschaften GmbH & Co. KG www.ernst-und-sohn.de

fédération internationale du béton International Federation for Structural Concrete www.fib-international.org

Journal of the fib

peer reviewed journal: Structural Concrete has been accredited with ISI Web of Knowledge since Vol. 10 (2009).

Johan Silfwerbrand Sustainable concrete is more than just durable concrete

61 62 63 63 64 65 66 67 67 68 69

fib-news fib Symposium 2013, Tel Aviv: Call for papers ICCS13: Call for papers fib Symposium Stockholm ICI-fib Workshop, New Delhi 60th Birthday of Harald Müller 65th Birthday of Joost Walraven Recent activities of fib Commission 8 Honorary doctorate to M. Curbach fib Bulletins Congresses and Symposia Acknowledgement

Products and Projects

www.wileyonlinelibrary.com, the portal for Structural Concrete online subscriptions

Bautechnik 81 (2004), Heft 1

Imprint The journal “Structural Concrete”, the official journal of the International Federation for Structural Concrete (fib – fédération internationale du béton), provides conceptual and procedural guidance in the field of concrete construction, and features peerreviewed papers, keynote research and industry news covering all aspects of the design, construction, performance in service and demolition of concrete structures. “Structural Concrete” is published four times per year completely in English. Except for a manuscript, the publisher Ernst & Sohn purchases exclusive publishing rights. Only works are accepted for publication, whose content has never been published before. The publishing rights for the pictures and drawings made available are to be obtained from the author. The author undertakes not to reprint his article without the express permission of the publisher Ernst & Sohn. The “Notes for authors” regulate the relationship between author and editorial staff or publisher, and the composition of articles. These can be obtained from the publisher or in the Internet at www.ernstund-sohn.de/en/journals. The articles published in the journal are protected by copyright. All rights, particularly that of translation into foreign languages, are reserved. No part of this journal may be reproduced in any form without the written approval of the publisher. Names of brands or trade names published in the journal are not to be considered free under the terms of the law regarding the protection of trademarks, even if they are not individually marked as registered trademarks. Manuscripts can be submitted via ScholarOne Manuscripts at www.ernst-und-sohn.de/suco/for_authors If required, special prints can be produced of single articles. Requests should be sent to the publisher. Publisher fib – International Federation for Structural Concrete Case Postale 88, CH-1015 Lausanne, Switzerland phone: +41 (0)21 693 2747, fax: +41 (0)21 693 6245 e-mail: fib@epfl.ch, Website: www.fib-international.org Publishing house Wilhelm Ernst & Sohn Verlag für Architektur und technische Wissenschaften GmbH & Co. KG Rotherstraße 21 12045 Berlin/Germany phone: +49 (0)30/47031-200 fax: +49 (0)30/47031-270 e-mail: info@ernst-und-sohn.de, Website: www.ernst-und-sohn.de Editor Dr.-Ing. Dirk Jesse, Verlag Ernst & Sohn Rotherstraße 21, D-10245 Berlin phone: +49 (0)30/47031-275, fax: +49 (0)30/47031-227 e-mail: dirk.jesse@wiley.com Technical editor Francisco Velasco, Verlag Ernst & Sohn Rotherstraße 21, D-10245 Berlin phone: +49 (0)30/47031-277, fax: +49 (0)30/47031-227 e-mail: francisco.velasco@wiley.com Advertising manager Fred Doischer, Verlag Ernst & Sohn phone: +49 (0)30/47031-234 Advertising Annekatrin Gottschalk, Verlag Ernst & Sohn Rotherstraße 21, D-10245 Berlin phone: +49 (0)30/4 70 31-2 49, fax: +49 (0)30/4 70 31-2 30 e-mail: annekatrin.gottschalk@wiley.com Layout and typesetting: TypoDesign Hecker GmbH, Leimen Printing: Meiling Druck, Haldensleben

Structural Concrete 13 (2012), No. 1

Editorial board Editor-in-Chief Luc Taerwe (Belgium), e-mail: Luc.Taerwe@UGent.be Deputy Editor Steinar Helland (Norway), e-mail: steinar.helland@skanska.no Members György L. Balázs (Hungary) Josée Bastien (Canada) Mikael Braestrup (Denmark) Tom d’ Arcy (USA) Michael Fardis (Greece) Stephen Foster (Australia) Tim Ibell (UK) S.G. Joglekar (India) Akio Kasuga (Japan) Gaetano Manfredi (Italy) Pierre Rossi (France) Guilhemo Sales Melo (Brazil) Petra Schumacher (Secretary General fib) Tamon Ueda (Japan) Yong Yuan (China) Current prices The journal Structural Concrete has four issues per year. In addition to “Structural Concrete print”, the PDF version “Structural Concrete online” is available on subscription through the online service Wiley Online Library. print print print + online (personal) (institutional) (personal) 120.00 €

460.00 €

138.00 €

print + online Single copy (institutional) (print) 529.00 €

34.00 €

Other currencies and bulk discounts are available on request. Members of the fib receive the journal Structural Concrete as part of their membership. Prices exclusive VAT and inclusive postage, errors and omissions excepted. Subject to change without notice. Prices are valid until 31 August 2012. A subscription lasts for one year. It can be terminated in writing at any time with a period of notice of three months to the end of the calendar year. Otherwise, the subscription extends for a further year without written notification. Bank details Commerzbank AG Mannheim account number 751118800 bank sort code 67080050 SWIFT: DRESDEFF670 Periodical postage paid at Jamaica NY 11431. Air freight and mailing in the USA by Publications Expediting Services Inc., 200 Meacham Ave., Elmont NY 11003. USA POSTMASTER: Send address changes to “Structural Concrete” c/o Wiley-VCH, 111 River Street, Hoboken, NJ 07030. Service for customers and readers Wiley-VCH Customer Service for Ernst & Sohn Boschstrasse 12, D-69469 Weinheim Tel.: +49 (0)800 1800 536 (with Germany) Tel.: +44 (0)1865476721 (outside Germany) Fax: +49 (0)6201 606184 cs-germany@wiley.com Quicklink: www.wileycustomerhelp.com © 2012 Ernst & Sohn Verlag für Architektur und technische Wissenschaften GmbH & Co. KG, Berlin

Inserts in this issue: Ortra Ltd., Tel Aviv 61092, Israel; Verlag Ernst & Sohn GmbH & Co. KG, 10245 Berlin

Products & Projects

Aluminium precasting mould system for greater efficiency and convenience Since the introduction of its aluminium precasting mould system, Magdeburgbased B.T. innovation has established a combination of FlyFrame mould carrier and ultra lightweight but immensely strong MagFly® AP magnets in the marketplace. More and more precasting plants in Germany and, indeed, all over the world, are opting for this innovative system. One of those is the Thisted plant, a very successful player in the Danish market. Thisted chose to install the FlyFrame®/ MagFly® AP combination because it gives the company a range of new opportunities. The magnets enable the mould carriers, which support the timber panels, to be positioned and fixed with millimetre accuracy.

totally unnecessary – just like with the well-known MultiForm MagFly® system. The new lightweight MagFly® AP magnet has an enclosed and fully integrated adapter that reliably prevents any cement slurry from leaking into the magnet. The simple handling gives the production manager at the Thisted plant more time to carry out stricter quality control measures on the three-layer sandwich panels the plant produces, or time for other, possibly problematic, production issues. And what at first glance seems to be a gain of just a few minutes, adds up to a considerable number of production hours over a year which have a direct bearing on the efficiency and economy of the production.

Faster, less stressful operations Aluminium not only cuts the weight substantially The use of aluminium plus the optimization of the material of the magnets and the steel components have reduced the weight considerably and at the same time increased the holding power. According to the manufacturer, the new generation of BT magnets therefore achieves the best performance of any such magnets currently available in the marketplace: 4.00 kN/kg (holding power 22 kN/5.5 kg mass). The new MagFly® AP magnets are also equipped with the tried-and-tested, patented MagFly® foot/spring system for effortless positioning. The integral adapter means they can be easily used with all BT mould carrier systems without the need for extra fittings or screw fixings.

Absolutely no chance of cement slurry leakage The ultra lightweight but stable FlyFrame® aluminium mould carrier enables moulds to be set up even faster and easier. Cranes or heavy steel frames are

B.T. innovation’s field staff discovered another very surprising and at the same time positive effect during their trials with German customers. The use of lightweight but stable material encourages personnel to carry out their work much faster but with less stress. The new FlyFrame®/MagFly® AP combination is therefore helping the precast concrete industry to achieve optimum production technology and working conditions plus optimum productivity. Further information: B.T. innovation GmbH, Sudenburger Wuhne 60, 39108 Magdeburg, Germany, tel. +40 (0)391 7352-0, fax +49 (0)391 7352-52, info@bt-innovation.de, www.bt-innovation.de TCT A/S, Stevnsvej 17, 7700 Thisted, Denmark, tel. +45 (0)97 922522, fax +45 (0)97 911522, salg@tct.dk, www.tct.dk

FlyFrame® precasting mould carrier with MagFly® AP magnet (©: BT innovation)

Products & Projects

Fair-face concrete answer for modern architecture in historical setting With its famous water features and, in particular, Hercules Monument, Kassel’s Bergpark can be regarded as the city’s best-known landmark. The monument itself crowns the highest point in the park and has now been given a new visitor centre. The client for the new building, which cost almost €3.5 million and has a floor area of approx. 750 m², was the Federal State of Hesse. Formwork panels from Westag & Getalit AG were used to comply with the high demands placed on the fair-face concrete surfaces. Fig. 1. The new visitor centre for the Hercules Monument in Kassel is the new reception point for the park’s many visitors.

Relief-type façade texture The task of integrating the modern architecture into the historical surroundings was solved with fair-face concrete. The architects from Berlin designed a polygonal figure with a special relief-type texture for the façade which could be built in class SB2 fair-face concrete. Located directly in front of the Hercules Monument, the entire structure is a direct reference to this local landmark. To achieve a perfect balance between the façade and the tuff stone of the monument, and between the new entrance to Kassel’s Bergpark and this historical site, the complete façade was subsequently sandblasted. As a contrast to the uneven look of the façade, the entire interior of the visitor centre was given an SB4 finish (the highest class of fair-face finish). The smooth fair-face concrete wall surfaces and soffits therefore provide a modern contrast to the wooden furnishings, but still fit in well with the style of the interior. Both inside and outside, all the details of this building have been carefully and thoroughly conceived, which meant that the various openings for doors and windows, even the recesses in the soffits for the flush-fitting lighting units, had to be constructed very accurately in order to do justice to the requirements placed on the design.

Extensive testing and sample panels

If you require special prints of selected articles from the journals at Ernst & Sohn? Please contact: Janette Seifert Verlag Ernst & Sohn Rotherstraße 21, 10245 Berlin Fax +49(0)30 470 31-227 jseifert@ernst-und-sohn.de

W i l e y

C o m p a n y

Structural Concrete 12 (2011), No. 4

0204200006_dp_88x63mm.indd 1

0204200006_dp

w ww w. e r n s t - u n d - s o h n . d e

Complying with the requirements placed on the immaculate fairface concrete surfaces required a considerable number of sample panels. The contractor responsible for the structure therefore prepared numerous samples in order to select the right formwork board, release agent and concrete. Indeed, a complete wall was built in advance with the relief-type texture in order to minimize further problems during construction. A total of approx. 1500 m² of Westaspan 450 SP was supplied for this project. This board, suitable for formwork to large areas, has a 450 g/m² film coating on both sides, which makes it ideal for all smooth, seamless concrete surfaces with a matt look. An-

Fig. 2. The façade was built with formwork in SB2 quality and subsequently sandblasted so that the modern structure would blend in with its historical surroundings, especially the tuff stone of the Hercules Monument.

other advantage for work on site was the factory-cutting of the panels: there isn’t a single 90° angle anywhere on the structure – a fact that placed high demands on the formwork. On site, the prefabricated panels were able to be directly erected for the walls and slabs without the need for time-consuming cutting to size. The boards were used in a panel formwork system. Fair-face concrete throughout posed a real challenge The formwork for the relief-type façade proved to be especially difficult and time-consuming. Boards in a huge variety of different lengths, widths and thicknesses had to be added to the formwork panels by hand to give the façade its special texture. To cope with the ensuing offsets within the textured façade, the site management decided to use a virtually self-compacting concrete (F6). Consequently, such a very fluid concrete mix called for well-sealed formwork to ensure that the desired final form was not spoiled. Another challenge was the inclusion of movement joints. In order to avoid disturbing the pattern of horizontal and vertical lines on the façade, all joints were incorporated into the layout of the formwork panels to create a homogeneous appearance. The façade with its formwork in SB2 quality was subsequently sandblasted to match its historical surroundings – especially the tuff stone of the Hercules Monument. Further information: Westag & Getalit AG, Hellweg 15, D-33378 Rheda-Wiedenbrück, tel. +49 (0)5242 17-0, fax +49 (0)5242 17-75000, zentral@westag-getalit.de, www.westag-getalit.de

01.12.2010 15:47:46 Uhr

Responsible for “Products & Projects” (sponsored content): Ernst & Sohn

Easy Design of Anchor Channels: JORDAHL速 EXPERT Software The JORDAHL速 EXPERT software makes it easy to design on the basis of European Technical Approval ETA-09 / 0338 and the European Standard CEN / TS 1992-4-3:2009. It can be used simply, using interactive 3D graphics, ensuring that the anchor channel, loads, and boundary

conditions are input in a rapid and clear manner. The design not only meets the requirements of the individual situation, but also carries out technical and economical optimisation to ensure the best possible solution. www.jordahl.de.

Deutsche Kahneisen Gesellschaft mbH

Products & Projects

Inspired by water, implemented with concrete The design team from Zaha Hadid Architects took its inspiration from the motion of flowing water when designing the exhibition for Roca, the bathroom furnishings and fittings supplier. The result was a style of architecture characterized by highly demanding organic forms that had to be realized in concrete. But the architects met with problems when trying to implement their ideas; most of the contractors they approached turned down the project, saying it was “unbuildable”. And that was exactly the aspect that attracted Kruno Stephan Thaleck, the owner of B & T Bau & Technologie GmbH based in Raubling in southern Germany. This company had become specialized in building extravagant architecture; of the 50+ applicants, it was therefore the only one qualified to construct the fair-face concrete elements plus the 3D plasterboard walls and ceilings. However, more than 2 1/2 years of intensive preliminary work were to pass before Kruno Stephan Thaleck could begin building any components. In this project he faced an even bigger challenge that could only be solved by developing a new building material – and that was the birth of CEton. CEton elements are made from a textile-reinforced fibre composite material based on concrete with an aluminium honeycomb core. It is this new building material that today enables B & T Bau & Technologie to fabricate concrete elements with a thickness of just 60 mm and a weight of no more than 50 kg/m². But that didn’t happen overnight… Prior to its own production, B & T Bau & Technologie GmbH worked closely together with Wacker Chemie AG from Munich. First of all, a concrete mix had to be designed for CEton which would guarantee that the fresh concrete would bond well with the aluminium and also exhibit an excellent flexural tensile strength (10 N/mm²) plus a high compressive strength (> 45 N/mm²). Starting with the specification for the fabrication of the CEton elements, the team selected the appropriate products from Wacker Chemie’s “Etonis” range and put together a tailored polymer blend. It proved to be particularly advantageous that the polymers enhanced the plastic deformability (ductility) of the concrete and hence prevented the concrete elements from cracking during transport and erection. In addition, the “Etonis” products guarantee that the concrete surfaces achieve a good resistance to abrasion. For the actual building of the exhibition elements, the contractor used CEton combined with challenging special formwork plus a suitable release agent. Kruno Stephan Thaleck was well on the way to achieving the designs of Zaha Hadid Architects,

Fig. 1. The unusual concrete elements of the Roca London Gallery remain fascinating even at night.

Structural Concrete 12 (2011), No. 4

Fig. 2. The new construction material CEton enables contractors to build lightweight concrete elements.

which required an interplay between white and grey components as well as curving contours. However, an appropriate cement was also needed and that search led Kruno Stephan Thaleck to the Holcim company. The Holcim range, includes a beautiful white cement which is an excellent basis for pigmented concrete. It was not long before Kruno Stephan Thaleck realized that Holcim White would be the right basis for CEton. Not only the high quality of the cement convinced him, but, in particular, the excellent professional support of Holcim in Vienna. It is the white cement from Holcim that now enables B & T Bau & Technologie to guarantee identical technical features every time, even with components in different colours produced by adding pigments. Dipl.-Ing. Claudia El Ahwany Further information: HOLCIM (SLOVENSKO) A.S., Plant 906 38, Rohoznik Na Zahori, Slovakia, tel. +421 (0)34776 5111, fax +421 (0)34776 5433 Wacker Chemie AG, Hanns-Seidel-Platz 4, 81737 Munich, Germany, tel. +49 (0)89 6279-0, fax +49 (0)89 6279-1770, info@wacker.com, www.wacker.com

Fig. 3. The designs of the architects were inspired by the soft flowing motion of water. (© 1 & 2: B & T Bau & Technologie GmbH; 3: Roca)

Responsible for “Products & Projects” (sponsored content): Ernst & Sohn

Products & Projects

Prestressed Concrete Design acc. to Eurocode 2 with RFEM With the RFEM add-on modules RF TENDON and RF-TENDON Design Dlubal can now offer two programs used to calculate prestressed concrete members with bonded post-tensioned concrete according to EN 1992-1-1 (concrete constructions) and EN 1992-2 (concrete bridges). RF-TENDON is the add-on module required to define tendons. With RF TENDON Design you perform the prestressed concrete design according to Eurocode 2 on the basis of the results calculated with RFEM and RF TENDON.

Working with RF-TENDON First one defines the structure, loads cases (load case for prestress without loading) and loads groups in RFEM. RF TENDON imports the RFEM data including cross-sections and materials. Then, one defines the tendons. The program database offers a variety of prestressing steels for selection. One can define straight and parabolic tendons automatically as well as manually. The tendon layout can be displayed in a 3D rendering overview. The load cases and groups are assigned to time intervals. Finally, the prestressing forces are calculated. They will be transferred automatically to RFEM where the internal forces are determined by calculation. In addition, RF-TENDON takes into account immediate loss of prestressing stress due to friction, anchorage set, relaxation and elastic deformation of concrete.

nal and shear reinforcements. One can just take advantage of the comfortable input tools provided by the program. RF-TENDON Design performs the required ultimate and serviceability limit state design according to EN 1992 1-1 for prestressed concrete members. Optionally, they can be calculated according to EN 1992-2. The design is carried out for axial force, biaxial bending, shear, torsion and combined internal forces. The program checks as well if the reinforcement rules according to Eurocode 2 are observed. The output is shown in clearly arranged results tables. Printout reports are generated in both add-on modules. The reports can be printed or exported in a RTF file. Further Information: Ingenieur-Software Dlubal GmbH, Am Zellweg 2, 93464 Tiefenbach, Germany, Tel. +49 (0)9673 9203-0, Fax +49 (0)9673 9203-51, info@dlubal.com, www.dlubal.com

Using RF-TENDON Design

Fig. 1. Design of crack width in RF-TENDON Design (© Dlubal)

HUGO S.L.C. HENS

Performance Based Building Design 1 From Below Grade Construction to Cavity Walls approx. 260 pages, approx. 172 ﬁgures, Softcover. approx. € 59,–* ISBN 978-3-433-03022-6

Order online! www.er nst-und-sohn.de

Date of Publication: April 2012

Package: Performance Based Building Design 1 and 2 approx. € 99,–* ISBN: 978-3-433-03024-0 Date of Publication: September 2012

Ernst & Sohn Verlag für Architektur und technische Wissenschaften GmbH & Co. KG

W i l e y

Q Just like building physics, performance based building design was hardly an issue before the energy crisis of the 1970ies. With the need to upgrade energy efficiency, the interest in overall building performance grew. As the first of two volumes, this book applies the performance rationale, advanced in applied building physics, to the design and construction of buildings. After an overview of materials for thermal insulation, water proofing, air tightening and vapour tightening and a discussion on joints, building construction is analysed, starting with the excavations. Then foundations, below and on grade constructions, typical load bearing systems and floors pass the review to end with massive outer walls insulated at the inside and the outside and cavity walls. Most chapters build on a same scheme: overview, overall performance evaluation, design and construction. The book should be usable by undergraduates and graduates in architectural and building engineering, though also building engineers, who want to refresh their knowledge, may benefit. The level of discussion assumes the reader has a sound knowledge of building physics, along with a background in structural engineering, building materials and building construction.

Customer Service: Wiley-VCH Boschstraße 12 D-69469 Weinheim

*€ Prices are valid in Germany, exclusively, and subject to alterations. Prices incl. VAT. Books excl. shipping. Journals incl. shipping. 0238100006_dp

When opening RF-TENDON Design, all data determined so far is imported. In addition to the prestressed concrete reinforcement already defined, one can specify non-prestressed longitudi-

Tel. +49 (0)6201 606-400 Fax +49 (0)6201 606-184 service@wiley-vch.de

C o m p a n y

Responsible for “Products & Projects” (sponsored content): Ernst & Sohn

Structural Concrete 12 (2011), No. 4

Products & Projects

Creative designs for floating cupola In France, numerous major buildings are located on the Place de la République. In Strasbourg, one of these is very large and circular in shape. It is surrounded by the Prefecture of the Département, the Ministry of Culture and Communication, and the Minstry of Finance – among other buildings, including the Bibliothèque Nationale et Universitaire. In the Franco-German War (1870–1871), the Dominikanerkirche church in Strasbourg was destroyed by Prussian artillery fire – along with its side wing which housed the metropolitan library and metropolitan archives; as a result, many irreplaceable historical documents were lost. On this site, the New Church (Temple Neuf) was erected. By contrast, the new library was order built in an area which at the time was called the Kaiserplatz. According to drafts by the architects Skjold Neckelmann and August Hartel, it was built between 1889 and 1894, in the style of historicism. It had two simultaneous functions: as a national library and a university library. Over time, it became the secondlargest library in France. However, the structural over 100 years old substance of this building which stands under monument conservation was increasingly unable to meet the modern requirements in terms of structural engineering, fireproofing etc. So it was decided to implement a complete refurbishment in which the exterior front, the roof and numerous inner structures were to remain intact, and this circumstance significantly complicated this undertaking. A total construction cost of € 61 million was proposed.

Construction work on existing buildings The building complex should be gutted to the greatest possible extent. In the process, no damage may occur to the fronts. This means that any access by machines, people and material must indeed be made by way of the existing access routes – which are, in principle, not designed for this type of access. The consortium consisting of Urban BTP (Illkirch) und Demathieu & Bard (branch Duppigheim) then addressed this task. Both companies were already founded in the 19th century, and are among the major construction companies in France; Urban is affiliated with the Vinci Group (over 160,000 employees). In addition, the very tight available space on the premises as a whole further complicate the matter, considering that the premises are just a few metres wider than the building itself. This space is largely taken up by construction containers as well as the construction vehicles carrying materials to and from the site. The surrounding streets were not to be blocked and traffic not impeded. This requires the particularly painstaking planning of logistics and deliveries whenever they are needed – and at that precise moment. A large crane which can transport even heavy loads to far-away locations then ensures a constant flow of replacement supplies.

Fig. 1. On the popular Place de la République stands the great building that is the Bibliothèque Nationale et Universitaire de Strasbourg.

– It is impossible for a crane to extend through the cupola. Nevertheless, heavy-duty concrete work must be performed up to a height of approx. 24 m. So, under these circumstances, what is the best course of action for maintaining the greatest possible level of efficiency? 1. The structurally-ineffective building components were removed, and a sub-structure (approx. 24 m high) was built from four towers to absorb the load of the cupola. After that, some space could be created. 2. On the four corners of the middle structure (at the greatest possible height – that is to say, below the cupola’s support structure and thereby in a less-than-ideal position), one pivot arm each of a hoist crane is installed. The maximal load is a mere 1.5 t. 3. A tetragonal binding beam made of in-situ concrete was constructed on which the cupola rests. The total weight of this structure alone amounts to more than 600 tonnes. It requires two weeks for the concrete to achieve its final load capacity. 4. On the corners which have remained clear, L-shaped concrete columns with a 32-mm concrete reinforcement rise to face the binding beam. These should ultimately carry the total load. Between the columns and the supporting walls, there is a spacing of merely 1.5 m. 5. As soon as the structural/design-engineering connection to the binding beam is successfully completed, the supporting scaffoldings will be removed. With that, the actual interior work can start.

The floating cupola The most difficult part of the renovation involves the gutting and support of the middle structure – crowned by the cupola which can be seen from afar. Here, too, access from above is impossible. The cupola should fill the inside with sunlight – all the way down to the basement floor. However, without sufficient structural support, the cupola threatened to topple into the interior! All work takes place under extremely difficult conditions which make typical construction work virtually impossible: – The available space is so limited that typical working-platforms cannot be applied to the formwork. Nevertheless, French law mandates safety installations which far surpass the given safety requirements in comparison to Germany.

A10 Structural Concrete 12 (2011), No. 4

Fig. 2. The sides of the cupola’s supporting structure are supported during the construction work by scaffoldings measuring up to heights of 23.8 m. In the corners, L-shaped concrete pillars rise upwards.

Responsible for “Products & Projects” (sponsored content): Ernst & Sohn

Products & Projects

Fig. 3. The cupola rests on a new, circumferential concrete beam with the low edge at a height of 23.80 m, supported on scaffolding towers.

Concrete and formwork works The performance of concrete work proved to be very highly sophisticated under these circumstances. In all of the planning and tender phases, those who considered the project practically impossible or unrealistically expensive in its intended form “jumped ship”. Ultimately, PASCHAL proved to be the only service providers who had a firm and durable concept. 1. The binding beam is 0.80 m wide, 1.10 m high and squareshaped (14.58 m on each side). In this context, the edges of the lengthwise sides are 1.45 m longer than those on the transverse sides, which gives the structure an approximate “H” shape. The formwork is built with horizontally-positioned elements of the LOGO.alu formwork system (up to 2.70 m long) which is very simple and easy to handle, due to its low weight – this is a necessary prerequisite in the tight confines of the dome. The LOGO.alu can be adjusted to the required length by 1-cm intervals – this renders the nuisance residual by site fillers unnecessary. As supporting system PASCHAL Deck was used. 2. As a formwork system for the L-shaped columns (3.80 × 3.00 m and 0.60 m thick), the LOGO.3 was selected. The LOGO.3 absorbs 70 kN/m2 of fresh-concrete pressure according to DIN 18218 – with its profiled flat-steel frame particularly suited to narrow conditions. One height cycle was 2.94 m high; there were seven of these, and the eighth and/ or the subsequent cycle was planned once again with the LOGO.alu. From the fourth height work cycle, the widths are a mere 3.00 × 3.00 m, which was also readily apparent in the planning process, which had also made its mark on may aspects of the planning – since tie points, element dimensions and work scaffoldings must be pre-set accordingly. The LOGO.3 was used as a climbing formwork with a suspended scaffold. A work cycle was done every two days. In total, two formwork sets were applied (each to diagonally-opposed columns). 3. There is no such thing as a serial work scaffolding for formwork which fits between the narrow walls and also fulfils the highest of safety requirements. So the PASCHAL structural Design Division developed two completely new climbing systems (one each for the front side and the rear side), which only have 1.05 m wide brackets (in which attached ladders are integrated) Ultimately, the construction oversight authorities were not the only ones enthused by this – it was even possible for the outside to (despite the rather modest performance of the small crane) to shift the formwork along with the work scaffolding, the climbing unit and a cane set – which in turn increased the level of efficiency even more.

Responsible for “Products & Projects” (sponsored content): Ernst & Sohn

Fig 4. The most sophisticated construction work is taking place below this cupola. In summary, the entire structure weighs 600 tonnes. (© PASCHAL)

The end of this formwork “detective story” Christian Bordier, the formwork consultant from PASCHAL, is absolutely thrilled. As he says, a construction site – particularly one with these requirements – is anything but child’s play, and there were constant modifications to be made, as well as a need for co-ordination, the occurrence of unforeseen difficulties – and even some smaller setbacks. “That is definitely a part of this process,” he says. “It always happens this way when one dares to try something new or difficult. “ No rewards without effort! As he maintains: “So little formwork, so much to consider – that’s good for your head!“ So this is how the result – which he describes as “art” – turned out, perfectly, only due to the co-ordination between all parties involved. Further information: PASCHAL-Werk G. Maier GmbH, Kreuzbühlstraße 5, 77790 Steinach, Tel. +49 (0) 78 32/71-0, Fax +49 (0) 78 32/71-209, service@paschal.de, www.paschal.de

HeidelbergCement opens new cement mill in Bangladesh Beginning of January HeidelbergCement officially inaugurated a new cement mill at its plant in the seaport Chittagong. The ball mill has a grinding capacity of about 0.8 million tonnes. The investment costs amount to approximately US$16 million. “We are very pleased that we are able to officially inaugurate our state-of-the-art cement mill today,” says Dr. Bernd Scheifele, Chairman of the Managing Board. “Bangladesh is an interesting market for HeidelbergCement. We expect the need for high quality cement to increase significantly in the coming years, especially due to new government infrastructure projects. The investment in Bangladesh is part of our long-term strategy to expand our cement capacities in attractive emerging markets by brownfield or greenfield projects.” The IMF forecasts a GDP growth of 6.1% for Bangladesh in 2012. Bangladesh has one of the lowest per capita cement consumption ratios in the world, but it is also one of the fastest growing markets worldwide. HeidelbergCement has been active in Bangladesh since 1998 and is one of the largest German investors in the country. Further Information: HeidelbergCement AG, Berliner Strasse 6, 69120 Heidelberg, Germany, Phone: +49 6221/481-0, Fax: +49 6221/481-13 217, info@heidelbergcement.com, www.heidelbergcement.com

Structural Concrete 12 (2011), No. 4

A11

Provider directory products & services

anchor channels

Deutsche Kahneisen Gesellschaft mbH Nobelstraße 51 D-12057 Berlin Tel. (0 30) 6 82 83-02 Fax (0 30) 6 82 83-4 97 e-Mail: info@jordahl.de Internet: www.jordahl.de JORDAHL® Ankerschienen, JORDAHL® Schrauben, Mauerwerksabfangungen, Trapezblechbefestigungen, Bewehrungstechnik, Durchstanzbewehrungen, Schubdorne

fastening technology

HALFEN Vertriebsgesellschaft mbH Katzbergstraße 3 D-40764 Langenfeld Phone +49 (0) 21 73 9 70-0 Fax +49 (0) 21 73 9 70-2 25 Mail: info@halfen.de Web: www.halfen.de concrete: fixing systems facade: fastening technology framing systems: products and systems

post-tensioning

reinforcement technologies

bridge accessories

Maurer Söhne GmbH & Co. KG Frankfurter Ring 193 D-80807 München Phone +49(0)89 32394-341 Fax +49(0)89 32394-306 Mail: ba@maurer-soehne.de Web: www.maurer-soehne.de Structural Protection Systems Expansion Joints Structural Bearings Seismic Devices Vibration Absorbers

literature

Ernst & Sohn Verlag für Architektur und technische Wissenschaften GmbH & Co. KG Rotherstraße 21 10245 Berlin Phone +49 (0) 30 4 70 31-2 00 Fax +49 (0) 30 4 70 31-2 70 E-mail: info@ernst-und-sohn.de Web: www.ernst-und-sohn.de

A12

Structural Concrete 13 (2012), No. 1

DYWIDAG-Systems International GmbH Max-Planck-Ring 1 40764 Langenfeld/Germany Phone +49 (0)21 73/7 90 20 Mail: dsihv@dywidag-systems.com Web: www.dywidag-systems.de

prestressed concrete

Max Frank GmbH & Co. KG Technologies for the construction industry Mitterweg 1 94339 Leiblfing Germany Phone +49 (0)94 27/1 89-0 Fax +49 (0)94 27/15 88 Mail: info@maxfrank.com Web: www.maxfrank.com

sealing technologies Paul Maschinenfabrik GmbH & Co. KG Max-Paul-Straße 1 88525 Dürmentingen/Germany Phone +49 (0)73 71/5 00-0 Fax +49 (0)73 71/5 00-1 11 Mail: stressing@paul.eu Web: www.paul.eu

software

DICAD Systeme GmbH CAD-Systems for reinforced concrete datailling Theodor Heuss Straße 92–100 D-51149 Köln Tel.: +49 (0) 22 03/93 13-0 info@dicad.de www.dicad.de

Ing.-Software DLUBAL GmbH Am Zellweg 2 93464 Tiefenbach Phone +49 (0) 96 73 92 03-0 Fax +49 (0) 96 73 92 03-51 Mail: info@dlubal.com Web: www.dlubal.de

stay cables

DYWIDAG-Systems International GmbH Max-Planck-Ring 1 40764 Langenfeld/Germany Phone +49 (0)21 73/7 90 20 Mail: dsihv@dywidag-systems.com Web: www.dywidag-systems.de

vibration isolation Max Frank GmbH & Co. KG Technologies for the construction industry Mitterweg 1 94339 Leiblfing Germany Phone +49 (0)94 27/1 89-0 Fax +49 (0)94 27/15 88 Mail: info@maxfrank.com Web: www.maxfrank.com

BSW GmbH Am Hilgenacker 24 D-57319 Bad Berleburg Phone +49(0)2751 803-126 Mail: info@berleburger.de Web: www.bsw-vibration-technology.com under-screed impact sound insulation with European Technical Approval, PUR foam & PUR rubber materials for vibration isolation

Editorial

Sustainable concrete is more than just durable concrete Sustainability is a sustainable word in the 21st century. Despite the fact that the media focus on a host of other world news items, sustainability issues such as greenhouse gases and global warming always seem to form the backdrop. Inevitably, concrete has the potential to be a sustainable material. The concrete structures of the Roman Empire, the zenith being the Pantheon in Rome, are obvious evidence of this statement. The fundamental meaning of sustainable is durable, and by surviving 20 centuries, concrete structures such as the Roman ones have shown that they are, indeed, durable. Today, we use the words sustainable and sustainability in a more abstract way. We need to consider ecological, economical and societal aspects. The global cement industry accounts for 5 % of CO2 emissions. We cannot just rely on the long service life that can be achieved by suitable concrete mix design, structural design and detailing, construction methods, use and maintenance. There are several tools we can employ to limit CO2 emissions: – – – – – – – –

Johan Silfwerbrand

Effective cement production and alternative fuels Alternative binders Optimized cement mixes Optimized cross-sections Optimized reinforcement Strengthening concrete’s advantages during the serviceability state Prolonging the service life Reuse and recycling

While cement producers have devoted considerable effort to the first point, most material researchers have concentrated on the second one. Recently, we have seen a boom in research on energy storage capacity, fire resistance, sound insulation, albedo, fuel consumption on concrete pavements and NOx reduction, i.e. properties where concrete has advantages over other materials. Durability issues have been an important research field since the 1980s, and optimizing the cross-section and the reinforcement even longer, despite the fact that optimization during the 1960s and 1970s was overshadowed by issues like constructability and productivity. There seems to be an increased interest in reuse and recycling. Very recently, issues such as carbon dioxide uptake through normal carbonization and Carbon Capture and Storage (CCS) have changed people’s thinking. Papers dealing with sustainability also feature regularly in “Structural Concrete”. This issue continues that trend: – Two papers on assessments conducted after the Christchurch earthquake in New Zealand one year ago, one dealing with façade systems, one with concrete bridges. – A paper on determining the in situ concrete compressive strength by using EN 13791.

Editorial – A paper on punching shear and the new design concepts in Model Code 2010. – One paper on full-scale tests on the ultimate strength of curved strand tendons in large concrete elements. – One paper on the development of design methods for the formwork pressure of highly flowable vibrated concrete and self-compacting concrete. The aim of assessment is to enable the continued use of a structure and, hence, the sustainability aspect is obvious in the three first papers. Fifteen years after its first appearance, the use of self-compacting concrete has still not achieved its full potential in in-situ concreting, despite its great working environment benefits. An improved working environment is an important part of the societal dimension of sustainability. The last three papers deal with structural design and detailing. Here, the sustainability aspect is less directly evident. However, the long-term goal of structural design and detailing is improved performance, improved safety or optimization, so at least there are economical aspects. If you have read so far, I am convinced that you are interested in sustainability and sustainable concrete structures. “Concrete Structures for Sustainable Community” is the title of the fib Symposium to be held in Stockholm in June of this year. I hope you will be able to join me at a symposium that will present some 170 papers on these very important issues.

Johan Silfwerbrand Chair of the Organizing Committee of fib Symposium 2012

Structural Concrete 13 (2012), No. 1

Articles Andrew Baird* Alessandro Palermo Stefano Pampanin

DOI: 10.1002/suco.201100040

Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake The magnitude 6.3 earthquake that struck Christchurch on 22 February 2011 caused widespread damage throughout Christchurch’s central business district (CBD), where a large proportion of the building stock consists of reinforced concrete (RC) buildings. Damage to the façades of these buildings was a clear contributor to overall building damage. This paper presents the damage assessment of the façade systems of these RC buildings. A survey of 173 RC buildings in the Christchurch CBD is conducted here, focusing on the damage to the façade systems of the buildings. The survey covers only buildings greater than three storeys in height, excluding the majority of unreinforced masonry façades, the damage to which has been well documented. The façade type and modularity is classified for each system, as well as the connection type where possible. The level of damage to each façade is determined in terms of the following performance levels: Operational, Immediate Occupancy, Life Safety and High Hazard. Further investigation is also made into three precast concrete panel systems. These case studies examine the damage, detailing and construction practice of each particular system. Keywords: facade, concrete panels, earthquake damage, performance levels, Christchurch earthquake

Introduction

Some 185 people lost their lives as a result of the earthquake that struck Christchurch on 22 February – the second largest toll from a natural disaster in New Zealand [1]. The proximity of the earthquake’s epicentre to Christchurch – approx. 10 km from the city at a shallow depth of 5 km – resulted in very strong ground shaking. The maximum felt intensity was MM IX and the maximum recorded peak ground acceleration (PGA) was 2.2g. The PGA recorded within the Christchurch Central Business District (CBD) ranged from 0.6g to 0.8g [2]. For further details on the performance of concrete structures and the level of ground motion within the CBD please refer to a companion paper in the journal Structural Concrete 4/2011 [3]. The earthquake caused widespread failure of older unreinforced masonry (URM) structures as well as the fail-

ure of two RC buildings. RC systems are the predominant multi-storey building type found in the Christchurch CBD. They have been widely used for commercial buildings in New Zealand since the 1931 Napier earthquake, which damaged a significant number of URM buildings [4]. Many buildings within the Christchurch CBD withstood the effects of the earthquake from a structural perspective but are considered unusable because of damage to façades, ceilings, partitions and contents. Current seismic design provisions typically require that non-structural components be secured so that they do not present a falling hazard; however, these components can still be so severely damaged that they cannot function [5]. It is not only damage to the façade that can cause a building to be unusable; there is also the risk of injury or death from falling panels, masonry, glass etc. It is also clear that façade systems are particularly vulnerable to earthquakes since new and continuing damage to façade systems has been observed throughout Christchurch in recent aftershock events. This paper presents the damage assessment overview of the façade systems of 173 RC buildings in the Christchurch CBD. Only buildings taller than three storeys in height were surveyed in order to exclude the majority of URM façades and to restrict the survey population. For buildings with multiple façade systems, multiple assessments are conducted. In total, 298 façade systems are surveyed. The survey is based on what is visible from outside the building, making it equivalent to a Level 1, or rapid safety, assessment [6]. Therefore, it was not possible to assess aspects such as the status of the connections or whether windows were jammed. The consequence of this is that the results of the survey will be conservative because less obvious forms of damage certainly exist. Only with a more detailed survey can the true extent of the damage be determined. Three case studies of damage to precast concrete panel systems are also investigated in further detail. Two of the case studies exhibit severe damage, not typical of the average performance. However, they are presented so that similar failures can hopefully be avoided in the future.

* Corresponding author: andrew.baird@pg.canterbury.ac.nz

New Zealand façade technology

Submitted for review: 14 August 2011 Revised: 29 August 2011 Accepted for publication: 7 September 2011

Façade systems can be grouped into two main types: claddings and infills. There is also the possibility of a

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

combination of the two, termed a mix system. Infills are constructed within the frame of the structure, whereas claddings are attached externally to the primary structure [7].

2.1

Claddings

Claddings often incorporate stiff, brittle materials such as glass, concrete and stone. Precast concrete panels have been the most popular cladding material used in new non-residential buildings in New Zealand over the past decade [8]. Panels of autoclaved lightweight concrete (ALC, also called autoclaved aerated concrete) feature on several buildings within the Christchurch CBD. This is also among the most widely used material for claddings in Japan [9]. Cladding connections can have numerous configurations; however, they are typically located on either the beams or columns. The generic connection method for heavy cladding consists of a bearing and a tie-back connection. The fixed bearing connections support the gravity loads of the cladding, whereas the ductile tie-back connections allow relative movement between cladding and structure. Tie-back connections must also be capable of accommodating the out-of-plane forces on the panel, including wind. Lightweight claddings are generally fixed to the structure with connections that do not allow movement, hence must be able to accommodate relative displacement within the system. Stick systems are a popular lightweight option in modern multi-storey buildings. The stick system consists of extruded aluminium frames holding panes of glass. A rubber seal is used to allow the glass within the frame to move while keeping the building weathertight. One of the more recent variations of the stick system is the double-skin façade system. The double skin consists of two layers of façade material (typically glass) that creates a sealed cavity to improve the thermal performance of the building. Double-skin façade systems are being increasingly employed in high-profile buildings, being touted as an exemplary “green” building strategy.

2.2

Infills

Infills have traditionally been made of heavy rigid materials, such as clay bricks or concrete masonry blocks. However, more lightweight infill panel options such as light steel- or timber-framed infill walls (dry walls) are available. Masonry infill construction has a long history in much of Europe and South America, but has been avoided in New Zealand for several decades, primarily because of concerns about its poor seismic performance and the complexity of its interaction with the structure [10]. It is typical for an infill panel to be combined with a glazing infill system. Glazing infill consists of an aluminium frame attached directly to the infill panel or structure. The frame has rubber gaskets to hold the panes of glass in place and keep the system watertight while allowing some in-plane movement. This type of system is simple to construct and is particularly prevalent in low- to medium-rise office structures. Infill panels are often clad both externally and internally to enhance thermal performance and improve aesthetics.

Structural Concrete 13 (2012), No. 1

2.3

Design standards

New Zealand design standards specify serviceability limit state (SLS) criteria for earthquakes in the form of deflection limits. These deflection limits are related to earthquake actions with an annual probability of being exceeded of 1/25 [11]. There is also an ultimate limit state (ULS) requirement that the façade should remain supported and not interfere with evacuation in a design level earthquake. Façade damage should be expected in an ULS event according to current design standards. This is because the SLS limits define deflections beyond which repairs can be expected. However, the damage should not be life-threatening.

Façade performance levels

The façade performance levels (or damage states) suggested by FEMA are the following: Operational, Immediate Occupancy, Life Safety and Hazards Reduced [5]. One of the problems with using these performance levels as a means of assessing damage is that they are intended for use in design. In particular, the Hazards Reduced level is aimed at preventing serious injury caused by large or heavy items falling. However, not all façades surveyed met this design criterion. In order to avoid confusion, the Hazards Reduced performance level is herein renamed the “High Hazard” performance level in order to include accurately any cases where there was a high risk of serious injury or fatality from façade damage. Fig. 1 shows photographs and graphic illustrations of the different façade performance levels sustained during the Christchurch earthquake. The basic requirements for setting façade performance objective levels are relatively simple. For example, the basic performance objective would be that a façade remains undamaged following frequent earthquakes and that it does not fail in large (very rare) earthquakes. However, this objective level means that the façade may be damaged to some degree in occasional earthquakes. Definitions of the performance levels that were used in the survey are described below and are based on those suggested by FEMA 356 [5]. It is important to distinguish that the level of structural and non-structural damage can be different and hence the structural and non-structural performance levels are not necessarily the same. It is generally expected that the damage level of the non-structural components will be worse than the damage level of the structure. Fig. 2 shows the performance-based design matrix that combines both structural and non-structural performance levels. A target building performance level consists of selecting a structural performance level and a nonstructural performance level [5]. The four squares highlighted represent the four target building performance levels suggested by FEMA 356 [5]. The four façade performance levels used in the assessment are subsequently explained.

3.1

Operational performance level

The façade is able to support its pre-earthquake functions, although minor clean-up and repair may be required.

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

Fig. 1. Photographs and graphic illustrations of performance levels (from left): Operational, Immediate Occupancy, Life Safety, High Hazard

3.2

Immediate Occupancy performance level

Damage to the façade is present but building access and life safety systems remain available and operable. Minor window breakage could occur. Presuming the building is structurally safe, occupants could safely remain in the building, although normal use may be impaired and some clean-up required. The risk of life-threatening injury due to façade damage is very low.

3.3

Life Safety performance level

Damage to the façade is present but the damage is not lifethreatening. Potentially significant and costly damage has occurred to the façade but the majority of the system has not become dislodged and fallen, threatening life safety ei-

ther inside or outside the building. Egress routes within the building are not extensively blocked, but may be impaired by lightweight debris. Whereas injuries may occur during the earthquake from the failure of façade components, overall, the risk of life-threatening injury is very low. Restoration of the façade may require extensive work.

3.4

High Hazard performance level

Damage to the façade is present, creating multiple falling hazards. Extensive damage has occurred to the façade with the potential of a serious threat to life safety outside the building. Widespread window breakage is likely and disconnection of components of the façade system from the structure is possible. Restoration of the façade is probably only possible by replacing the system completely.

Fig. 2. Post-earthquake building performance level, taken from FEMA 356 (left and centre) [5], graphical illustration of building performance levels reproduced from FEMA 389 (right) [12]

Structural Concrete 13 (2012), No. 1

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

Building survey

The building survey was conducted within the four avenues (Bealey, Deans, Moorhouse and Fitzgerald) that border the Christchurch CBD. In total, 173 RC buildings were surveyed. The focus was solely on RC building types, which make up 80 % of buildings taller than three storeys in the Christchurch CBD. These can be broken down by their primary structural system: concrete frame with reinforced concrete walls, concrete frame with concrete masonry walls and moment-resisting concrete frame. Fig. 3 presents the information regarding the construction information of the 173 buildings. After the 22 February earthquake, all buildings were inspected and given either a green, yellow or red placard to indicate the safety of the building. A green placard meant that a building had been assessed and no apparent structural or other safety hazards were found. A yellow placard meant that a building had restricted access and a red placard meant a building must not be entered because it was deemed unsafe [6]. Some 79 % of the buildings in the survey were given either a yellow or red placard.

The predominant occupancy type of the buildings is office use, accounting for 69 %, and the majority are lowto medium-rise buildings. Building age was estimated at the time of survey or found from city records following further investigations. Following a large boom in construction after the 1960s, the majority of buildings are less than 50 years old.

Façade damage

A total of 298 façade systems were surveyed on the 173 buildings. A maximum of two façade systems were surveyed per building and a façade system was only surveyed if it occupied at least 10 % of the building’s surface area. The survey classified the façade systems by 11 individual typologies based on those used on the Post-earthquake Building Performance Assessment Form [13]. These typologies were grouped by the sub-categories: heavy cladding, lightweight cladding and infill, as outlined in Fig. 3. The damage sustained has been organized by these sub-categories. Heavy claddings can be defined as having a mass > 80 kg/m2, whereas a lightweight cladding is < 80 kg/m2 [14].

Fig. 3. Building construction information (clockwise from top left): building location and placard composition, RC construction type, occupancy type, building construction age, building height

Structural Concrete 13 (2012), No. 1

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

Fig. 4. Façade information: facade typology groups (top left), composition of facade typologies surveyed (bottom left), overall façade performance (right)

The age of the façade in relation to the building was recorded. Some 96 % of façade systems appeared to be the same age as the building, with the remaining systems having been retrofitted. The performance level of each façade system was determined according to the criteria discussed above. Overall façade performance can be seen in Fig. 4 and shows that 63 % of façade systems surveyed were deemed Operational. This is an excellent result, considering that 79 % of RC buildings had either a yellow or red placard. However, it should be remembered that the survey is based on what is visible from outside the building and less obvious forms of damage certainly exist. There were also 31 façade systems (10 %) that were deemed High Hazard, falling outside the objectives of the design standards.

5.1

Heavy cladding

The majority of heavy claddings surveyed were precast concrete panel systems. Precast panels can be either storey-height panels that provide multiple architectural functions or panels that are purely aesthetic. The function of spandrel panels, for example, is typically only to hide RC members from view. The proportion of storey-height panels and aesthetic panels in the survey was approximately equal. Storey-height heavy panels commonly have openings for windows. The window system inside the panels could have been classified as a glass infill; however, for this survey they have been included as part of the panel system. This decision was taken because the surrounding panels have such a high in-plane stiffness that allowance for movement is not required for these window systems.

Heavy cladding damage covered the range of damage states. The majority of heavy claddings were deemed Operational because they exhibited little to no evidence of damage, as shown in Fig. 5. Complete disconnection of panels was also observed as well as cracking and spalling of panels. The performance level Life Safety was used for claddings that were clearly damaged, e.g. concrete corners or similar small amounts of concrete missing, as shown by two of the photographs in Fig. 5. Only one case was deemed High Hazard and was the result of several spandrel panels shearing off their bolted connections and falling onto the sidewalk below. Fortunately, no one was killed by these falling panels; however, there was the risk of multiple fatalities as the heavy panels fell on approx. 20 m of sidewalk. This building is covered in detail later as a case study. Minor damage was also observed in the form of panels having residual displacements and/or rotations. The ejection of sealing joints between panels was also common. The performance level Immediate Occupancy was used for claddings that showed evidence of cracking or where it was clear that the panels had residual displacements and/or rotations. It could be concluded that heavy claddings performed better than most façade systems, with 94 % of heavy claddings deemed either Operational or Immediate Occupancy. However, it is possible that a more thorough assessment of the connections from inside the buildings may lower this percentage. More importantly, the possible consequence of heavy claddings falling is severe, which means their treatment requires further attention. Complete disconnection of large concrete panels was also observed in the magnitude 6.3 aftershock on 13 June 2011. However, these were attached to a two-

Structural Concrete 13 (2012), No. 1

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

Fig. 5. Heavy cladding performance (clockwise from top left): composition of performance levels, corner damage to spandrel panel, structure showing six missing spandrel panels, corner damage to storey-height panels

storey building so are not included in the survey. This damage is covered in detail later as a case study.

5.2

Lightweight cladding

Lightweight cladding includes a broad range of façade systems, and each typology of lightweight cladding can also include a large range of systems. For example, the curtain wall typology includes endless arrangements of extruded aluminium members infilled with glass or lightweight panels. Often, lightweight cladding incorporates a large amount of glazing. They can therefore appear a lot more lightweight than they in fact are, with some systems (such as the double skin) containing a substantial amount of weight. The composition of performance levels for lightweight claddings can be seen in Fig. 6. A large number of High Hazard cases were observed. This was usually due to a significant portion of the glazing falling from the system. Overall, 82 % of lightweight claddings were deemed either Operational or Immediate Occupancy, exhibiting either no or very minor damage such as ejected window seals or cracked glass. Lightweight claddings of all ages showed various levels of damage. Newer systems were proportionately less likely to exhibit moderate to severe damage. However, issues do still exist with current design and construction techniques since several lightweight cladding systems less

Structural Concrete 13 (2012), No. 1

than 20 years old were heavily damaged. One system installed only 12 months prior to the earthquake was very badly damaged. One of the worst-performing lightweight cladding systems was a curtain wall no more than about 15 years old. Multiple sections of the curtain wall had become completely detached from the building, with the aluminium frame and glazing falling onto the sidewalk, as shown in Fig. 6. Closer inspection showed that the aluminium frame was screwed to a wooden subframe and the failure was a result of the screws either shearing off or tearing out of the wood. The glass damage was recorded for all lightweight cladding that contained glass. The damage was categorized according to FEMA 356 [5]. These can be simplified to the following: the glass is unbroken, the glass shatters but remains in the frame, the glass falls out of the frame. The difference in damage levels corresponds to different levels of risk to pedestrians. As can be seen in Fig. 6, nearly half of all glazed lightweight claddings had glazing damage and 39 % presented a falling hazard. Inadequate allowance for movement of the glass panes was the likely cause for the majority of glazing damage. Several buildings with older, non-seismic glazing frames were re-glazed between September and February, only to be damaged again in the February earthquake. Considerable heavy damage was observed in “spider glazing”, as can be seen in Fig. 7. Spider glazing is a rea-

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

Fig. 6. Lightweight cladding performance (clockwise from top left): composition of performance levels, disconnection of curtain wall frame, heavy glazing damage, composition of glazing damage, heavy glazing damage

sonably modern system so we might expect it to perform better than other systems; however, this was not the case. It appeared that damage originated around the “spider” that holds each glass pane, probably as a result of the spider creating stress concentrations in these regions due to the restraint of the connection to the structure.

5.3

Infill

Infill systems include masonry and glazing systems that are located within the frame of the structure. Infill façades performed the worst among the façade groups, as can be seen in Fig. 8. Only 60 % of infill systems were deemed either Operational or Immediate Occupancy, 17 % were deemed High Hazard, the highest of the façade groups. Older glazing infill systems were particularly susceptible to damage. These systems typically consist of highly modulated glazing frames that do not contain any in-plane

movement allowance apart from the small gaps around each pane of glass. These gaps are typically only a few millimetres wide and consequently only allow a minimal amount of in-plane drift before the glass begins to carry force. Once this occurs, the stiff, brittle glass is at high risk of cracking and being dislodged from the frame. The glass damage was recorded for all glazing infill systems. As can be seen in Fig. 8, over half of all infill glazing systems had glazing damage that resulted in the glass falling out of the frame. Typically, modern glazing infill performed well and did not exhibit any breakages. However, since the survey was visual only, it is possible further damage exists to the façade systems which is not clearly visible. For example, many residential homes exhibited warping of their glazing frames without any cracks forming in the glass. This warping made opening windows and doors impossible in some instances. Therefore, it is possible that some glazing infill cases were also distorted.

Fig. 7. Spider glazing damage: cracked glass concentrated around spider (left), fallen glass panes from external lift shaft (centre), fallen glass panes of newly constructed system (right)

Structural Concrete 13 (2012), No. 1

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

Fig. 8. Infill performance (clockwise from top left): composition of performance levels, failed infill panel, heavy glazing infill damage, short column effect due to infill, composition of glazing damage, heavy masonry infill damage to St Elmo Courts (–43.532, 172.631)

The vulnerability of masonry infill was clearly showcased by the damage sustained by the eight-storey St. Elmo Courts building (−43.532, 172.631), pictured in Fig. 8. The collapse hazard of this building resulted in surrounding buildings and streets being completely off-limits for several weeks. This building has now been demolished. Other URM infill cases also showed significant damage. Reinforced masonry infill did not typically show much damage other than small cracks. However, it was evident that the infill had an effect on the seismic performance of the primary structure [15], as can be seen in Fig. 8, where the infill had a short column effect, causing shear cracking in the column.

Precast concrete case studies

Three case studies are presented here to illustrate the performance of precast concrete cladding panels. Of the three cases, two experienced failure and one just minor damage. The case studies examine the detailing of each system, the damage sustained and any obvious issues arising. Case study 1: flat spandrel panels This was the only survey case in the Christchurch CBD that was categorized as High Hazard. Six panels fell from a height of approx. 10 m onto the sidewalk below. The panels were between 7 and 10 m long and weighed between 1.3 and 1.9 t. The panels had been attached to a car park (−43.534, 172.635) and therefore their function was to provide an aesthetic finish as well as prevent people from walking or driving off the edge of the structure. Each panel was attached to the structure by four bolted connections. These connections were located on the side of the column above the beam as can be seen in Fig. 9. Even

Structural Concrete 13 (2012), No. 1

though they were attached to the column and not the beam, the panels acted as beam spandrel panels. The connections consisted of metal angle sections welded to cast-in plates in the columns. All but two of the 24 connections remained attached to the structure. Four of the remaining connections had also been partially torn out from the column. The panels were attached to the angle by a bolt that was fixed to the panel by a cast-in anchor. Horizontal slots were present in all metal angles to allow the bolt to slide, as shown in Fig. 9. Upon inspection, many of these bolts had failed, with some bolt heads still remaining in the metal angle. The slotted connections should theoretically have prevented large in-plane forces being carried in the panels. This is because slotted connections allow relative movement between the structure and the panels. However, on closer inspection it was observed that the bolt heads had not been able to move along the slots because their washers had been welded to the metal angle, as shown in Fig. 9. With the connections essentially fixed, the high in-plane stiffness of the panels would have resulted in significant forces being transferred through the panels under in-plane deformation of the structure. The bolts would have been required to transfer these forces and it is highly likely that they were unable to and consequently failed in shear. It is also possible that the failure could have been an out-of-plane failure, due to the significant accelerations experienced during the earthquake. The large mass of the panels would have induced significant inertial forces, possibly causing the bolts to fail in tension. This would also explain why four of the metal angles had been bent perpendicularly to the frame. However, the plastic hinges evident in many of the beams indicate that the structure underwent significant in-plane deformation. This suggests that shear failure of the bolts is the most likely reason for failure.

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

Fig. 9. Flat spandrel precast panel case study (–43.534, 172.635): structure and location of panels (left), slotted connections and failed bolts (centre), bolt and washer still in connection showing no movement (right)

Case study 2: storey-height coffered panels The magnitude 6.3 aftershock on 13 June 2011 caused the complete disconnection of six storey-height precast concrete panels as shown in Fig. 10. The building is located outside the CBD in a western suburb (−43.522, 172.582). A magnitude 5.6 aftershock, which led to the evacuation of the area, occurred 80 minutes prior to the magnitude 6.3 event, so luckily no one was injured. The panels were 3.2 m wide, 2.6 m high and weighed approx. 2.4 t each. They had small central window openings and were coffered with a 160 mm deep by 100 mm wide section protruding out around the back edge of the panel. The panel connections were vertical bolts through the coffered section. The coffered section also allowed the panel to be seated on the primary structure. The weight of the panel was transferred through this bearing connection, which was fixed with two vertical 20 mm bolts. The panels were secured at the top with two vertical 16 mm bolts. These bolts connected the top of the panel to the structure via a metal angle. The top connections did not appear to include any allowance for movement because the metal angle appeared to be flush with the remaining panels. Each panel was also connected to adjacent panels with five 16 mm bolts. A small gap is present between panels but the connection between the panels is clearly very strong. As such, the panels effectively behave as one long panel with numerous top and bottom connections. The fallen panels showed evidence of bolt tension failure and concrete anchorage failure, according to how the top bolt was configured. The 16 mm top bolts were either screwed into a cast-in socket that was placed in the edge of the coffered region, or they passed through the coffered region and were fixed with a nut and washer. All of the cast-in sockets remained in the panels. This was probably because they were located close to the mesh reinforcement of the panels. The bolts that were screwed into the cast-in sockets all appeared to have failed in tension, a fact revealed by the ends of the bolts left hanging

from the metal angle, as shown in Fig. 10. Some degree of concrete failure is also evident where the bolts pulled out from the cast-in sockets. The bolts that were fastened with a nut after passing through the coffered section all still had their nuts attached. The corresponding locations on the panels had large sections of concrete missing where the bolts passed through, clear evidence of a wedge-type concrete failure. The top bolts were located very close to the corners of each panel, heightening the risk of a wedge failure of the concrete. There was also no reinforcing steel evident in the coffered region where the bolts passed through. The bottom connections were galvanized metal sleeves that allowed 20 mm bolts to pass through. Some of

Fig. 10. Storey-height coffered precast panel case study (–43.522, 172.582) (clockwise from top left): failed connection and imminent failure, fallen panels during removal, un-grouted bearing connections, damage to concrete below panels that had not fallen

Structural Concrete 13 (2012), No. 1

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

these bolts had been grouted but some had not, like that shown in Fig. 10. It is not clear whether this had an effect on the performance but it does bring into question the quality of workmanship and checks by the contractor during installation. Reinforcing steel was present in the panel where the metal sleeve was located; however, no reinforcing steel was evident in the slab between the bolt and the concrete edge. The ungrouted bolts remained in the concrete slab, whereas the grouted bolts tore out a wedge of concrete from the slab. The panels on the other three sides of the building did not fall off; however, the concrete slab on which the panels were seated exhibited significant cracking where the bottom connections were located, as shown in Fig. 10. It was not clear whether damage of this nature was visible following earlier earthquakes, but it was clear that the connections had been undermined. As such, the remaining panels were pulled down. Witnessing the panels being pulled down by an excavator confirmed that the panels all came down together along each side of the building. Case study 3: storey-height flat panels A nine-storey tower (−43.532, 172.635) in the Christchurch CBD is an excellent example of a building clad almost entirely in precast concrete panels, as shown in Fig. 11. Most of the panels are storey-height panels with central window openings. Smaller panels are used at the corners of the building. The panels performed well with only minor damage. The performance level was deemed Immediate Occupancy. Damage in the form of cracking was observed in the corners of two panels, as shown in Fig. 11. Minor damage to the sealing joints between panels was also observed. Heavier damage was observed in the lower corner of the corner panels. This would have presented a falling hazard to pedestrians due to the likely size of concrete pieces. However, the two-storey podium around the base of the tower virtually eliminated this risk. The tie-back connections can be seen in Fig. 11. The length of the bolts of these connections shows that these panels are designed to allow a reasonable amount of differential displacement between the panels and the structure. It did not appear that the bolts had been taken outside of their elastic range as there was no permanent deformation visible. There was also no evident residual displacement of the panels. The damage to the panels can be reasonably easily remediated with patching and grouting of the damaged regions.

Discussion

In order to reduce damage to façade systems in the future, both technical and regulatory issues need to be addressed. Improvements are required in order to understand better the behaviour of many façade systems and whether the methods used to isolate them are satisfactory. Design guidelines are required for both designers and installers of façade systems. Communicating common errors that should be avoided is also important. It is evident from what has been presented that many façade systems are not meeting their design criterion since they still pose a reasonable risk to life safety. In order to improve this situation, it is necessary to assign the respon-

Structural Concrete 13 (2012), No. 1

Fig. 11. Storey-height flat precast panel case study (–43.532, 172.635) (clockwise from top left): view of structure and panels, corner crack in lower panel, corner cracks in corner panels, view of connections from inside wall cavity

sibility of ensuring that a building’s façade is seismically proficient. This, coupled with mandated regulations for suppliers and installers, will help to ensure suitable design and construction. Currently, there are no standards written specifically for the design and/or installation of façade systems. Consequently, there is no way quality control can be enforced since there is no regulation that needs to be adhered to! Further, the installation of the façade system is usually performed by a subcontractor to the main contractor. This becomes an issue since there is an incentive for the main contractor to obtain the cheapest possible façade installation in order to win the contract. The case studies above focus predominantly on describing the system and the damage observed. It should be remembered that storey-height panels serve many more functions that those of spandrel panels. Storey-height panels, as with most façade systems, need to keep water out of the building, prevent air leakage, control the passage of light, heat and sound as well as provide an aesthetic finish to the building. Whether the performance of any of these systems has been jeopardized by the earthquake has not been investigated. A long-term research programme into improving existing façade systems, developing integrated design procedures and investigating cost-effective, damage-free façade solutions is currently being conducted at the University of Canterbury [7]. Solutions to improve building performance (structural and non-structural) include improving the disconnection of the façade from the structure, using the façade for controlled stiffening or damping and integrating the façade into the structure.

Conclusion

Earthquake damage to façade systems undoubtedly poses a major threat to life. The economic implications of façade damage are also significant due to business downtime and repair costs. Many buildings within the Christchurch CBD remain unoccupied due to non-structural damage despite the building retaining its structural integrity. In addition to the damage sustained in Septem-

A. Baird/A. Palermo/St. Pampanin · Façade damage assessment of concrete buildings in the 2011 Christchurch earthquake

ber 2010 and February 2011, continued façade damage has occurred in the subsequent aftershocks. A survey of 173 RC buildings and their 298 respective façade systems has revealed all types of damage to all the different typologies of façade systems. The survey has shown that in order to achieve façade systems that do not incur significant damage in design level earthquakes, major improvements are still required. References 1 New Zealand Police http://www.police.govt.nz (accessed 10 Feb 2012). 2. GeoNet. http://geonet.org.nz (accessed 16 Jul 2011). 3. Kam, W. Y., Pampanin, St.: The seismic performance of RC buildings in the 22 February 2011 Christchurch earthquake. Structural Concrete, 2011, vol. 12, pp. 223–233. 4. Johnston, J. A. R.: A brief history of damage in earthquakes in Wellington City and developments in multi-storey building construction in New Zealand. 2nd World Conference on Earthquake Eng., Tokyo, Japan, 1960, pp. 457–471. 5. Federal Emergency Management Agency. Prestandard and Commentary for the Seismic Rehabilitation of Building. 2000, FEMA 356. 6. Applied Technology Council. Procedures for Postearthquake Safety Evaluation of Buildings & Addendum. 1989, ATC-20. 7. Baird, A., Palermo, A., Pampanin, S., Riccio, P., Tasligedik, A. S.: Focusing on Reducing the Earthquake Damage to Façade Systems. Bulletin of the New Zealand Society for Earthquake Engineering, 2011. 8. Page, I.: Cladding Types in New Buildings. BUILD, 2008, pp. 55–56 9. Okazaki, T., Nakashima, M., Suita, K., Matusmiya, T.: Interaction between Cladding and Structural Frame Observed in a Full-Scale Steel Building Test. Earthquake Engineering & Structural Dynamics, 2007, vol. 36, pp. 35–53. 10. Kodur, V. R., Erki, M. A., Quenneville, J. H. P.: Seismic design and analysis of masonry-infilled frames. Canadian Journal of Civil Engineering, 1995, vol. 22, pp. 576–587. 11. Standards New Zealand. Structural Design Actions, Part 5 Earthquake Actions – New Zealand. 2004, NZS 1170.5. 12. Federal Emergency Management Agency. Communicating with Owners and Managers of New Buildings on Earthquake Risk: A Primer for Design Professionals. 2004, FEMA 389. 13. Applied Technology Council. Database on the Performance of Structures near Strong-Motion Recordings: 1994 Northridge, California, Earthquake. 2000, ATC-38. 14. Standards New Zealand. Timber Framed Buildings. 1999, NZS 3604. 15. Dolsek, M., Fajfar, P.: The Effect of Masonry Infills on the Seismic Response of a Four-Storey Reinforced Concrete Frame – A Deterministic Assessment. Engineering Structures, 2008, vol. 30, pp. 1991–2001.

Andrew Baird Ph.D. Candidate Department of Civil & Natural Resources Engineering University of Canterbury, New Zealand andrew.baird@pg.canterbury.ac.nz

Alessandro Palermo Senior Lecturer Department of Civil & Natural Resources Engineering University of Canterbury, New Zealand alessandro.palermo@canterbury.ac.nz

Stefano Pampanin Associate Professor Department of Civil & Natural Resources Engineering University of Canterbury, New Zealand stefano.pampanin@canterbury.ac.nz

Structural Concrete 13 (2012), No. 1

Articles Alessandro Palermo* Liam Wotherspoon Lucas Hogan Mitchel Le Heux Elena Camnasio

DOI: 10.1002/suco.201100041

Seismic performance of concrete bridges during Canterbury earthquakes In less than six months, the city of Christchurch, New Zealand, experienced two major earthquakes: on 4 September 2010 and 22 February 2011. The former was generated by the rupture of the previously unknown Greendale fault, releasing a magnitude Mw 7.1 earthquake 30–40 km away from the Central Business District (CBD); the latter event, of magnitude Mw 6.2, was less than 10 km from the CBD on an unknown buried fault at the edge of the city. There was widespread damage to the lifelines covering not only Christchurch City but also the closest districts of Selwyn and Waimakariri. The different nature of the fault ruptures and locations of the two events resulted in a variation in damage levels between the earthquakes throughout the region. Teams from various organizations performed inspections on over 800 bridges throughout the affected Canterbury region. No major collapses were registered among concrete bridges and only 20 bridges required closure due to damage caused by the two earthquakes. Owing to the nature of the Canterbury soils, extensive liquefaction and lateral spreading occurred throughout the region. It was this lateral spreading that caused most of the traffic disruption and closure of bridges, due to damage to the abutments and approaches, foundation settlement and rotation. The authors aim to give a detailed overview of the damage assessment and seismic performance of the Canterbury bridges during these two earthquakes, emphasizing unexpected issues that are still not properly detailed in New Zealand and overseas standards.

City, damage was mainly confined to the bridges in the central and eastern regions, with lateral spreading affecting bridges spanning both the Avon and Heathcote rivers, whereas in the Waimakariri district, the town of Kaiapoi sustained massive damage due to liquefaction during the Darfield earthquake (Fig. 1). Very few bridges developed significant damage on non-liquefiable sites. Abutments, approaches and piers suffered varying levels of damage, with very little damage observed in bridge superstructures. This article is a summary of the authors’ field observations and preliminary investigations undertaken after both events [1, 2], presenting an overview of the damage to the bridge inventory in the Canterbury region. A joint research group from the University of Canterbury and University of Auckland plus international overseas bridge experts inspected several bridges, collaborating with designers engaged by Christchurch City Council and the New Zealand Transportation Agency (NZTA) to perform post-earthquake visual assessments. Observations will mainly comprise Christchurch City Council road bridges, pedestrian and highway bridges. Particular attention is also given to non-structural components, for example damaged pipelines running along bridges, and approaches, thus highlighting the need for an integrated design approach to structures and utilities.

Introduction

The region in and around Christchurch, comprising the Christchurch City, Selwyn and Waimakariri districts, contains over 800 road, rail and pedestrian bridges. Some 55 % of those bridges are integral reinforced concrete structures or hybrid designs, i.e. precast concrete deck and cast-in-place substructure (piers, foundations). All bridges have symmetric characteristics, with small to moderate total spans of between 10 and 30 m, and very stiff superstructures. Most of the damage to these bridges caused by the two earthquakes (Darfield, 4 September 2010, and Christchurch, 22 February 2011) coincided with areas highly susceptible to liquefaction. In fact, for Christchurch * Corresponding author: alessandro.palermo@canterbury.ac.nz Submitted for review: 15 August 2011 Revised: 30 August 2011 Accepted for publication: 31 August 2011

Fig. 1. Aerial view of Christchurch and surrounding region, indicating the locations of a selection of damaged bridges, strong motion stations (red lettering), Central Business District (CBD, red square) and the epicentres of the major earthquakes (Google Inc., 2011)

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

Seismic demand

On 4 September 2010 the Mw 7.1 Darfield earthquake occurred with an epicentre near the town of Darfield, 30–40 km west of the Christchurch Central Business District (CBD). Large vertical accelerations were registered in the region of fault rupture, typical of the near-source strong motion recordings. The 22 February 2011 Christchurch earthquake (Mw 6.2) had an epicentre less than 10 km from the Christchurch CBD, between Lyttelton and the south-eastern edge of the city [1]. The close proximity and shallow depth of this event caused higher intensity shaking in Christchurch compared with the Darfield event. Despite its short duration (15–20 s), the 22 February 2011 event recorded one of the highest maximum peak ground accelerations (PGAmax = 2.2 g, close to epicentre) ever experienced close to a major city. In the CBD, ground motions were also characterized by large vertical accelerations. Further aftershocks occurred dur-

Table 1. Comparison of peak ground accelerations in the vicinity of Christchurch for the Darfield, Christchurch and Sumner earthquakes (geometric mean of horizontal PGAs)

Event

Fig. 2. Response spectra of the geometric mean of the horizontal accelerations at strong motion station recordings in central and eastern Christchurch compared with NZS1170.5 design response spectrum for Christchurch, site subsoil class D for 500-year return period: a) Darfield earthquake; b) Christchurch earthquake. The four-letter abbreviations stand for the different strong motion stations (see Fig. 1 for locations).

PGA Maximum horizontal

Maximum vertical

Darfield, 4 September 2010

7.1

0.8 g

1.3 g

Christchurch, 22 February 2011

6.2

1.7 g

2.2 g

Sumner, 13 June 2011

6.0

2.0 g

1.1 g

ing the following months, with one of the strongest being the Mw 6.0 event on 13 June 2011, with an epicentre again on the south-eastern edge of the city (Table 1). The approximate period range of Canterbury bridges can be roughly estimated as < 0.8 s, as indicated by the rectangles in Figs. 2a and 2b. Acceleration response spectra of typical sites from both the Darfield and Christchurch events are compared with the New Zealand design spectra [4] for site subsoil class D, 500-year return period. The geometric means of the horizontal acceleration response spectra for five strong motion stations (CCCC, HPSC, HVSC, PRPC, SHLC) close to bridges damaged during the Canterbury earthquakes are considered for comparison. Within the period range of 0–0.8 s, it is immediately evident that during the Darfield event (Fig. 2a) the spectral accelerations were less than or equal to the design spectral accelerations. Recorded values of peak horizontal ground accelerations (PGA) were 0.15–0.3g in the Christchurch CBD, compared with the 0.4g design PGA. In the Port Hills area (Heathcote Valley Primary School, HVPS) a PGA of 0.6g was recorded due to site amplification effects. During the 22 February Christchurch earthquake, horizontal accelerations of 1.7g and vertical accelerations of 2.2g were recorded near the epicentre at the HVPS strong motion station. In the area of south-east Christchurch, Fig. 2b shows that spectral acceleration values were much higher than the design level in the period range of most New Zealand road and highway bridges. More importantly, due to the shallowness of the fault and the vicinity of the bridges to the epicentre, exceptional vertical acceleration values were also registered, especially compared with those from other events shown in Table 1. This certainly contributed to unexpected phenomena identified in the pier damage at some bridges.

Damage associated with liquefaction

During the Darfield earthquake, extensive liquefaction and lateral spreading occurred in Kaiapoi, Brooklands, Selwyn and eastern Christchurch suburbs. The majority of bridges with damage of sufficient severity to cause traffic disruption were all located in these areas of liquefaction and lateral spreading. This strong geographical link indicated that liquefaction-induced lateral spreading was the primary mechanism for bridge damage. The effects of

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

River suffered much less damage, with less extensive liquefaction. Apart from Ferrymead Bridge (–43.5584, 172.7086) at the mouth of the Heathcote, all bridges were either undamaged or suffered only minor damage. Typical damage was minor approach settlement, with little impact on bridge abutments and superstructure.

Fig. 3. UC drive-through liquefaction survey map after 22 February 2010 event (University of Canterbury) [courtesy of Misko Cubrinovski]

the Christchurch earthquake were much more localized; however, the significant ground motions in the city resulted in large areas of severe liquefaction damage. Accordingly, bridges along both the Avon and Heathcote rivers within the city suffered from varying levels of damage due to lateral spreading, with ground conditions and distance from the fault rupture influencing this response. Fig. 3 shows the regions of liquefaction identified during a survey of the Canterbury area after the 22 February 2011 event and bridges that were severely damaged. Red lines indicate regions of moderate to servere liquidation, orange low to moderate liquefaction, purple liquefaction on roads only, and finally blue regions with no liquefaction. The type of bridge damage along the Avon was fairly consistent, with settlement and lateral spreading of approaches, back rotation and cracking of the abutments, some pier damage and severe damage to many lifelines crossing bridges. In most cases the bridge decks restrained the movement of the top of the abutment, limiting longitudinal movements but inducing internal actions in the structure. Moderate to severely damaged bridges typically had pile foundations, on which lateral spreading forces placed large demands, likely resulting in plastic hinging below grade. The approach fill of several bridges subsided by up to 1.0 m due to the cumulative effect of these earthquakes, resulting in temporary bridge closures. In most cases, settlement and spreading of the approaches impacted on the serviceability and operation of the bridge. The Christchurch CBD bridges crossing the Avon River performed well, with the most common damage being minor lateral spreading, compression or slight slumping of approach material and minor cracking in abutments. Compared with the Avon River, bridges crossing the Heathcote

Road bridges

Most of the bridges severely damaged by the earthquakes were located in the areas highly prone to liquefaction. However, bridge structures themselves suffered only a moderate amount of damage compared with residential and commercial buildings. This is mainly because the gravity design of road bridges intrinsically results in an axially strong superstructure that can resist the axial demand placed on the bridge by lateral spreading forces. So although the majority of bridges were not designed to resist lateral spreading forces, their intrinsic strength and stiffness enabled them to cope with such actions. Nevertheless, as some bridges critical to the city’s infrastructure sustained substantial approach damage, extensive traffic disruption was evident immediately following the events. Fig. 4 shows the critical road bridges after the Christ-church earthquake. Damaged bridges can be subdivided further: according to their type of construction, as listed in Table 2. This subdivision indicates that since the 1960s, the precast typology has become predominant. Moreover, it can be shown that monolithic structures showed a relatively stiff and sturdy response; however, the induced internal forces and moments caused structural damage, especially to the piers.

3.1

Abutments and foundations

Abutments exhibited two different damage types: residual displacements/rotations of the substructure and damage to the superstructure (cracks, concrete crushing and spalling). Both were caused by lateral spreading of the approaches and riverbanks. One commonly observed effect of lateral spreading on bridges was the tendency for the abutments to rotate backwards towards the riverbanks or to converge inwards, meaning that the bridge superstructure acted as a rigid strut, whereas the foundations underwent forced rotations or movements in the direction of lateral ground flow (Fig. 5a). This effect was most prevalent on bridges with deep abutments or closely spaced abutment piles. These abutment types are in contrast to the current design philosophy for lateral spreading banks which uses a shallow beam-type abutment and a small number of deep piles [4]. These are designed to allow the laterally spreading slope to spill past the abutment.

Table 2. Damaged Canterbury bridges listed according to type and year of construction

Integral bridges

Precast bridges

Bridge

Year

Bridge

Year

Pages Road Bridge Gayhurst Road Bridge Avondale Road Bridge Moorhouse Avenue Bridge

1931 1954 1962 1964

Fitzgerald Avenue Twin Bridges Ferrymead Bridge (Ferry Road) Bridge Street Bridge Anzac Drive Bridge

1964 1965 1980 2000

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

Fig. 4. Overall views of some of the damaged bridges after 22 February 2011; clockwise from top left: Gayhurst Road Bridge, Fitzgerald Avenue Bridge, Ferrymead Bridge, Bridge Street Bridge, Moorhouse Avenue Bridge, Avondale Bridge a)

Fig. 5. Damage to abutments: a) sketch of lateral spreading forces acting on abutments and foundations; b) back-rotated abutments of Bridge Street Bridge [photo: M. Bruneau]; c) back-rotated abutments of Fitzgerald Avenue Bridge [courtesy of OPUS International Ltd]; d) damaged abutments of Avondale Road Bridge [courtesy of OPUS International Ltd]

During the Darfield earthquake, the western abutment of Bridge Street Bridge (−43.5252, 172.7241) rotated by approximately 5° due to lateral spreading, and light cracking was observed on the tension face of the abutment piles after the event. This pile damage was exacerbated during the Christchurch earthquake, with further abutment rotation to > 12° at the western end of the span (Fig. 5b) and plastic hinging clearly visible on the western abutment piles (Fig. 6a). Great uncertainties remain with regard to potential plastic hinging below the ground sur-

Fig. 6. Damage to pile foundations: a) plastic hinging in abutment piles of Bridge Street Bridge [photo: M. Bruneau]; b) exposed piles during the securing works at Avondale Road Bridge [courtesy of OPUS International Ltd]; c) exposed piles during the securing works at Gayhurst Road Bridge [courtesy of OPUS International Ltd]; d) flexural cracks in Fitzgerald Avenue Bridge pile [photo: L. Hogan]

face. Hinging is likely to have occurred in many of the bridges with visible abutment damage; this has been witnessed in bridges with similar soil conditions in recent earthquakes [6]. Following the 13 June 2011 aftershocks, the exposed pile height beneath the abutment was 1.3 m. Other bridges that suffered foundation damage due to abutment rotation are shown in Figs. 5c and 5d. In particular, the piles of Avondale Road Bridge (−43.5006, 172.6878) and Gayhurst Road Bridge (−43.5216, 172.6728) were inspected during the securing works following the 22 February 2011 event (pile damage is shown in Figs. 6b

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

Fig. 7. Damage to bridge decks: a) Colombo Street Bridge, buckling of steel arch [photo: M. Yashinsky]; b) pounding of deck on Bridge Street Bridge abutment after Christchurch event [photo: A. Kivell]; c) damage to Fitzgerald Avenue Bridge superstructure due to abutment rotation [photo: L. Hogan]

and 6c). Plastic hinging, spalling of the concrete and yielding of the reinforcing steel were also evident at Bridge Street Bridge and Fitzgerald Avenue Bridge (−43.5262, 172.6506) (Figs. 6a and 6d).

3.2

Superstructure

Overall, road bridges exhibited little damage to their superstructure, even if most of the City Council bridges constructed in the 1950s and 1960s were designed using rudimentary seismic codes. Deck damage was limited in these bridges due to their overall robustness. Fig. 7 shows clear examples of this minor damage, caused by deck-abutment pounding and/or permanent displacements.

3.3

Piers

Pier damage was caused either by ground shaking, with a combination of vertical and horizontal components, or by lateral spreading. Bridge piers performed well in most cases, without experiencing extensive plastic hinging due to ground shaking. Flexural cracks are an expected consequence of the design structural failure mechanism of the bridges, and the limited damage did not compromise the structural integrity of the bridges. However, unrepaired cracks may become a potential threat to the design life of

bridges close to marine environments, providing an accelerated pathway for chloride ingress and compromising the overall durability of the structure [7, 8]. Some bridges experienced pier cracking as a result of extensive lateral spreading. For example, Gayhurst Road Bridge exhibited a single horizontal crack along one face of a pier, approx. 1 m from the deck soffit (Fig. 8a). Lateral spreading exerted a lateral force on the pier base and caused a large moment at the stiff pier-deck interface, thus cracking the pier. The liquefied soil layer would have also reduced the lateral stiffness of the pier foundation system, allowing rotation of the bottom of the pier towards the middle of the river. Ferrymead Bridge was undergoing a major upgrade (deck widening) when the earthquake occurred. Lateral spreading caused significant damage to the replacement structure and some temporary construction platforms. The existing structure also sustained damage due to lateral spreading, with permanent rotation and cracking of a number of the piers situated in the estuary (Fig. 8b). Only Moorhouse Avenue Overbridge (−43.5399, 172.6367) suffered significant pier damage during the Christchurch earthquake due to transverse ground shaking. The overall performance of the structure was unsatisfactory, with significant shear cracking and buckling of the piers. The structure was constructed in three separate sec-

Fig. 8. a) Damaged pier at Gayhurst Road Bridge [photo: L. Hogan]; b) cracked and rotated piers at Ferrymead Bridge [photo: Christchurch City Council]

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

joint on only one side of the bridge induced irregularity in the bridge structure. In fact, with the west and central part of the bridge linked, the bridge pier at the eastern expansion joint suffered extensive displacement demand. The slenderness of the pier affected the vertical load-carrying capacity of the structure along with the lateral capacity. The columns also had widely spaced transverse reinforcement, making the structure susceptible to a brittle failure mechanism (Fig. 10a). Observations after the Christchurch event indicated that the damaged columns had started to buckle (Fig. 10b), putting the central span at risk of collapse. In this instance, the damage was induced by extensive ground shaking; large transverse horizontal accelerations may have caused a flexural-buckling failure mechanism in the columns. Due to the higher displacement demand in the west/central part of the bridge, the deck pounded against the south-west abutment of the bridge, causing extensive spalling and bar buckling (Fig. 10c).

3.4

c) Fig. 9. Moorhouse Avenue Overbridge: a) bridge elevation showing locations of expansion joints and steel rod linkages; b) plan of bridge showing qualitative displacement profile under transverse loading; c) typical pier elevation and horizontal section through pier with expansion joint

tions, linked by expansion joints (Fig. 9). The bridge sustained damage to one column near the north-east approach where a deck expansion joint was located. The insertion of steel rod linkages in the deck at the expansion

Approaches

In general, approach damage was caused by liquefactioninduced lateral spreading, with movement of the riverbanks towards the river. In some instances, although the bridge structure did not suffer serious damage, the approaches were severely damaged, making access to the bridge difficult or impossible. Differential movements led to the formation of large vertical offsets between bridge deck and approach. A typical example is Davis Road Bridge (−43.5222, 172.6601), where the bridge itself, based on a post-earthquake survey undertaken by Selwyn District Council, did not settle and remained relatively unmoved. However, a vertical movement of the laterally spreading bank led to the formation of a large differential vertical settlement between the bridge deck and the approach of approximately 500 mm (Fig. 11a).

Fig. 10. Damage to Moorhouse Avenue Overbridge: a) shear failure mechanism of pier at point B’ in Fig. 9 [photo: A. Palermo]; b) close-up view of failed pier showing buckled bar on column side of expansion joint and tension on opposite face [photo: M. Bruneau]; c) concrete spalling and bar buckling at southwest abutment [photo: A. Kivell]

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

Fig. 11. Relative vertical movement between approach and deck at: a) Davis Road Bridge, Lincoln [photo. L. Wotherspoon]; b) Gayhurst Road Bridge [photo: L. Hogan]; c) Bridge Street Bridge [photo: L. Wotherspoon]; d) Snell Bridge [photo: M. Le Heux]

Another form of damage was cracks in the approach due to soil–bridge interaction. One such example is Gayhurst Road Bridge, where cracks in the approaches parallel to the axis were observed (Fig. 11b). These cracks provided evidence of soil–bridge interaction induced by the monolithic span of the bridge resisting lateral spreading. Similar cracks are shown in Figs. 11c and 11d. The formation of these cracks parallel to the longitudinal axis of the bridge can be explained by the “strut effect” analogy. This considers a bridge as a strut spanning across the river and bearing on the river banks. As lateral spreading occurs, the river banks place a large compression demand on the bridge, i.e. the strut; therefore, the cracks running parallel to the bridge axis are tensile stress cracks in the bearing area of the bridge on the riverbanks.

Temporary repairs

Moorhouse Avenue Overbridge was the only bridge out of service to all traffic for almost a month after the 22 February Christchurch earthquake. Initially, temporary strengthening works were erected around the failed pier (Fig. 12a). These consisted of two built-up square block concrete columns with timber blocking to prop the bridge while further works were undertaken. A multi-span struc-

Structural Concrete 13 (2012), No. 1

tural steel frame spanning between the two failed columns was designed and constructed. This left the tapered deck beams bearing on the steel portal with a smaller cross-section, which was only intended as a temporary solution to stop complete collapse of the damaged section. This solution did not provide any additional lateral stability and therefore the overbridge was not reopened to vehicle traffic until 31 March 2011, when the final temporary repair solution consisting of dual cross-bracing units at each of the weakest piers was implemented (Fig. 12b). Bridge Street Bridge was temporarily repaired after both the Darfield and Christchurch events. Its closure and the consequent detour increased travel time by two to four hours because of the disruption associated with damage to the rest of the road network. This damage was remediated by building up the approach with gravel, and lightweight traffic was able to use the bridge again within four days. Ferrymead Bridge was temporarily repaired with prestressed rods connecting the bottom of the piers to the abutments (Figs. 12c and 12d). This prevented further inward movements of the piers towards the river/estuary which might compromise overall stability of the bridge. This bridge, which carries 30 000 vehicles per day, still had heavy traffic restrictions six months after the Christchurch earthquake.

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

Fig. 12. a) Temporary propping to Moorhouse Avenue Overbridge [photo: G. Whitla]; b) temporary repair solution for Moorhouse Avenue Overbridge [photo: A. Palermo]; c) securing works for Ferrymead Bridge [courtesy of OPUS International Ltd]; d) close-up view of inclined rods at Ferrymead Bridge [courtesy of OPUS International Ltd]

Highway bridges

Few highway bridges were severely damaged during the Darfield and Christchurch events. One of the reasons for this success is a recent seismic retrofit programme to reduce the seismic risk of the national highway bridges. The majority of bridges at risk in the Canterbury region underwent some form of seismic retrofitting during the past 20 years in particular, after an overall seismic screening of the national highway network [9]. One of the most widespread retrofit programs was undertaken by Transfund New Zealand (now incorporated into NZTA) [10], which saw the installation of tie rods and steel brackets acting as the transverse shear keys at the pier–deck and deck–abutment junctions respectively. Similar devices were installed on bridges on most of the highways in the region. Post-earthquake investigations have not been detailed enough to ascertain the level of demand that these devices experienced during the seismic shaking. However, no retrofitted bridges were severely damaged structurally during the Darfield earthquake. Nevertheless, lateral spreading damage was observed on some key lifeline bridges. For example, Chaney’s Overpass (−43.4297, 172.6464) on State Highway 1 north of Christchurch (Fig.

13, top left) was found to be structurally sound, and was tied to the abutment walls to prevent unseating. However, due to liquefaction of the soils surrounding the bridge (Fig. 14b), the approach to the southbound lanes of State Highway 1 settled by approximately 5–10 cm and large cracks were observed in the built-up approach. After a brief closure for inspection, this busy route was reopened with signage reducing the speed to 30 km/h (down from 100 km/h) due to the rough road surface. As with road bridges, the Christchurch event caused more damage to the highway bridges. The Port Hills and Horotane Overbridges (−43.5725, 172.6947), (Fig. 13, top right and bottom left), which are within 200 m of each other and close to the epicentre, suffered different structural damage. The two bridges had been recently retrofitted under the retrofit programme mentioned above. Both retrofits were designed to link spans together and to the abutments in order to create an “integral cast-in-place bridge” concept. This method appears to have worked well, and despite both bridges sustaining damage, they were able to carry traffic again soon after the earthquake. The six-span Port Hills Overbridge suffered pier damage due to transverse ground shaking as shown in Fig. 14a, and the linkages between the span and the abutments had elongated. The Horotane Overbridge suffered abutment

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

Fig. 13. Overall views of some damaged highway bridges after 22 February 2011; clockwise from top left: Chaney’s Overpass [photo: M. Anagnostopoulou]; Port Hills Overbridge [courtesy of OPUS International Ltd]; Anzac Drive Bridge [photo: E. Camnasio]; Horotane Overbridge [courtesy of OPUS International Ltd]

Fig. 14. a) Buckling of reinforcing steel at bases of piers to Port Hills Overbridge [courtesy of OPUS]; b) liquefaction ejecta around the piers of Chaney’s Overpass [photo: M. Anagnostopoulou]; c) backward rotation of abutments of Anzac Drive Bridge [photo: A. Palermo]; d) sheared bolt at Horotane Overbridge abutment retrofit [photo: J. Allen]

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

damage due to incipient slope failure of the embankments. The ties between spans and at abutments also elongated and pulled out as they had in the Port Hills Overbridge. Additionally, 60 % of the bolts that attached the soffit of the precast concrete beams to the abutment seat extension had sheared off (Fig. 14d). If these spans had not been tied together and the seats not extended, it is quite likely that the spans would have collapsed. Similar to other road bridges, Anzac Drive Bridge (−43.5009, 172.7011), shown in Fig. 13, bottom right, developed permanent abutment rotations of up to 6° during the 22 February 2011 event (Fig. 14c). This amount of rotation and the lateral spreading loads placed large demands on the steel piles below the abutments, and likely resulted in plastic hinging below grade. After the 13 June 2011 aftershocks, piles were exposed over a length of 1300 mm in the south abutment.

Pedestrian bridges

Pedestrian bridges experienced damage similar to road bridges during both earthquakes, primarily due to lateral spreading. However, due to their relatively low deck stiffness and strength, axial compressive forces as a result of lateral spreading and settlement of the riverbanks led to deformation and complete failure of many pedestrian bridge decks under compressive buckling modes of failure. This type of failure affected many bridges during the Darfield earthquake, with further damage to the same bridges following the Christchurch earthquake. Figs. 15 and 16 show a selection of the most seriously damaged pedestrian bridges after the latter event.

After 4 September 2010, Dallington and Porritt Park pedestrian bridges had visible damage to their abutments. At Dallington Bridge (43.5179, 172.6770) (Fig. 15), a posttensioned double-flange concrete arch deck, a threepinned arch formed in response to lateral spreading during the Darfield earthquake. Two positive flexural cracks formed at the abutments at each end of the bridge in the horizontal plane (Fig. 15, detail 1), whereas at mid-span a negative flexural hinge developed (Fig. 15, detail 2). In the 2011 event, the damage was compounded, with continued superstructure deformation at mid-span, spalling concrete cover under negative bending and torsional deformation of the superstructure about the longitudinal axis of the bridge. In concrete bridges with flat decks, such as precast panels or cast-in-place slabs, a compressive load passing through the deck is unlikely to cause a hinge to form as it did in this case. This is because there is no lever arm between the point of application of the compressive force at the riverbanks and at the point of resistance at the centre of the bridge, which means there is no moment demand. Dallington Pedestrian Bridge had a lever arm due to the arch form, so there is a moment demand, thus leading to a plastic hinge forming at mid-span where the section has low strength and stiffness. The same bridge also exhibited severe damage to the pilecaps of the abutment on the side of the river that was subjected to extensive lateral spreading (Fig. 15, detail 1). The bridge underwent a relatively expensive retrofit programme more than 10 years ago because it also acts as a “utility” bridge, carrying two 66 kW power cables. This bridge is used as a case study to assess fundamental lateral spreading principles in [11]. The other case, Porritt Park Pedestrian Bridge (−43.5145, 172.6829), shown in Fig. 16a, experienced

Fig. 15. Dallington Pedestrian Bridge, clockwise from top left: overall view [photo: A. Palermo]; approach and abutment damage [photo: A. Palermo]; schematic of three pinned arch mechanism; close up view of plastic hinge at midspan from the side [photo: A. Kivell] and from underneath [photo: M. Le Heux]; damage to abutment with close up of crack [photo: M. Le Heux]

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

Fig. 16. Porritt Park Pedestrian Bridge: a) end view [photo courtesy of OPUS International Ltd]; b) backward rotation of abutment and exposure of piles [photo courtesy of OPUS International Ltd]; c) repairing damage to approach and riverbank lateral spreading [photo: L. Hogan]

large rotation and translation of the south abutment. This abutment sits on six small piles approx. 200 mm in diameter (Figs. 16b and 16c). It is likely that the piles do not extend very deep and, as such, they provided minimal lateral stiffness to the system, only serving to increase the surface area of the soil wedge with which they interact and thus exacerbating the problem. This rotation resulted in extensive cracking at the abutment face as well as hinging at the deck–abutment connection. Rotation imposed on the bridge by the abutment also led to a plastic hinge forming in the double-T beam at 1.5 beam depths from the abutment. This distance is a likely point at which joint reinforcement was curtailed. After the aftershocks on 13 June 2011, the deck lost bearing support and collapsed. Fifield Terrace Footbridge (−43.5552, 172.6594) exhibited significant superstructure damage following the Christchurch earthquake, which rotated the bridge about its vertical plane (Fig. 17a). The steel studs linking the pier to the superstructure was the weak point in the deck, which would have been further influenced by the significant rotation of the steel piers (Fig. 17c). Deformations were so extreme that the rectangular hollow steel section at the extremity of the deck ruptured through its weld (Fig. 17b).

Damage to lifelines associated with bridges

Ground shaking and liquefaction/lateral spreading were devastating not only for bridge structures but also for a)

pipeline systems crossing these bridges. Many utilities are located along the longitudinal bridge axis under the deck or within the bridge deck. Lifeline networks were severely damaged along the Avon River due to extensive liquefaction and lateral spreading. Quite surprisingly, areas where soil-bridge interaction occurred in lateral spreading reported damage to pipelines as well. In fact, several pipes were damaged due to differential settlement between the bridge and the surrounding soil. This confirms that pipe connections were not appropriately designed to accommodate deck-to-pipe, or abutment-to-pipe relative displacements. A typical example is the Dallington Pedestrian Bridge mentioned previously, with lateral spreading creating extensive cracking parallel to the bridge and perpendicular to the alignment of the road and buried pipes (Fig. 18f), potentially leading to damage to sewer pipes in the road. As mentioned earlier, this bridge has two 66 kW power cables placed under the bridge deck, providing electricity for 20 000 inhabitants, making this modest pedestrian bridge a structure with strategic importance. Some moderate to extensive damage was observed in many road bridges where pipes were distorted and/or leaking in the proximity of deck-abutment and abutment-approach connections. As shown in Fig. 18, the main issues arose with stiff pipes, such as sewage and water pipes, as they are fully fixed to the deck and usually run through the abutments. On the other hand, the flexibility of power and/or telephone cables was able to accommodate larger displacement demands. c)

Fig. 17. Fifield Terrace Pedestrian Bridge: a) overall view after 22 February 2011; b) damage to deck; c) backward rotation of pier

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio 路 Seismic performance of concrete bridges during Canterbury earthquakes

Fig. 18. a) Damage to pipelines at Kainga Road Bridge after 4 September 2010 [photo: A. Palermo]; b) Gayhurst Road Bridge: damage to lifelines [photo: A. Palermo]; c) repair work at Bridge Street Bridge [photo: A. Palermo]; d) close-up view of damaged pipes along the deck of Ferrymead Bridge [courtesy of OPUS]; e) damaged pipelines at Bridge Street Bridge after 13 June 2011 [photo: E. Camnasio]; f) Dallington Pedestrian Footbridge: damage to electrical services for Dallington area [photo: M. Le Heux]

Conclusions

Overall bridge structures performed well during the 2010 and 2011 Canterbury earthquakes, confirming the design expectations [5]. The robustness of Christchurch City Council road bridges built in the 1940s and 1950s without any seismic design criteria certainly helped to sustain earthquake loadings comparable with or higher than the current design levels. However, the significant damage to

the approaches due to lateral spreading of the riverbanks caused disruption which could be mitigated through possible low-damage approach solutions. Life safety of bridges is still the primary focus of current design standards; however, further advanced technology should be implemented to preserve not only the integrity of the structure [12, 13], but also the foundations and approaches. Importantly, utilities and pipes that are commonly carried over bridges sustained severe damage due to a lack of

Structural Concrete 13 (2012), No. 1

A. Palermo/L. Wotherspoon/L. Hogan/M. Le Heux/E. Camnasio · Seismic performance of concrete bridges during Canterbury earthquakes

a design-integrated approach. New Zealand Standards as well overseas codes are deficient in terms of the aforementioned aspects and further research needs to be implemented.

Acknowledgements The authors wish to acknowledge the following institutions and consulting engineering companies for their collaboration and technical assistance and for supplying data: NZTA (John Reynolds), Christchurch City Council (Stewart Smith, Lloyd Greenfield, David McNaughton), Opus International Consultants Ltd (Michael Cowan, Anthony Rooke), KiwiRail (Samuel Grave), ELMAC Consulting Engineers (John MacKenzie). We also acknowledge the support and assistance of other researchers (Allan Scott, Anton Kivell, Michel Bruneau, Mark Yashinky, Aaron Bradshaw, John Allen) and students (Tiziana Cristini, Riccardo Diaferia, Francesco Sarti, Giacomo Paganotti, Simona Giorgini, Maria Brando) during the visual inspections of the bridge performed in the survey period. References 1. Palermo, A., Le Heux, M., Bruneau, M., Anagnostopoulou, M., Wotherspoon, L., Hogan, L.: Preliminary findings on performance of bridges in the 2010 Darfield earthquake. Bulletin of the New Zealand Society for Earthquake Engineering, 2010, 43, No. 4, pp. 412–420. 2. Palermo, A., Wotherspoon, L., Hogan, L., Kivell, A., Yashinsky, M., Bruneau, M.: Preliminary findings on performance of bridges in the 2011 Christchurch earthquake. Reconnaissance report, 2011; website: http://www.nzsee.org.nz/ link to Clearing House Darfield and Christchurch. 3. Geonet website: http://www.geonet.org.nz/earthquake/historic-earthquakes/; section 22 Feb Christchurch, 4 Sept Darfield earthquake. 4. Standards New Zealand: NZS 1170.5:2004. Structural Design Actions. Part 5: Earthquake actions – New Zealand, 2004. 5. Transit New Zealand: New Zealand Bridge Design Manual, 2nd ed., 2004. 6. Berrill, J., Yasuda, S.: Liquefaction and pile foundations: some issues. Journal of Earthquake Engineering, vol. 6, Special Issue 2002, pp. 1–41. 7. Akiyama, M., Frangopol, D. M.: On life-cycle reliability under earthquake excitations of corroded reinforced concrete structures. Proc. of 2nd International Symposium on Life-Cycle Civil Engineering, 27–31 Oct 2010, Taipei, Taiwan. 8. Biondini, F., Palermo, A., Toniolo,G.: Seismic performance of concrete structures exposed to corrosion: case studies of lowrise precast buildings. Structure and Infrastructure Engineering, First published on 1 Apr 2010, pp. 1744–8980. 9. Transit New Zealand: Manual for seismic screening of bridges, SM110, Wellington, New Zealand, 1998. 10. Chapman, H. E., Lauder, M. K., Wood, J.: Seismic assessment and retrofitting of New Zealand state highway Bridges. Proc. of New Zealand Society Earthquake Engineering Conference (NZEES), 11–13 Mar 2005, Wairakei (New Zealand), CD-ROM. 11. Le Heux, M., Palermo, A., Mackenzie, J.R.: Darfield earthquake 2010 – Lateral spreading actions on the Dallington pedestrian bridge. Proc. of 9th Pacific Conference on Earth-

Structural Concrete 13 (2012), No. 1

quake Engineering – Building an Earthquake-Resilient Society (PCEE), 14–16 Apr 2011, Auckland (New Zealand), CDROM. 12. Palermo, A., Pampanin, S., Calvi, G. M.: Concept and Development of Hybrid Solutions for Seismic Resistant Bridge Systems. Journal of Earthquake Engineering, 2005, 9, No. 6, pp. 899–921. 13. Palermo, A., Mashal, M.: Accelerated bridge construction (ABC) in earthquake-prone areas, international trends and New Zealand needs. Proc. of New Zealand Concrete Society Conference (NZCSC), 7–8 Aug 2011, Rotorua (New Zealand), CD-ROM.

Alessandro Palermo, Senior Lecturer University of Canterbury, New Zealand Dept. of Civil & Natural Resources Engineering Private Bag 4800, Christchurch 8140 Tel: +64 3 364 2987 ext. 8867 alessandro.palermo@canterbury.ac.nz

Liam Wotherspoon, EQC Research Fellow in Earthquake Engineering, University of Auckland, New Zealand

Lucas Hogan, PhD Student, University of Auckland, New Zealand

Mitchel Le Heux, Master’s Student Rose School of Pavia, Italy

Elena Camnasio, PhD Student Politecnico di Milano, Italy

Articles R.D.J.M. Steenbergen* A.H.J.M. Vervuurt

DOI: 10.1002/suco.201100031

Determining the in situ concrete strength of existing structures for assessing their structural safety EN 13791 applies when assessing the in situ compressive strength of structures and precast concrete components. According to the code itself, it may be adopted when doubt arises about the compressive strength of a concrete. For assessing the structural safety of existing structures, however, the method given in EN 13791 does not seem to be applicable and may lead to an unsafe approach for determining the characteristic concrete strength. This paper presents an alternative method for determining the characteristic concrete strength from cylinders obtained in situ. The method proposed is based on EN 1990 (Eurocode basis of design) and the corresponding Annex D. The method according to EN 13791 is outlined in this article. Moreover, the practical implementation of the method in accordance with clause 5.2 of EN 1990 is explained and an example is given. Finally, both methods are compared with each other. It is demonstrated that EN 13791 does not apply to safety assessments of existing concrete structures and the use of this code may lead to unsafe situations. Keywords: concrete strength, structural safety, codes

Determining the characteristic concrete strength to EN 13791

European Standard EN 13791 [1] prescribes methods and procedures for assessing the in situ compressive strength of concrete in structures and precast components. It provides principles and guidance for establishing the relationship between test results obtained from drilled cores and the in situ concrete strength. In most cases, to assess the in situ concrete strength, 6–12 cores are drilled, depending of the possibilities with respect to the structure under investigation. In this case, if the concrete compressive strength is determined from tests on (< 15) in situ drilled cores, according to clause 7.3.3 of EN 13791, the characteristic concrete compressive strength fck of the tested region may be determined using the lower value of:

and fck = fc;min + 4

(2)

where: fcm fc;min k

mean in situ compressive strength obtained from tested cores minimum value of in situ measured compressive strength of tested cores factor associated with number of test results k = 7 for 3 < n < 6 (n = number of cores tested) k = 6 for 7 < n < 9 k = 5 for 10 < n < 14

From the text in EN 13791 it appears that the method in the standard is primarily intended for situations in which there is doubt about whether the actual strength of the concrete as supplied complies with the requirement. In such situations it is important that the value of fck determined is neither too high nor too low; both the consumer’s and the producer’s risk play a role. Taerwe [9] and Caspeele and Taerwe [10] elaborate on this concept in depth. Notwithstanding, EN 13791 also states that the calculated value of the characteristic concrete strength may be used for redesign. However, proving with a certain confidence that the concrete provided meets the requirements is essentially different from proving the safety level of a structure based on actual measurements of the strength. Both assessments require fundamentally different strength parameters. The following example illustrates that EN 13791 is not fit for use in structural safety issues. Assume n = 6 cores (100 mm dia., 200 mm long) where the following concrete compressive strength values fc [N/mm2] have been measured: 80.0, 83.5, 77.4, 92.9, 67.1 and 92.4. The mean value is fcm = 82.2 N/mm2.

(1)

fck = fcm < k

= 82.2 – 7 = 75.2 N/mm2 Eq. (1) gives: fck = fcm - k Eq. (2) gives: fck = fc;min + 4 = 67.1 + 4 = 71.1 N/mm2

* Corresponding author: raphael.steenbergen@tno.nl Submitted for review: 27 June 2011 Revised: 17 October 2011 Accepted for publication: 27 November 2011

According to EN 13791, the characteristic concrete compressive strength is the lower of these two values, i.e. 71.1 N/mm2. Based on the core dimensions, this value is equivalent to the cylinder strength. To translate this value

R.D.J.M. Steenbergen/A.H.J.M. Vervuurt · Determining the in situ concrete strength of existing structures for assessing their structural safety

Table 1. Values for the student-t distribution (v = n-1 and p = 0.05)

v = n-1

⬁

tn-1

6.31

2.92

2.35

2.13

2.02

1.94

1.89

1.86

1.83

1.81

1.72

1.70

1.64

back into a strength class value, according to EN 13791, it should be divided by 0.85, thus increasing it by 18 %. This is based on two notes in table 1 of EN 13791. Note 1 states implicitly that the in situ compressive strength may also not be less than that measured on standard specimens. In that case the 18 % increase is certainly unsafe. It is questionable whether note 2, which states that the 0.85 figure is part of γc in EN 1992-1-1, is true; it is not in agreement with the fib Model Code 2010 [11] and the work done by fib SAG 9 [12]. For the shear rules, note 2 in table 1 of EN 13791 is certainly not the case because these rules are calibrated on the concrete compressive strength of tested beams to result in the required value of the reliability index in combination with γc = 1.5. Thus, the shear strength in EN 1992-1-1 is based on the assumption that the characteristic concrete compressive strength in the structure equals the value of the strength class and not 15 % less. In the example above, in line with EN 13791, the measured cylinder strength of 71.1 N/mm2 is translated by interpolation into compressive strength class C83/98 according to table 1 of EN 13791. The strength class value of 83 N/mm2 (cylinder strength) is thus even higher than the mean value of the cores tested. This is not acceptable for an assessment of structural safety. EN 1990 provides the background to assessing the strength for a structural safety calculation and does not comply with EN 13791, as will be seen in the next section.

Determining a characteristic strength according to EN 1990

Eurocode EN 1990 [2] establishes principles and requirements for the safety, serviceability and durability of structures. To do that, EN 1990 describes the basis for their design and verification and gives guidelines for related aspects of structural reliability. When determining concrete compressive strength based on tests, clause 5.2 of EN 1990 presents a general method that can be applied to determine a reliable value for the characteristic concrete compressive strength based on in situ drilled cores. In the method, the statistical uncertainty depends on the number of cores tested. In order to obtain the characteristic concrete compressive strength according to EN 1990, the strength is assumed to have a log-normal distribution (see [3] and [4]). The characteristic concrete compressive strength (defined as that strength value with a probability of non-exceedance of 5 %) follows from: ¨« 1 ¬« fck = exp fcm (Y ) u exp ©<tn<1, p=0.05 u s(Y ) u 1 + n «® ª«

{

}

(3)

where: fck in situ concrete compressive characteristic strength

Structural Concrete 13 (2012), No. 1

fcm(Y) mean of natural logarithm of in situ measured strength values n number of cores tested s(Y) standard deviation of natural logarithm of measured in situ strength values tn–1 value of tn–1 according to student-t distribution (Table 1) Eq. (3) is based on a Bayesian update procedure in which a vague prior distribution is assumed for both mean and standard deviation. Each time a new set of cores is analysed, the mean value, but especially the standard deviation, can differ from the values belonging to the previous set. Annex D of EN 1990 takes into account this statistical uncertainty, see Eq. (3). In this equation, the uncertainty in the standard deviation is incorporated by using the student-t distribution. The factor 兹苵 (1+ 苵苵1苵/n) 苵苵苵 takes into account the uncertainty in the mean value. As can been seen from Eq. (3), the statistical uncertainty decreases as the number n of cores tested increases. EN 13791 indicates that the calculated value of the characteristic concrete strength calculated according to this code may be used for redesign. However, the differences between this standard and EN 1990 are considerable. For example, in the case of n = 5, EN 13791 prescribes a factor k = 7. Combining with the value tn–1 = 2.13 from the student-t distribution in Table 1 results in a standard deviation for the concrete compressive strength of 3.3 N/mm2. This value is unrealistically low, given values in, for example, fib Model Code [5] and Rackwitz [4]. Similar results are obtained for other numbers of tested cores. EN 13791 therefore leads to higher characteristic values when compared with the values derived with the Eurocodes. The use of values based on EN 13791 may lead to lower levels of safety than intended by the Eurocodes. Using EN 13791 for assessing the structural safety of concrete structures is therefore not recommended. In the UK, complementary guidance to EN 13791 is given in BS 6089:2010 [6]. This standard specifies a method for determining the characteristic in situ strength which is more in line with EN 1990.

Determining the characteristic concrete strength for existing structures using a modified EN 1990 method

In section 2, the EN 1990 method was used to determine the characteristic concrete strength. That method includes the statistical uncertainty in the mean and standard deviation obtained from the cores tested. However, the following phenomenon also has to be taken into consideration. Incidentally, the standard deviation of the strength of the cores tested may be very low and therefore the characteristic strength of the structure may become quite large. This could also occur when all cores tested are obtained

R.D.J.M. Steenbergen/A.H.J.M. Vervuurt · Determining the in situ concrete strength of existing structures for assessing their structural safety

from one batch or pour, whereas an entire structure mostly consists of several concrete pours. In assessing the strength of an existing structure, we are interested in the strength of the section that is decisive for the stress distribution. Firstly, this section is not exactly known beforehand. Secondly, concrete cores are never drilled from that section; they are drilled from a more or less unimportant edge of the structure. For this reason, a translation has to be made between the statistical distribution of the measured strength of cores somewhere in the structure and the statistical distribution of the strength of the section governing the structural analysis. An accidentally small standard deviation obtained from tests on cores often taken from an edge of a structure may therefore be not representative for the governing section, especially if that low standard deviation is much smaller than the average standard deviation obtained from many experiments on similar types of concrete structures. For this reason, in order to establish a safe method, a minimum standard deviation smin is introduced which is to be applied if the measured standard deviation turns out to be less than this smin. This minimum standard deviation is considered to be a fixed property of the population considered; it is a safe value established by expert judgment. No statistical uncertainty is involved because the value of smin is only used in case a real, measured standard deviation is judged to be too small for safety reasons. In the proposed new method, the characteristic value of the concrete compressive strength is determined by the minimum of two values: A) the strength based on the measured test results, including the statistical uncertainty, and B) the strength based on the mean values of the measured test results and a minimum, constant standard deviation smin. This may be considered as a safe approach because very small, measured standard deviations are fixed at a larger smin. Based on a log-normal strength distribution, the characteristic concrete compressive strength is then obtained from the lowest value of: method A: ¨« 1 ¬« fck = exp fcm (Y ) u exp ©<tn<1, p=0.05 u s(Y ) u 1 + n ®« ª« and

{

}

(4)

method B: ¨« 1 ¬« fck = exp fcm (Y ) u exp ©<1.64 u smin(Y ) u 1 + n «® ª«

{

}

(5)

Subsequently, the value for smin(Y) in the log-normal domain can be calculated from: 2¥ £ £ s ¥ smin(Y )= ln ²1 + ² min ´ ´ ² ¤ fcm ¦ ´ ¤ ¦

(6)

The test results of cores taken from existing structures may be adopted for determining the value of the minimum standard deviation smin. Similar types of structures or structures that are produced identically (e.g. prefabricated

Fig. 1. Cumulative distribution of the standard deviation of the concrete compressive strength for two types of existing structure (types I and II)

Table 2. Minimum value of standard deviation (smin) for two types of structure and different probabilities of being exceeded

smin in N/mm2 related to probability of being exceeded type

80 %

50 %

35 %

20 %

10 %

beams versus in situ cast concrete) may serve as a solid basis for determining smin. This is illustrated in Fig. 1. The figure shows the cumulative distribution function for the standard deviation obtained from a number of existing structures of different types (type I = slab structures, type II = slender T-beams). It should be noted that, for practical use, a more generalized and internationally accepted value for smin is recommended. The value smin can be obtained based on the cumulative distribution function (Fig. 1) and a safe value for the probability of exceedance. According to Fig. 1, for structure types I and II, values for smin at different probabilities of exceedance are given in Table 2. In the literature, the standard deviation of the concrete compressive strength at t = 28 days is generally assumed to be constant at s = 5 N/mm2 (e.g. fib Model Code [5]). This is confirmed by numerous test results in concrete plants. From Fig. 1 it is concluded that a higher value for s is found for 80 % of the type I structures. For the type II structures, this figure rises to 99 %. Therefore, it is concluded that for existing structures, the standard deviation may be expected to be greater than generally assumed for new structures and a larger value for smin is suggested. An smin value equal to or less than the mean value in Fig. 1 would be appropriate. Applying an excessively large value would – unreasonably – cancel lower measured standard deviations of concrete strength. The fixed value smin is only needed to correct for possibly excessively low measured standard deviations of core strengths that are not taken from the governing section in the structure; for that

Structural Concrete 13 (2012), No. 1

R.D.J.M. Steenbergen/A.H.J.M. Vervuurt · Determining the in situ concrete strength of existing structures for assessing their structural safety

reason smin should be chosen belonging to, at most, a 50 % probability of exceedance according to Table 2. The observed difference between the standard deviation for new structures (s = 5 N/mm2) and existing structures may have various causes: – The value s = 5 N/mm2 is obtained from specimens tested at 28 days. Hardening over time results in an increase in both the mean value of the concrete compressive strength as well as the standard deviation. – The standard deviation of s = 5 N/mm2 is based on single batches of one pour. In existing structures, several batches and pours are present when testing a limited number of cores from the structure. It may be expected that this leads to an increase in the standard deviation when the complete structure is considered. – As the curing conditions in a structure differ from place to place, it may be expected that the hardening process is different at various positions in the structure. – The value s = 5 N/mm2 generally does not take into account working conditions with respect to the casting of the concrete and the drilling of the cores. When the results of in situ drilled cores are used, these aspects are included implicitly. Finally, it should be noted that the value for the standard deviation of s = 5 N/mm2 was originally intended to obtain the mean value of the concrete compressive strength from the characteristic strength (e.g. fib Model Code [5]). In such a situation, a low value for the standard deviation leads to a safe approach. However, when a low value for the standard deviation is used for determining the characteristic concrete compressive strength, an unsafe value is obtained. If the strength is derived using the method described above, no further assessment is required when the structure complies with the requirements. However, if the structure does not comply with the requirements, additional research may still lead to approval of the structure. Such additional investigation may consist of: – A detailed analysis of the measured test results. Suggestions for such analyses are provided in British Standard BS 6089 [6]. – A detailed study of the group of structures to which the structure considered belongs. This might also result in adjusting some of the relevant parameters (e.g. smin). – Considering redistribution capacity with respect to the failure mechanism (strength reserves).

Applying the modified EN 1990 method

In this section, the modified EN 1990 method is applied to some test results. Assume an experiment of n = 6 cores and the following measured concrete compressive strength fc values: fc;1

fc;2

fc;3

fc;4

fc;5

fc;6

80.0

83.5

77.4

92.9

67.1

92.4

ln(fc;1)

ln(fc;2)

ln(fc;3)

ln(fc;4)

ln(fc;5)

ln(fc;6)

4.38

4.42

4.35

4.53

4.21

4.53

From the above values, the following values can be calculated: mean value of fc;1, fc;2, …, fc;6:

fcm = 82.2N/mm2

standard deviation of fc;1, fc;2, …, fc;6:

s = 9.8 N/mm2

mean value of ln(fc;1), ln(fc;2), …, ln(fc;6):

fcm(Y) = 4.40 (N/mm2)

standard deviation of ln(fc;1), ln(fc;2), …, ln(fc;6):

s(Y) = 0.122 (N/mm2)

value tn-1 according to the student-t distribution (p=0.05):

tn–1 = 2.02

If a value of 10 N/mm2 is taken for smin, the value of smin(Y) is calculated from: 2¥ £ £ smin ¥ ² ´ s (Y ) = 0.121 smin(Y ) = ln 1 + ² ² ¤ fcm ´¦ ´ min ¤ ¦

Subsequently, the characteristic concrete compressive strength is calculated from the minimum of: method A: ¨« 1 «¬ fck = exp fcm (Y ) u exp ©<tn<1, p=0.05 u s(Y ) u 1 + n ®« «ª

{

¨« 1 ¬« fck = exp 4.40 u exp ©<2.02 u 0.122 u 1 + = 62.6 N /mm2 6 ®« «ª

{ }

and method B: ¨« 1 «¬ fck = exp fcm (Y ) u exp ©<1.64 u smin(Y ) u 1 + n «® «ª

{

Structural Concrete 13 (2012), No. 1

}

¨« 1 ¬« fck = exp 4.40 u exp ©<1.64 u 0.121 u 1 + = 65.9 N /mm2 6 «® ª«

{ }

Based on the minimum of method A and B, the characteristic concrete compressive strength is determined from method A: fck = 62.6 N/mm2. As mentioned earlier, in the example above, smin is assumed to be 10 N/mm2. According to Table 2, for existing type I structures, this value is exceeded in 20 % of the structures tested. Therefore, it a safe value for smin. Nevertheless, for this particular example, method A remains dominant for the calculated value of fck.

}

Comparing EN 13791 and EN 1990

In the Netherlands, more than 600 cores have been obtained from more than 120 concrete structures ín recent

R.D.J.M. Steenbergen/A.H.J.M. Vervuurt · Determining the in situ concrete strength of existing structures for assessing their structural safety

characteristic concrete compressive strength calculated using EN 13791 is structurally higher than the value of fck calculated using the proposed method according to EN 1990. Based on the results of a large number of structures investigated, it is concluded that EN 13791 leads to a value for fck that is about 15 % too high. In some cases the strength class value obtained using EN 13791 is even higher than the mean value of the cores tested. EN 13791 may therefore be unsafe and not appropriate for safety assessments of concrete structures.

References Fig. 2. Comparison of concrete compressive strength according to EN 13791 and EN 1990 (smin = 10 N/mm2)

years ([7] and [8]). The cores have been tested for determining the concrete compressive strength. The effect of using EN 13791 rather than EN 1990 on the measured characteristic concrete compressive strength has been studied based on a subset of 70 concrete slab structures (type I in Fig. 1). Fig. 2 shows the results of this comparison. In the figure, the ID number of the structure is given on the horizontal axis, and fck is plotted on the vertical axis for both methods (EN 13791 and EN 1990). The ID numbers are sorted by order of magnitude of fck according to EN 1990. The results are based on smin = 10 N/mm2, which corresponds to a 20 % probability of exceedance for type I (see Table 2). This value is based on the experimental results and is assumed to be realistic for these types of structure. It can be seen from the figure that the strength as calculated using EN 13791 is higher for all structures when compared with EN 1990. The strength according to EN 13791 is about 15 % higher compared with EN 1990, in some cases even up to 30 % higher. For smin = 5 N/mm2, the difference is slightly less (about 12.5 %). Therefore, it is concluded that the difference is mainly dominated by the method adopted and less by the value of smin.

EN 13791, Assessment of in situ compressive strength in structures and precast concrete components, 2007. 2. EN 1990, Basis of Design, 2002 3. Faber, M., Vrouwenvelder, A. C. W. M.: Probabilistic Model Code, Joint Committee on Structural Safety, 2001. 4. Rackwitz, R.: Predictive Distribution of Strength under Control, Materials & Structures, 16, 94, 1983. 5. Walraven, J., et al.: fib Model Code, International Federation for Structural Concrete, 2010. 6. BS 6089; Assessment of in situ compressive strength in structures and precast concrete components. Complementary guidance to that given in BS EN 13791, 2010. 7. Steenbergen, R. D. J. M., Vervuurt, A. H. J. M.: Characteristic concrete strength based on compression tests using either vague prior information or prior information from rebound tests. 7th International Probabilistic Workshop, Delft, 2009. 8. Steenbergen, R. D. J. M., Vervuurt, A. H. J. M.: Method for the determination of the characteristic concrete compressive strength on the basis of drilled cores, TNO, 2010 9. Taerwe, L.: Evaluation of compound compliance criteria for concrete strength. Materials and Structures, 1988, pp. 13–20. 10. Caspeele, R., Taerwe, L.: Conformity control of concrete based on the “Concrete Family” concept, Beton- und Stahlbetonbau 103, 2008. 11. Model Code 2010, fib, Fédération Internationale du Béton, Lausanne 12. fib SAG 9, Revision of partial safety factors, Technical Report, 2010.

Conclusions

EN 13791 applies when assessing the in situ compressive strength of structures and precast concrete components. In practice, EN 13791 seems to apply to the determination of the strength when there is doubt about the concrete strength class. However, for assessing the structural safety of existing structures, the method given in EN 13791 is not applicable and leads to an unsafe approach. Therefore, for determining the concrete strength based on in situ drilled cores, it is suggested to use the methodology based on EN 1990 and the additional provisions presented in this paper. A comparison of the two methods shows that the

Dr.ir. R.D.J.M. Steenbergen TNO, PO Box 49, 2600 AA Delft, The Netherlands Tel: 0031 88 866 3423 raphael.steenbergen@tno.nl

Dr.ir. A.H.J.M. Vervuurt TNO, PO Box 49, 2600 AA Delft, The Netherlands Tel: 0031 88 866 3295 adri.vervuurt@tno.nl

Structural Concrete 13 (2012), No. 1

Articles Aurelio Muttoni Miguel Fernández Ruiz*

DOI: 10.1002/suco.201100032

The levels-of-approximation approach in MC 2010: application to punching shear provisions In order to address how new knowledge influences design expressions, design codes have in most cases become significantly more complex over the last decades. However, this tendency is leading to codes that are too complicated for preliminary design but still not sufficiently accurate for assessing existing structures (where even more realistic models of behaviour are sometimes required). An alternative code strategy is that proposed by codes based on a levels-of-approximation (LoA) approach. This approach is based on the use of theories based on physical parameters where the hypotheses for their application can be refined as the accuracy required increases. The approach proposes adopting safe hypotheses during the first stages of design, leading to relatively quick and simple analyses. In cases where such a degree of accuracy is not sufficient (e.g. design of complex structures, assessment of existing structures, significant potential economic savings), the hypotheses can be refined in a number of steps, leading to better estimates of the behaviour and strength of members. This approach, recently adopted in the first complete draft of Model Code 2010 for a number of design issues, is discussed within this paper with reference to punching shear provisions. Keywords: levels-of-approximation approach, design codes, Model Code 2010, assessment of structural safety, critical shear crack theory

Introduction

Most experienced engineers have always tackled the problem of designing new structures or assessing the strength of existing ones by following a levels-of-approximation (LoA) approach. This is rather intuitive as the limits to the strength of a structure can usually be calculated with fairly simple models. Provided that such models are based on sound theories, some of their physical parameters (e.g. angles of compression struts or effectiveness factors) can be better estimated by devoting more time to their analysis, leading to improved (typically higher) estimates of the strength of a member. However, this design strategy has not always been reflected in codes of practice, where in many instances it is not possible to refine the parameters used in their design expressions. This is typically the case with empirical for* Corresponding author: miguel.fernandezruiz@epfl.ch Submitted for review: 12 July 2011 Revised: 20 September 2011 Accepted for publication: 20 September 2011

mulas based on geometrical dimensions and material properties, but not on physical parameters. As a consequence, codes seldom overrule most aspects of design (which is usually time-consuming and leaves little opportunity for designers to use advanced state-of-the art design methods) or are excessively open (which might be dangerous in the hands of inexperienced designers). In order to address the influence of new knowledge on design expressions suitably, codes are also increasing in complexity (particularly when empirical models are used). This tendency is leading to codes that are sometimes too complicated for preliminary design but still not sufficiently accurate for assessing existing structures (where even more realistic models of behaviour are sometimes required). In many countries this is giving rise to a debate on the need for “concise codes” for designing simple structures and “assessment codes” for existing structures. However, their design models are not always consistent and this leads to confusion for designers. An alternative code strategy is that proposed by the LoA approach [1, 2], see Fig. 1. This approach proposes using theories based on physical models. When preliminary estimates of the strength of a member are required, the mechanical parameters of the design expressions can be assessed in a simple (yet safe) manner. This allows the limits of strength to be determined even though very little time needs to be devoted to the analyses, which is normally sufficient for preliminary design purposes and even for many structural members without a given governing failure mode. However, in cases where such a degree of accuracy is not sufficient (e.g. critical elements, detailed design), the values of the mechanical parameters can be refined in a number of steps. This means devoting more time to analyses, see Fig. 1, but leads to better estimates of

accuracy IV III II I

levels of approximation

time devoted to analysis

Fig. 1. Levels-of-approximation approach: accuracy of the estimate as a function of the time devoted to analyses (adapted from [1])

A. Muttoni/M. F. Ruiz · The levels-of-approximation approach in MC 2010: application to punching shear provisions

the behaviour and strength of members. The LoA approach thus allows the preliminary design phases as well as advanced designs and assessments to be covered with the same set of expressions. With respect to the design of new structures, this approach allows for a gradual increase in the accuracy of the analyses as the project evolves from preliminary design studies to a construction project. This helps to expend the necessary time at each design stage. It also allows a refined design to be carried out for unusual elements with special significance regarding the safety of the structure (e.g. discontinuity regions or coupling members). And the incremental approach is also very convenient when assessing existing structures which, even if they were correctly designed according to codes of practice at the time they were built, might not comply with current code recommendations due to changes in loads or more stringent code provisions. This does not mean, however, that such structures are unsafe. Design rules are provided to cover a series of uncertainties and to be applied to a wide number of cases, although they might be excessively conservative in some situations. In these cases, the use of more refined analysis methods to assess the structural safety is fully justified (even if they are more timeconsuming) as expensive strengthening can be avoided. The LoA approach was formally presented [1] and implemented for the calculation of second-order effects in the Swiss Code for structural concrete [3] in 2003. Recently, it has also been considered in the first complete draft of Model Code 2010 [4, 5] for shear, punching shear and buckling design. In this paper, the fundamentals of this approach are presented with reference to the punching shear provisions of Model Code 2010. A practical example of the use of this approach is also introduced, helping the reader to understand the increase in accuracy expected as higher levels of approximation are employed.

(a)

(b)

The LoA approach for punching shear in Model Code 2010

Punching shear has been a topic of research in structural concrete since the 1960s. The first rational approach to punching shear design was developed in Sweden by Kinnunen and Nylander [6]. This approach successfully explained the behaviour and strength of punching shear in flat slabs without transverse reinforcement. Although the approach of Kinnunen and Nylander was rather satisfying, it resulted in somewhat complicated design expressions. As a consequence, its implementation in codes of practice was difficult and currently most codes of practice both in Europe [7] and in America [8] are still based on empirical expressions without a physical basis for the punching shear design of members without transverse reinforcement. In order to provide a better understanding of the phenomenon, intensive research has been performed in recent decades. A detailed state of the art report and comparisons of approaches can be found in specialized publications [9, 10] and research works [11, 12]. Following these investigations, and contrary to previous editions of the Model Code, the punching shear provisions in Model Code 2010 are based on a physical theory rather than on empirical formulas. The theory behind MC

(c)

Fig. 2. Critical shear crack developing through the compression strut: (a) location of strut and critical shear crack [12], (b) failure envelopes for reinforced concrete slabs as a function of slab rotation (results for specimens with effective depth of 95–450 mm, flexural reinforcement ratio of 0.4–1.6 %, concrete strength of 15–60 MPa, aggregate size of 8–32 mm and column diameter of 100–200 mm), and (c) comparison of failure band and results from 99 punching shear tests [12]

2010 provisions is critical shear crack theory (CSCT). The basic principles of CSCT with respect to punching shear design were developed by Muttoni and Schwartz in 1991 [13] and were later refined and extended to shear design of one-way members by Muttoni [14]. A series of recent experimental and theoretical works have provided justifica-

Structural Concrete 13 (2012), No. 1

A. Muttoni/M. F. Ruiz · The levels-of-approximation approach in MC 2010: application to punching shear provisions

tion for its mechanical model [12, 15, 16] and have also extended its use to members failing in shear after developing plastic strains in the flexural reinforcement [17, 18], shearreinforced slabs [19, 20] and other topics. An extended summary of recent developments and applications can be found elsewhere [21].

2.1

Mechanical model of CSCT

Critical shear crack theory is based on the assumption that the shear strength in members without transverse reinforcement is governed by the width and roughness of a shear crack that develops due to the inclined compression strut carrying shear [12, 15], see Fig. 2a. The shear strength resulting from this assumption can be calculated by assuming two rigid bodies with kinematics at failure characterized by the rotation of the slab (developed in agreement with test measurements [16]). Assuming such kinematics, both tensile stresses and stresses due to aggregate interlock (relative slip between the lips of the crack) develop along the critical shear crack. The shear strength can thus be calculated by integrating both contributions (concrete in tension and aggregate interlock) along the failure surface (dowel action is neglected due to spalling of the concrete cover to the flexural reinforcement, Fig. 2a). Fig. 2b shows the results obtained by performing such integration, using the numerical approach detailed in [16] and for significant variations of the mechanical parameters implied. The plot is normalized on both axes to account for support region size, concrete compressive strength, member depth and aggregate size. It can be seen that the punching shear strength decreases as rotation and effective depth increase (leading to greater openings of the critical shear crack). This is logical because wider cracks reduce both the concrete in tension and aggregate interlock contributions. It is also interesting to note that failures occur in a well-defined and rather narrow band for all cases (Fig. 2b). Comparing the failure region with the results of 99 punching shear tests (whose data is detailed elsewhere [12]) is shown in Fig. 2c, showing a very satisfactory agreement. For design purposes, and taking into account the narrow width of the failure band, detailed calculation of the failure envelopes by integrating concrete in tension and aggregate interlock contributions is not usually necessary. For these cases, a simplified failure criterion was proposed by Muttoni [14, 12]. It assumes that the punching shear strength (traditionally correlated to the square root of the concrete compressive strength after the works of Moody et al. [22]) is a function of the width and roughness of a shear crack as justified by the previous mechanical model: VR = fc u f w, dg b0 u dv

(

)

(1)

where: VR b0

shear strength shear-resisting control perimeter (set at dv/2 of the edge of the support region assuming a uniform distribution of shear forces)

Structural Concrete 13 (2012), No. 1

Fig. 3. Comparison of failure region (Fig. 2c) and average and characteristic CSCT failure criteria

fc w dg

shear-resisting effective depth of member (distance between centroid of flexural reinforcement and the surface at which the slab is supported) compressive strength of concrete width of critical shear crack maximum size of aggregate (accounting for the roughness of the lips of the cracks)

In order to evaluate the width of the critical shear crack w, Muttoni and Schwartz [13] assumed it to be proportional to the slab rotation ψ multiplied by the effective depth of the member (see Fig. 2a): (2)

w |s u d

Based on these assumptions, the following failure criterion was proposed by Muttoni [14, 12] for members without shear links and assuming average values for the strength: VR b0 u dv u fc

3/4 s ud 1 + 15 dg0 + dg

(3)

where: dg0 d

reference aggregate size equal to 16 mm is to be introduced in [mm].

In Fig. 3 this equation is compared with the failure band calculated on the basis of the mechanical model, showing good agreement. For design purposes, a characteristic failure criterion has to be adopted (target 5 % fractile, refer to Muttoni [12] and Model Code 2010 [5]), see Fig. 3: VRd f b0 u dv u ck ac

1 ) 0.6 1.5 + 0.6 u s u d u kdg

(4)

where kdg is a coefficient accounting for the maximum aggregate size dg, whose value can be calculated as kdg = 48 [mm]/(16+dg).

A. Muttoni/M. F. Ruiz · The levels-of-approximation approach in MC 2010: application to punching shear provisions

(a)

Fig. 4. Calculation of failure point according to CSCT: intersection between failure criterion and load–rotation curve

2.2

Calculation of failure load

(b)

The punching shear strength of a slab without shear reinforcement can be directly calculated using the CSCT failure criterion. To do so, the intersection between the failure criterion and the actual behaviour of the slab (characterized by its load-rotation curve) has to be calculated, see Fig. 4. It should be noted that this procedure allows not only the calculation of the punching strength but also the estimation of the deformation capacity (rotation) at failure. This provides the designer with valuable information on the behaviour of the structure (e.g. ductility, brittleness). Moreover, the rotation (as an estimate of the shear crack opening) can be used to calculate the activation of the transverse reinforcement for shear-reinforced slabs [19] or to estimate how fibres contribute to punching shear strength [21], accounting for their softening behaviour.

2.3

Application to shear-reinforced slabs

The theory can also be consistently applied to shear-reinforced slabs. A number of potential failure modes can develop [19, 20] such as: punching within the shear-reinforced area, punching outside the shear-reinforced area, crushing of concrete struts, delamination of concrete core, shear reinforcement pull-out, flexural failures. Details of the way these failure modes can be treated within the frame of CSCT are investigated in depth elsewhere [19, 20, 21]. Of particular significance is the failure mode by punching within the shear-reinforced zone, see Fig. 5a. The strength in this case depends on the contributions of the concrete and the transverse reinforcement: VRd = VRd,c + VRd,s

(5)

This fact has been acknowledged by most design models. However, most codes of practice still propose empirical formulations for estimating the contributions of the two terms. For instance, a constant reduction in the concrete contribution with respect to the strength of members without shear reinforcement is provided for in EC-2 (25 %) [7] and ACI 318-08 [8] (50 %). These codes also give empirical formulas or constant values for the stress in the shear reinforcement. The CSCT approach is, however, rather different and takes advantage of the physical hypotheses of the theory. This approach can be understood with the help of Fig. 5a,

(c) Fig. 5. Slabs with transverse reinforcement: (a) activation of shear reinforcement by critical shear crack, (b) concrete and shear reinforcement contributions, and (c) sum of concrete and shear reinforcement contributions as a function of slab rotation

which shows that the transverse reinforcement is activated as the critical shear crack opens (Fig. 5b). This means that the stresses in the reinforcement increase until they eventually reach their yield strength. On the other hand, the concrete’s contribution to the strength decreases with the opening of the shear crack (Fig. 3). This is consistent with the assumptions of empirical codes but allows the calculation of a suitable reduction in the contribution of the concrete (VRd,c/VRd,c0 ratio in Fig. 5c) for each specific case on the basis of the rotation (deformation capacity) at failure. With respect to the activation of the shear reinforcement, suitable analytical laws have been derived elsewhere [19, 20] as a function of the slab rotation (correlated to the critical shear crack opening) and bond conditions of the reinforcement. A code-like formulation of these models [19, 26], accounting for bond and inclined reinforcement, has recently been introduced in the first complete draft of Model Code [5]. In its general formulation, this equation is as follows:

m swd =

£ Ess f d¥ u (sin _ + cos _ ) ² sin _ + bd ´)f 6 fywd qw ¦ ywd ¤

(6)

where: α fbd fywd φw

angle between slab axis and shear reinforcement value of bond strength (which for design purposes can be adopted as 3 MPa for ribbed bars) yield strength of shear reinforcement diameter of shear reinforcement

Structural Concrete 13 (2012), No. 1

A. Muttoni/M. F. Ruiz · The levels-of-approximation approach in MC 2010: application to punching shear provisions

2.4

Calculating the load-<rotation behaviour using an LoA approach

£1 e ¥ msd = VEd ² + u ´ ¤ 8 2bs ¦

Various methods can be used to estimate the load–rotation behaviour necessary to calculate the punching shear strength (Figs. 4 and 5c). Model Code 2010 proposes both simplified formulas derived on the basis of the analytical formulas [12] and numerical procedures. With respect to the analytical formulas, the general form of the load–rotation relationship proposed by MC 2010 [5] is: r f £m ¥ s = 1.5 u s yd u ² sd ´ d Es ¤ mRd ¦

1.5

fyd Es msd mRd

distance from column axis to line of contraflexure of radial bending moments yield strength of flexural reinforcement modulus of elasticity of flexural steel average moment per unit length for calculating flexural reinforcement in support strip average flexural strength per unit length in support strip

The values of the various mechanical parameters in the formula can be assessed with different degrees of accuracy, leading to the levels-of-approximation (LoA) approach.

2.4.1 LoA I For preliminary design purposes, a safe hypothesis can be adopted by assuming msd = mRd. This implies that, at failure, bending reinforcement yields over the entire width of the support strip, thus leading to large crack openings (which decreases the punching shear strength). This therefore leads to a safe estimate of the strength because if punching strength satisfies this condition, the strength of the slab will be governed by its bending capacity. Furthermore, slabs satisfying this condition exhibit a very ductile behaviour, avoiding brittle failure problems. The load-rotation equation is therefore:

s = 1.5 u

rs fyd d Es

(8)

where for regular flat slabs the value of rs can be calculated on the basis of the span length as rs ≈ 0.22 艎 (calculated based on the geometry of the structure).

2.4.2 LoA II Approximation level II is a simplified estimate of the moment acting in the support strip msd. This is carried out with an analytical expression relating the moment in the support strip to the shear force acting VEd and the moment transferred from the slab to the support region (characterized by its eccentricity eu). For instance, for inner columns of flat slabs:

Structural Concrete 13 (2012), No. 1

where: VEd/8

average moment for calculating flexural reinforcement acting in support strip without moment transfer VEd · eu moment transferred to column 2bs width where this transferred moment acts (half the moment acting on each side of the column)

(7)

where: rs

(9)

In spite of its simplicity, this expression provides excellent estimates of the load-rotation behaviour of a slab as shown elsewhere [12, 21]. For instance, the ratio between the measured and calculated punching shear strengths for the tests in Fig. 2c is 1.07, with a coefficient of variation of 9 % (significantly better than most design codes [12]).

2.4.3 LoA III If a linear-elastic analysis is performed for designing the flexural reinforcement in a flat slab, the resulting moment field can be used to improve the estimate of the mechanical parameters of Eq. (7). This is, for instance, simple for the values of rs and msd, where bending and torsion moments can be integrated directly for the latter. In this case, and due to the better estimate of the various parameters, coefficient 1.5 in Eq. (7) can be replaced by 1.2 (leading to stiffer behaviours and thus to higher strengths for equal mechanical parameters):

s = 1.2 u

rs fyd d Es

£m ¥ u ² sd ´ ¤ mRd ¦

1.5

(10)

2.4.4 LoA IV In some special cases, the load-rotation behaviour of a flat slab can be investigated by integrating the moment-curvature diagrams of the structure directly. This is justified, for instance, if expensive strengthening can be avoided by performing a more refined analysis. However, it should be borne in mind that such analyses are very time-consuming and that the accuracy of LoA III is already very satisfactory. Significant improvements in the strength by using this level should only be expected for slabs with fairly low reinforcement ratios over columns (with significant tensionstiffening effects) or when large redistributions of bending moments between column and mid-span regions are expected. Procedures for the numerical integration have been presented elsewhere [23, 24, 26]. Many commercial software packages also provide tools for such analyses. These methods are, however, generally sensitive to the choice of the parameters involved and should be applied by experienced users who have previously checked the results of the numerical simulations against actual tests.

A. Muttoni/M. F. Ruiz · The levels-of-approximation approach in MC 2010: application to punching shear provisions

(a)

(c)

(b)

Fig. 6. Example of application: (a) geometry and region investigated, (b) cross-section over column, and (c) governing load case

Choice of a suitable LoA

The choice of a suitable level of approximation depends mostly on the context of the analysis performed (preliminary or detailed calculations) and on the potential savings that can be achieved if a more refined level of approximation is performed. In general, this choice is a decision that belongs to the designer, but some objective criteria to guide this choice are provided below. LoA I provides simple and safe hypotheses for evaluating the physical parameters of design equations. It is quick and simple and thus is in most cases sufficient for preliminary design purposes. Another significant use of LoA I is to check whether a given failure mode cannot govern. This is the case for structures where sufficient strength is provided even under the safe assumptions of LoA I. In such cases, it is unnecessary to perform further analyses using more accurate levels of approximation. For more accurate levels of approximation, the physical parameters of design equations are evaluated through simplified analytical formulas. These levels are again quick and are usually sufficient to cover most design cases. Their use is advised for tender and detailed design of most new structures as well as for assessing existing structures. As the most refined level of approximation, numerical methods can be used to estimate the value of the physical parameters considered by the design equations (numerical integration of the moment-curvature diagrams of the structure). The use of such levels is typically very time-consuming and only advised for the de-

tailed design of very complex structures or for the assessment of critical existing structures. This is justified when a more accurate estimate can lead to significant savings for the client (avoiding or limiting the strengthening of structures).

Example of application

This section explains the use of the levels-of-approximation approach with the help of a practical example. It consists on the assessment of the strength of an existing flat slab built in the 1970s in Switzerland. The slab has a constant thickness of 0.52 m and is supported on a series of walls around its periphery and by two inner columns (with a diameter of 1.20 m at the slab support point), see Fig. 6a. Concrete compressive strength was updated by tests to a value of fck = 59.4 MPa (accounting for long-term effects) and the maximum aggregate size is 32 mm. The slab has large amounts of flexural reinforcement in the column region, with average ratios in the support strips of ρ = 1.7 % in the x direction and ρ = 1.4 % in the y direction. The reinforcement yield strength is fyd = 390 MPa. The slab also has a number of bent-up bars (16 sections of bars 16 mm diameter inclined at 45° and intersected by the conical punching failure surface). It is subjected to self-weight, earth cover and traffic loads. The geometry and governing load case for punching shear strength are shown in Figs. 6a and 6b. The structure was analysed and eventually strengthened following the guidelines of MC 2010. In this paper, only the evaluation of its punching shear strength will be discussed.

Structural Concrete 13 (2012), No. 1

A. Muttoni/M. F. Ruiz · The levels-of-approximation approach in MC 2010: application to punching shear provisions

(a)

(b)

Fig. 7. Shear field analysis: (a) shear field of flat slab for governing load case, and (b) distribution of shear forces along control perimeter

4.1. LoA I A preliminary estimate of the punching shear strength can be obtained using LoA I. The total shear acting on the control perimeter can be obtained from a linear analysis of the structure (column reaction minus forces acting within control perimeter) as VEd = 5.54 MN. The minimum strength corresponds in this case to the direction of maximum rotation (maximum span length), which can be calculated from Eq. (8):

s max = 1.5

rs fyd 0.22 u 11.39 390 = 1.5 = 1.63% d Es 0.45 200000

(11)

For estimating the strength, the shear-resisting control perimeter b0 has to be reduced with respect to the total perimeter available at a distance dv/2 from the border of the support region (designated the basic control perimeter b1 in Model Code 2010) to account for concentrations on the shear field. This can be done by applying the following relationship [5]: b0 = ke u b1

(12)

Approximated values for the coefficient of eccentricity ke are provided in Model Code 2010 [5]. However, as stated in the code, due to the presence of significant concentrated loads in the vicinity of the support region, larger shear concentrations than those considered by the values of Model Code 2010 can occur. Thus, instead of a value ke = 0.90 (corresponding to inner columns of flat slabs subjected to distributed loads according to Model Code 2010) a safer value ke = 0.80 will be adopted [25]. This choice will be discussed and refined below. The concrete and reinforcement contributions can thus be directly calculated for the governing rotation: VRd = VRd,c + VRd,s =

b0 u dv u fck / a c + 1.5 + 0.6s maxd u kdg

+ Asw u ke u m swd u sin _ = 1.63 + 1.87 = 3.50

(13)

where the shear reinforcement contribution can be calculated on the basis of Eq. (6), leading to σswd = 390 MPa (yield strength). In this case, other potential failure modes

Structural Concrete 13 (2012), No. 1

(crushing of concrete struts, punching at other perimeters) are not governing. The compliance factor (n = VRd/VEd) is then 0.63, which can be considered as very low, advising urgent measurements.

4.2

LoA II

A better estimate of the punching shear strength can be obtained by using the results of a linear-elastic analysis. Coefficient ke can be estimated on the basis of the shear field of the structure around the column region, see Fig. 7. This figure shows concentrations of the shear field near the concentrated loads to the left of the column (where heavier concentrated loads are applied). In such a case, according to Model Code 2010, the coefficient of eccentricity can be calculated as: ke =

Vd 1 u = 0.78 v perp,d,max b1

(14)

where vperp,d,max corresponds to the maximum value of the shear force component per unit length perpendicular to the basic control perimeter, see Fig. 7b. This value shows that the previous estimate (level I) for the coefficient of eccentricity was already rather good. In LoA II, the load–rotation curves are calculated based on Eq. (7) and estimating msd according to the expressions is provided by level II of MC 2010, msd = VEd (1/8 + eu/(2bs)). The parameters required, rsx (= 0.22 艎x = 2.51) and rsy (= 0.22 艎y = 2.22), can be obtained by considering the geometry. The results are plotted in Fig. 8a. In this figure, the concrete strength (VRd,c, estimated according to Eq. (4)), the contribution of the shear links (VRd,s, estimated according to Eq. (6)) and the load-rotation behaviour of the slab (estimated according to Eqs. (7) and (9)) are plotted as a function of the rotations in the x and y directions (maximum punching shear strength is not governing). The result yields VRd = 4.22 MN, with the y direction governing. The compliance factor (n = VRd/VEd) is 0.76, still rather low but not so critical as in LoA I. It can be seen that in this case the stress in the transverse reinforcement is σswd = 271 MPa, which is below the yield strength (contrary to the outcome of LoA I). This is

A. Muttoni/M. F. Ruiz · The levels-of-approximation approach in MC 2010: application to punching shear provisions

be seen that the structure will most probably not have sufficient strength (VRd < VEd) even if more refined levels of approximation are used (the strength of the slab is only higher than the action for very limited rotations). However, the analysis with higher levels of approximation can still be interesting for the optimization of the potential strengthening [20]. This was the case in this example, which justified more in-depth analyses by using higher LoAs.

4.3

(a)

Taking advantage of the linear analysis performed, values of msd and rs can be estimated in a more accurate manner from the moment field as described in MC 2010. The values of rs are 2.48 and 2.61 m for the x and y directions respectively (fairly in agreement with the estimates of LoA II). Integrating the bending moments leads to the result shown in Fig. 8b. The results provide a slightly higher strength prediction, although with a limited gain. The punching shear strength is VRd = 4.35 MN, leading to a compliance factor of n = 0.78.

4.4

(b)

(c)

Fig. 8. Results for the various levels of approximation: (a) LoA II, (b) LoA III, and (c) LoA IV

due to the fact that the rotation at failure (ψ = 0.64 %) is smaller than that estimated in LoA I. However, although the contribution of the shear reinforcement decreases, the total strength increases because the concrete contribution is larger for smaller rotations. From the plot of Fig. 8a, it can also

LoA IV

Finally, a refined analysis of the load-rotation behaviour was performed using a non-linear finite element model. The model was local (slab between mid-span axes) and consisted of 32 regions where the different reinforcement near the column region, the support strips and the field where considered. The results are shown in Fig. 8c, where it can be seen that the result is only a limited increase in the strength, with a value of VRd = 4.46 MN leading to a compliance factor of n = 0.80. This result is consistent with other works [12], showing that the MC 2010 expressions for LoA II and III are quite accurate for the flexural behaviour of slabs with large amounts of bending reinforcement and safer for slabs with low amounts of flexural reinforcement (where tension-stiffening effects can play a significant role).

4.5

LoA III

Comments on the results

The results obtained confirm that the estimate of the accuracy of the punching shear strength can be refined progressively by using the levels-of-approximation approach. This is possible because a better estimate of the physical parameters of the design model is provided in each subsequent level of approximation (LoA). LoA I and II can be performed in just a couple of minutes. LoA III required more than twice that time. LoA IV, as performed in this example, took some days. The gain in the estimated strength is, however, limited, and LoA II or III are usually sufficient for detailed designs and assessments. In this case (due to the complexity of the location of the flat slab) all levels were performed in order to optimize the amount of shear reinforcement and the method to be used during retrofitting.

Conclusions

This paper has discussed the main ideas of the levels-of-approximation (LoA) approach for designing and assessing

Structural Concrete 13 (2012), No. 1

A. Muttoni/M. F. Ruiz · The levels-of-approximation approach in MC 2010: application to punching shear provisions

structures. Its use in the new Model Code 2010 is also explained with reference to the punching shear chapter and an example of an application to a real structure is presented. The main conclusions of the paper are: 1. The LoA approach is based on the idea that consistent (physically sound) theories should be used for design. The various mechanical parameters used within these theories can be estimated using different degrees of accuracy. 2. When little work is devoted to the analysis (first levels of approximation), safe (yet realistic) values of the strength and behaviour of structural members should be provided by this approach. 3. The accuracy can be increased thereafter by performing additional analyses. This allows a better estimation of the physical parameters required by the design equations. 4. Such an approach is convenient for designing or assessing structures: a. With respect to design, it allows an increase in the accuracy of the analyses as the project evolves from a conceptual design to a construction project. b. With respect to the assessment of existing structures, it allows the refinement of some hypotheses adopted for design (typically safe and introduced to cover a broad range of cases). In these cases, devoting a significant amount of time to the analyses is certainly justified if expensive (and unnecessary) strengthening can be avoided. 5. The first complete draft of Model Code 2010 proposes following the LoA approach with respect to shear, punching shear and buckling. Consistent design models have been obtained, encouraging the extension of the LoA approach to other domains.

Acknowledgements The authors have implemented the ideas of the LoA approach into Model Code 2010 within the scope of fib Task Group 4.2. The authors would like to thank the other members of the editorial group of TG 4.2 (Evan Bentz, Stephen Foster and Viktor Sigrist) as well as Joost Walraven. The authors are also appreciative of the comments and suggestions of Karel Thoma and Bruno Zimmerli (i-beratung GmbH, Switzerland) for the example presented within this paper.

Notation Asw Es V VEd VR VRd VRd,c VRd,max VRd,s b0

cross-sectional area of shear reinforcement modulus of elasticity of reinforcement shear force design value of shear force acting punching shear strength design punching shear strength design concrete contribution to punching shear strength maximum punching shear strength design shear reinforcement contribution to punching shear strength shear-resisting control perimeter

Structural Concrete 13 (2012), No. 1

b1 bs dv d dg dg0 eu fbd fc fck fyd fywd kdg ke 艎 msd mRd n rs vperp,d,max w α

γc φw ρ σswd ψ ψmax

basic control perimeter strip width shear-resisting effective depth effective depth maximum diameter of aggregate reference aggregate size (16 mm) load eccentricity with respect to centroid of basic control perimeter design bond strength average compressive strength of concrete (cylinder) characteristic compressive strength of concrete (cylinder) design yield strength of flexural reinforcement design yield strength of shear reinforcement coefficient for aggregate size coefficient of eccentricity span length average moment per unit length (design of flexural reinforcement) in strip average flexural strength per unit length in support strip compliance factor (= VRd/VEd) distance between column and line of contraflexure of moments maximum shear force perpendicular to basic control perimeter critical shear crack opening angle between slab axis and shear reinforcement partial safety factor for concrete diameter of shear reinforcement flexural reinforcement ratio design stress in shear reinforcement rotation of slab outside column region maximum rotation of slab outside column region

Reference 1. Muttoni, A.: Introduction to SIA 262 code (in French: Introduction à la norme SIA 262), Documentation SIA, D 0182, Zürich, Switzerland, 2003, pp. 5–9. 2. Muttoni, A., Fernández Ruiz, M.: Design through an incremental approach: the Swiss experience, 2010 Joint IABSE-fib Conference, Dubrovnik, Croatia, 2010, p. 8. 3. SIA. Code 262 for Concrete Structures, Swiss Society of Engineers and Architects, Zürich, Switzerland, 2003, p. 94. 4. Fédération Internationale du Béton (fib), Model Code 2010 – First complete draft, fédération internationale du béton, Bulletin 55, Lausanne, Switzerland, 2010, vol. 1, p. 318. 5. Fédération Internationale du Béton (fib), Model Code 2010 – First complete draft, fédération internationale du béton, Bulletin 56, Lausanne, Switzerland, 2010, vol. 2, p. 312. 6. Kinnunen, S., Nylander, H.: Punching of Concrete Slabs Without Shear Reinforcement, Transactions of the Royal Institute of Technology, No. 158, Stockholm, Sweden, 1960, p. 112. 7. Eurocode 2, Design of concrete structures – Part 1-1: General rules and rules for buildings, CEN, EN 1992-1-1, Brussels, Belgium, 2004, p. 225. 8. ACI, Building Code Requirements for Structural Concrete (ACI 318-08) and Commentary (ACI 318R-08), American Concrete Institute, Farmington Hills, Mich., USA, 2008, p. 473.

A. Muttoni/M. F. Ruiz · The levels-of-approximation approach in MC 2010: application to punching shear provisions

9. fib TG 4.3, Punching of structural concrete slabs, Bulletin 12, fédération internationale du béton, Lausanne, Switzerland, 2001, p. 307. 10. Polak, M. A.: Punching Shear in Reinforced Concrete Slabs, American Concrete Institute, Special Publication SP-232, Farmington Hills, Mich., 2005, p. 301. 11. Hegger, J., Häusler, F., Ricker, M.: Critical review of the punching shear provisions according to Eurocode 2 (in German: Zur Durchstanzbemessung von Flachdecken nach Eurocode 2), Beton- und Stahlbetonbau, vol. 103, No. 2, 2008, pp. 93–102. 12. Muttoni, A.: Punching shear strength of reinforced concrete slabs without transverse reinforcement, ACI Structural Journal, vol. 105, No. 4, 2008, pp. 440–450. 13. Muttoni, A., Schwartz, J.: Behaviour of Beams and Punching in Slabs without Shear Reinforcement, IABSE Colloquium Stuttgart, vol. 62, IABSE, Zurich, Switzerland, 1991, pp. 703–708. 14. Muttoni, A.: Shear and punching strength of slabs without shear reinforcement, (in German, “Schubfestigkeit und Durchstanzen von Platten ohne Querkraftbewehrung”), Beton- und Stahlbetonbau, vol. 98, 2003, pp. 74–84. 15. Muttoni A., Fernández Ruiz, M.: Shear strength of members without transverse reinforcement as function of critical shear crack width, ACI Structural Journal, vol. 105, No. 2, 2008, pp. 163–172. 16. Guidotti, R.: Punching of flat slabs subjected to very large column loading (in French: Poinçonnement des planchersdalles avec colonnes superposées fortement sollicitées), PhD thesis, École Polytechnique Fédérale de Lausanne, Switzerland, 2010, p. 187. 17. Guandalini, S., Burdet, O., Muttoni, A.: Punching tests of slabs with low reinforcement ratios, ACI Structural Journal, vol. 106, No. 1, 2009, pp. 87–95 18. Vaz Rodrigues, R., Muttoni, A., Fernández Ruiz, M.: Influence of shear on the rotation capacity of R/C plastic hinges, American Concrete Institute, Structural Journal, vol. 107, No. 5, 2010, pp. 516–525. 19. Fernández Ruiz, M., Muttoni, A.: Applications of the critical shear crack theory to punching of R/C slabs with transverse reinforcement, ACI Structural Journal, vol. 106, No. 4, 2009, pp. 485–494. 20. Fernández Ruiz, M., Muttoni, A., Kunz, J.: Strengthening of flat slabs against punching shear using post-installed shear reinforcement, ACI Structural Journal, vol. 107, No. 4, 2010, pp. 434–442. 21. Muttoni, A., Fernández Ruiz, M.: MC2010: The Critical Shear Crack Theory as a mechanical model for punching shear design and its application to code provisions, fédération internationale du béton, Bulletin No. 57, 2010, pp. 31–60. 22. Moody, K. G., Viest, M., Elstner, R. C., Hognestad, E.: Shear Strength of Reinforced Concrete Beams – Part 1: Tests of

23.

24.

25.

26.

Simple Beams, ACI Journal, Proceedings vol. 51, No. 4, 1954, pp. 317–332. Muttoni, A. (ed.), Fernández Ruiz, M., Fürst, A., Guandalini, S., Hunkeler, F., Moser, K., Seiler, H.: Structural safety of parking garages (in French: Sécurité structurale des parkings couverts), Doc. D 0226 SIA, Société Suisse des ingénieurs et des architectes, Zurich, Switzerland, 2008, p. 105. Vaz Rodrigues, R.: Shear Strength of Reinforced Concrete Bridge Deck Slabs, Thèse EPFL, No. 3739, Lausanne, Switzerland, 2007, p. 289. Muttoni, A., Fernández Ruiz, M., Guandalini, S.: Punching of slab bridges (in French: Poinçonnement des ponts-dalles), 4. FBH / ASTRA – study conference „Neues aus der Brückenforschung“, Doc. D0223 SIA, Societé suisse des ingénieurs et architects, Zurich, Switzerland, 2007, pp. 85–94. Tassinari, L.: Asymmetric punching of R/C slabs with shear reinforcement (in French : Poinçonnement asymétrique des dalles en béton armé avec armature de poinçonnement), Thèse EPFL No. 5030, Lausanne, Switzerland, 2011, p. 197.

Prof. Dr. Aurelio Muttoni Ecole Polytechnique Fédérale de Lausanne – ENAC Station 18 Lausanne CH-1024 Switzerland

Dr. Miguel Fernández Ruiz Ecole Polytechnique Fédérale de Lausanne – ENAC Station 18 Lausanne CH-1024 Switzerland

Structural Concrete 13 (2012), No. 1

Articles Johann Kollegger Susanne Gmainer* Klaus Lehner Josef Simader

DOI: 10.1002/suco.201100027

Ultimate strength of curved strand tendons In a bridge construction process, where large concrete elements are rotated with the aid of strand lifting units (lowering of arch halves, balanced lift method), the tendons have to be deviated over saddles with small radii of curvature. Since information on the ultimate strength of curved strand tendons was not available at the time, a test programme was carried out in order to determine the ultimate strength as a function of radius of curvature (R = 0.5 m, R = 1.0 m, R = 3.0 m, straight tendons), number of strands (1 to 55) and type of strand (0.5’’ and 0.6’’). The results of the full scale experiments showed almost no decrease in efficiency with regard to single strand tendons and only moderate losses regarding multistrand tendons. Based on the experimental results and a biaxial failure criterion, an analytical model was developed which is able to predict the ultimate strength of curved tendons.

through 90° is to deviate the strand tendons, which act as tension ties in the final position connecting nodes B and BI, over saddles. Since the tendons should be located in the centre of the cross-section in the final position of the bridge girder, the radii of the saddles will be between approx. 0.5 m (see example in section 7) and 2.5 m for larger bridge projects.

Keywords: heavy lifting, balanced lift method, strand tendons, post-tensioning, radius of curvature, full scale tests

Rmin = 0.6 Fpk [ MN ] * 0.6 m

are permitted for tendons with loop anchorages, where it is implicitly assumed that only very small relative displacements between strands and ducts occur at the posi-

Introduction

The balanced lift method for building bridges [1, 2, 3, 4] requires joints that can be subjected to rotations of between approx. 60 and 150°. Depending on the size of the bridge project, tensile and compressive forces of up to 10 000 or 20 000 kN have to be transferred across the joints when the bridge girders are rotated from the original vertical to the final horizontal position. Two possible applications of the balanced lift method are shown in Fig. 1. For bridges with high piers, a design that involves lifting joint A is advantageous (Fig. 1a). For bridges with piers of small or medium height, an auxiliary pier is installed and node C is moved from the original high to the final low position with the aid of heavy lifting plant (Fig. 1b). The transfer of the compression forces at joints A and B is accomplished by steel plates and a connecting bolt, e.g. as used for lowering the arches of the Deza Viaduct [5], or by concrete roller bearings with steel plates on the outside [4]. The most economical solution for transferring the tensile force at node C while rotating the bridge girder * Corresponding author: susanne.gmainer@tuwien.ac.at Submitted for review: 30 May 2011 Revised: 18 October 2011 Accepted for publication: 27 November 2011

Minimum radii of curvature of post-tensioning tendons

Applicable minimum radii of curvature Rmin as recommended by the fib Model Code [6] are shown in Fig. 2 as a function of the characteristic ultimate load Fpk. The smallest values of (1)

(a)

(b) Fig. 1. Balanced lift method for building bridges: a) lifting node A for bridges with high piers, b) lowering node C for bridges with piers of small and medium height

J. Kollegger, S. Gmainer, K. Lehner, J. Simader · Ultimate strength of curved strand tendons

Fig. 3. Layout of curved and straight tendons Fig. 2. Comparison of minimum radii of curvature of post-tensioning tendons according to [6] and radii of curvature of ultimate strength tests

tion of the curvature, i.e. the curved part of the tendon acts as a fixed end anchorage. According to Fig. 2, minimum radii of curvature of saddles for external post-tensioning tendons are calculated with Rmin = 1.4 Fpk [ MN ] * 2.0 m

(2)

in cases where high-quality polyethylene tubes with adequate thickness for external cables are used. Quite large radii of curvature according to the formula Rmin = 2.8 Fpk [ MN ] * 2.5 m

(3)

are specified for internal post-tensioning tendons with corrugated steel strip sheaths or plastic ducts. Internal unbonded monostrand tendons with greased and sheathed strands are permitted with minimum radii of curvature of 2.0 and 2.5 m for 0.5 and 0.6 inch strands respectively. In the balanced lift method, relative displacements occur between tendons and saddles at node C. Therefore, the minimum radii of loop tendons cannot be applied. Since the recommended radii of curvature for tendons with relative displacements between tendon and saddle (external post-tensioning) or curved ducts (internal post-

tensioning) according to Fig. 2 are much larger than those required for applications for node C in the balanced lift method, an experimental test programme was set up in order to determine the ultimate strength of curved tendons.

Full-scale tests on curved tendons

The properties of the tendons for the tests are given in Table 1. Seven-wire prestressing strands with areas of 100 mm2 (0.5 inch) and 150 mm2 (0.6 inch) were used, with characteristics as indicated in Table 1. The six tendon configurations were tested for cable layouts with radii of curvature equal to 0.5 m, 1.0 m, 3.0 m and infinity (straight tendons). The radii of curvature of the test series and the corresponding breaking loads Fpm are compared with the recommended values in Fig. 2 (red points). The tendons were placed in a 1.5 m high reinforced concrete block measuring 4.5 × 4.5 m on plan. The recommendations of the ETA [7] regarding the placement of anchorages and additional reinforcement were considered. The final tendon layout can be seen in Fig. 3. A concrete quality of C 40/50 was selected due to the high radial pressures exerted on the concrete by the large tendons with small radii of curvature. A complete description of the experimental setup can be found in [4, 8]. Hydraulic post-tensioning jacks were used to stress the tendons from one end until the maximum force was reached. The jacks had been calibrated just before the

Table 1. Properties of strands and tendons for ultimate load tests

tendon

single strand Ap

fpm

fp0,01

fp0,2

Agt

Ap,tendon

Fpm

duct

tendon

[mm2]

[N/mm2]

[%]

[mm2]

[kN]

[mm]

5-1 5-19 5-55

100 100 100

1891 1891 1891

1353 1353 1353

1694 1694 1694

5,5 5,5 5,5

100 1900 5500

189 3593 10400

25/30 80/87 135/145

6-1 6-12 6-31

150 150 150

1937 1914 1914

1524 1512 1512

1789 1741 1741

4,6 6,0 6,0

150 1800 4650

291 3445 8900

25/30 75/82 120/127

Structural Concrete 13 (2012), No. 1

J. Kollegger, S. Gmainer, K. Lehner, J. Simader · Ultimate strength of curved strand tendons

Fig. 5. Force-elongation diagram for 5-55 tendons Fig. 4. Stressing a 6–31 tendon

tests were started. Force measurements in the tests relied on recording the hydraulic pressure and the nominal area of the jacks. Loads were applied in steps of 10 % of the breaking load Fpm according to Table 1 up to a load level of 50 %, and from then on in 5 % steps until the tendon failed. Fig. 4 shows a jack with 11 000 kN capacity being installed for stressing a 6-31 tendon. This photograph is included in the paper in order to demonstrate that the results of the experiments were obtained under site conditions, not laboratory conditions. It is known that the use of jack wedges will lead to premature cable failure at about 90 % of Fpm. Therefore, the jack wedges were removed and anchor heads with standard wedges according to [7] were installed at the rear end of the jacks as can be seen in Fig. 4. Fig. 5 shows the applied forces plotted against the measured elongations for the four tests with 5-55 tendons. The elongations were calculated by subtracting the measured wedge draw-ins at the fixed end anchorages from the displacements of the jack piston. The stress-strain diagram (Fig. 6) was obtained by dividing the applied forces by the area of 5500 mm2 and by dividing the elongations by the lengths of the tendons. The mean tendon lengths of the 5-55 tendons were 7.72 m (R = 3 m), 5.48 m (straight tendon), 4.58 m (R = 1 m) and 3.79 m (R = 0.5 m). The measured wedge draw-ins are shown in Fig. 7. A complete compilation of all test results can be found in [8]. After evaluating the stress-strain diagrams of all 24 tests, it can be stated that: – The ultimate strains of straight and curved single strand tendons were between 15 and 30 ‰. – The ultimate strains of curved multistrand tendons were about 10 ‰. – Ultimate strains of 15, 18, 33 and 11 ‰ were obtained for straight 5-19, 5-55, 6-12 and 6-31 tendons respectively. Fig. 8 compares the quotients of ultimate test loads Fpm,exp divided by the ultimate loads Fpm based on the strengths of the strands according to Table 1. It can be seen that there was practically no loss of strength for tendons 5-1 and 6-1. The tendon efficiency of the medium-sized tendons 5-19 and 6-12 dropped to about 90 % for R = 0.5 m. Larger

Structural Concrete 13 (2012), No. 1

Fig. 6. Stress-strain diagram for 5-55 tendons

Fig. 7. Force-wedge draw-in diagram for 5-55 tendons

losses as well as a stronger influence of the curvature were measured for the 5-55 and 6-31 tendons. After the ultimate load tests, the strand bundles were removed from the concrete block and the locations of the wire fractures recorded. The inspection of the strands distinguished between fracture at the stressing anchor head, at the beginning of the curved part close to the stressing anchorage, within the curved part and at the fixed end anchor head. The results of this inspection are given in Table 2.

J. Kollegger, S. Gmainer, K. Lehner, J. Simader · Ultimate strength of curved strand tendons

the strands could only be inspected after the ultimate load test and after removing the strands from the concrete block. The following observations can be made concerning the locations of the wire fractures: – Straight tendons and tendons with a radius of 3 m failed at the anchor heads with the exception of tendon 5-55 mentioned previously, where fractures were also detected in the curved part. – Single-strand tendons failed at the anchor heads, with one exception: tendon 6-1 (R = 0.5 m) experienced wire fractures at the beginning of the curved part. – All multistrand tendons with radii of 1.0 and 0.5 m showed wire fractures within the curved part or at the beginning of the tendon curvature. Several tendons also exhibited wire fractures at the stressing anchor head.

Fig. 8. Efficiency of tendons as a function of curvature 1/R

A few fractures were recorded at two locations, e.g. tendon 5-55 with a 3 m radius suffered wire fractures at the stressing anchor head and within the curved part. Obviously, it was not possible to record the location of the initial wire fracture and the sequence of fractures because

4 4.1

Table 2. Locations of wire fractures in tendons

stressing beginning within anchor of curved head curved part part

The wire fractures in the curved part of the tendons were all similar to the example shown in Fig. 9 for a strand of the 6-12 tendon with a radius of 0.5 m. The fracture plane was tilted by approx. 45° with respect to the wire axis. This indicates that the failure was initiated by a multi-axial stress state. For a uniaxial tensile stress state, some lateral constriction and a fracture plane normal to the wire axis could be expected.

fixed end anchor head

Stresses in curved tendons Axial stresses in strands

Tendon

R [m]

5-1

•

6-1

•

6-12

•

5-19

•

6-31

•

5-55

•

5-1

6-1

6-12

5-19

6-31

5-55

5-1

6-1

6-12

5-19

6-31

5-55

α [%] = 100 – 0.77 · Fpm [MN]

5-1

0,5

6-1

0,5

The factor α takes into account the influence of site conditions.

6-12

0,5

5-19

0,5

6-31

0,5

5-55

0,5

Tendons are designed in order to withstand large axial stresses in practical applications. For example, the axial stresses for internal post-tensioning correspond to approx. 75 % of the characteristic tensile strength fpk [6]. In static load tests according to ETAG 013 [9], the measured maximum load should not be less than 95 % of the actual strength Fpm. The measured efficiencies of the straight tendons, shown in Fig. 8, are plotted in Fig. 10 as a function of the actual strength Fpm. The efficiencies were around 100 % for single-strand tendons, 95.9 and 98.2 % for the medium-sized tendons, and 91.3 and 93.9 % for the large tendons. This trend of efficiency decreasing as tendon size increases can also be observed in static load tests on anchorages in the laboratory. An important difference between the tests described in this paper and laboratory tests according to [9] is that the multistrand tendons were not stressed with a perfectly parallel placement of the strands within the bundle because some crossovers of the strands most likely occurred at the anchorages. The decrease in efficiency α as a function of tendon size was determined by linear regression based on the six available tests on straight tendons (Fig. 10):

4.2

(4)

Axial stresses due to bending of strands

In curved tendons, additional stresses in the axial direction arise due to the bending moments in the strands.

Structural Concrete 13 (2012), No. 1

J. Kollegger, S. Gmainer, K. Lehner, J. Simader · Ultimate strength of curved strand tendons

Fig. 9. Wire fractures in a strand of 6-12 tendon with R = 0.5 m

Table 3. Elastic bending stresses fbend [N/mm2] in strands according to Eq. (6)

Fig. 10. Efficiency of straight tendons as a function of tendon strength Fpm

Based on the relationship between radius of curvature R and bending moment M, the axial bending stresses fbend can be calculated with 1 M = R Ep u I fbend = M u

(5) deq 2uI

Ep u I deq Ep u deq u = 2uI 2uR R

(6)

It can be seen from Eq. (6) that the bending stresses in a strand due to placing the strand in a curved duct depend linearly on the equivalent diameter deq of the strand. An equivalent diameter has to be used because the bending stiffness of a strand is much lower than the corresponding stiffness of a solid section. For example, the moment of inertia of a 0.6 inch, seven-wire strand (Ap = 150 mm2) with a central wire with a diameter 3 % larger than the diameter of the outer wires would be equal to 2032 mm4 using the formulas for a solid section. Through experiments [10], and assuming a modulus of elasticity of 195 000 N/mm2, an average moment of inertia of 270 mm4 was determined for such a strand. A moment of inertia of 270 mm4 can be calculated for an equivalent circular solid section with diameter deq = 8.6 mm. In [11], an equivalent diameter of 10 mm is proposed and in Germany [12], deq is assumed to be half of the nominal strand diameter, which is equal to 7.85 mm for the strand under consideration. For calculating the bending stresses in the strands, the accurate specification of the equivalent diameter deq is of minor importance compared with the assumptions that have to be made for calculating the stresses for a stressed multistrand tendon, where the relative movements of the strands with respect to each other are partially constrained and which is pulled over a saddle. An equivalent diameter deq of 50 mm and a modulus of elasticity of 165 000 N/mm2 for a stressed multistrand tendon that is pulled over a saddle is chosen in [11].

Structural Concrete 13 (2012), No. 1

Radius in [m]

single strand tendon deq = 8,6 mm Ep = 195 000 N/mm2

multistrand tendon deq = 50 mm Ep =165 000 N/mm2 according to [10]

3 1 0,5

280 839 1677

1375 4125 8250

The unfavourable situation where a straight part of a multistrand tendon is pulled through a curved part (comparable with the saddle situation described above) also arose during the experiments. The straight tendon adjacent to the curved part was moved into the curved region by the wedge draw-in (2–3 mm) at the fixed end anchorage and by the elongation (approx. 7–14 mm) in the 1 m straight section between fixed end anchorage and curved part (see Fig. 3). However, in the strands of the failed tendons, no wire breakages were found at the rear part of the curvature close to the fixed end anchorage. As indicated in Table 2, wire failures occurred at the anchorages, at the beginning of the curved part close to the stressing anchorage or within the curved part. It seems that bending stresses fbend in the tendons according to Eq. (6) and Table 3 were of minor importance in our ultimate load test. It can be assumed that the high elastic stresses, due to the bending of the strands, disappear under ultimate load conditions, which are characterized by plastic behaviour of the strands.

4.3

Axial stresses due to different strand lengths

Fig. 11 shows 31 strands (0.6 inch) in a 120/127 duct. The distance between the top and the bottom strands in Fig. 11 was measured to be 60 mm. If the tendon has a layout with a quarter circle as shown in Fig. 3, the same distance of ΔR = 60 mm will occur between the innermost and outermost strands, leading to a length difference of

6l =

1 u 2 u 6 R u / = 94.3 mm 4

(7)

which will be present in all 6-31 tendon layouts with 3, 1 and 0.5 m radius of curvature. The largest influence of the length difference on the ultimate load will be present for the shortest tendon with 0.5 m radius of curvature. For the 6-31 tendon with 0.5 m radius of curvature, the lengths of

J. Kollegger, S. Gmainer, K. Lehner, J. Simader · Ultimate strength of curved strand tendons

Determining the maximum transverse force of a single strand in a curved multistrand tendon is much more complicated because the strands on the outside (e.g. top strands in Fig. 11) press against the strands located inside the bundle (e.g. bottom strands in Fig. 11). A cable factor k is proposed in [13] based on numerical simulations. With the aid of the cable factor, the average transverse force per unit length for a single strand can be determined according to Weiher et al. [13] from pstrand =

Ptendon 1 u uk N R

(9)

where N is the number of strands in the tendon and the cable factor k is calculated using k=

the shortest and longest strands are equal to 3666.1 and 3760.4 mm respectively. Conservatively assuming an elastic-brittle Rankinetype material model for the strands (only for the purpose of estimating how the different lengths of the strands influence the ultimate load of the multistrand tendon), failure will occur when the tensile stress in the shortest strand is equal to fpm with an ultimate strain of 1914 N/mm2/195 000 N/mm2 = 0.00982. The elongation of the shortest strand at this point will amount to 0.00982 × 3666.1 mm = 36.0 mm, which would correspond to the displacement of the stressing anchorage fixed to the jack (Fig. 4). As this displacement of the stressing anchor head is the same for all strands, the strain in the longest strand will be equal to 36.0/3760.4 = 0.00957 and the stress will be 0.00957 × 195 000 = 1866 N/mm2. If for the sake of simplicity the mean stress in the tendon is taken as the average of the two extreme values for the shortest and longest strands, we get a value of (1914 + 1866) × 0.5 = 1890 N/mm2. The stress difference between fpm and the mean value is only 24 N/mm2, which corresponds to 1.25 % of the actual strand strength fpm. Even for the very unfavourable assumption of an elastic-brittle material behaviour, the influence of different strand lengths on the ultimate load is small. Using realistic material models, which account for the plastification of the strands under ultimate conditions, the influence of the different strand lengths on the ultimate load would turn out to be negligible.

Radial stresses due to transverse pressure

Large radial stresses occur in curved parts of axially stressed tendons. The transverse force per unit length of a curved tendon can be readily obtained from the axial force in the tendon Ptendon and the radius of curvature R: ptendon =

Ptendon R

(10)

where: dP nominal diameter of strand inner diameter of duct Di

Fig. 11. Section through tendon with 31 strands (0.6 inch)

4.4

(8)

An average radial stress in the strand is obtained by dividing the transverse force per unit length from Eq. (9) by the nominal strand diameter dp and by substituting k by the expression given in Eq. (10): fradial, av =

Ptendon 1 u R Di

(11)

It should be noted that the same average radial stress would be obtained by dividing the transverse force per unit length of the tendon as given in Eq. (8) by the inner diameter of the duct Di.

Analytical model for ultimate strength of curved tendons

In the previous section it could be shown that the axial stresses due to bending of the strands and the influence of the different lengths of the strands were of minor importance for the ultimate loads of the tests on curved tendons. Therefore, in developing an analytical model, only the axial stresses caused by the jack force and the radial stresses caused by the transverse pressure in the curved part will be considered. Failure has to be expected at the strand with the highest radial stress. Weiher et al. suggest [13] that the highest radial stress is equal to twice the average stress. Furthermore, in ducts made of corrugated steel strip sheaths, the actual support condition of the strands corresponds to point supports rather than to a line support. Therefore, the average radial stresses as given in Eq. (11) have to be multiplied by a factor β that takes into account the increase in stress for the strand with the highest radial stress and the fact that the strands are supported on the ribs of the duct only. It turned out that a factor β = 2.5 resulted in a good correlation between experimental data and analytical model. The radial stress of the innermost strand at the contact area between strand and duct is thus estimated as:

Structural Concrete 13 (2012), No. 1

J. Kollegger, S. Gmainer, K. Lehner, J. Simader · Ultimate strength of curved strand tendons

Fig. 12. Comparison of analytical and experimental efficiencies for 0.5 inch tendons

fradial = fradial, av u ` =

Ptendon 1 u u` R Di

Fig. 13. Comparison of analytical and experimental efficiencies for 0.6 inch tendons

(12)

The axial stress of the tendon is: faxial =

Ptendon Ap,tendon

(13)

The radial stress can be expressed as a function of the axial stress: (14)

fradial = faxial u C where: C = Ap,tendon u

1 1 u u` R Di

(15)

Resorting to a Von Mises yield criterion [14] for the biaxial stress state yields: 2 2 2 2 u( faxial + faxial u C + faxial · C 2 = fpm

_ 2 ) 100

Relevance of the analytical model for heavy lifting operations

Usually, a global safety factor approx. 2.5 with regard to the guaranteed ultimate tensile strength of the strands is applied in heavy lifting operations with strand lifting units [4, 5]. The expression given in Eq. (17) can be used to determine the axial failure stress of a curved multistrand cable if the strand cable is guided over a saddle during lifting or lowering operations. In real applications, the saddle with a smooth surface will offer more favourable support conditions for the innermost strands than the corrugated metal ducts used in the experiments described in this paper. Using a value of β = 2.5 will therefore be a safe assumption.

Example of an application of the analytical model

In a full-scale test of the balanced lift method [2], 12 tendons with 0.6 inch strands were deviated over saddles with

(16)

which finally gives an expression for the axial stress when failure of a curved tendon will occur: faxial =

fpm 1+ C

+ C2

_ 100

(17)

Taking α from Eq. (4), fpm from Table 1 and C according to Eq. (15), the failure loads of the curved tendons were calculated using Fpm,model = faxial u Ap,tendon

(18)

and compared with the relevant experimental test data in Figs. 12 and 13 for 0.5 and 0.6 inch tendons respectively. As already mentioned above, it was possible to achieve quite good agreement between experimental and calculated failure loads by selecting a value of 2.5 for β.

Structural Concrete 13 (2012), No. 1

Fig. 14. Full scale test of the balanced lift method with curved strand tendons at the top of the bridge girder

J. Kollegger, S. Gmainer, K. Lehner, J. Simader · Ultimate strength of curved strand tendons

Fig. 15. Close view of strand tendons and concrete saddle after rotating the bridge girders into the horizontal position

radii of curvature equal to 0.534 m at node C (Fig. 1b). During the lowering of the bridge girders (Fig. 14), the 12 strands served as tension ties connecting joints B and BI, where the bridge girders and compression struts were connected with each other. A detail of the saddles, which were produced by inserting Styrofoam sheets in the formwork, is shown in Fig. 15. This photograph was taken after rotating the bridge girders into their final position. In the design of the tension ties, the nominal strength of 1860 N/mm2 was reduced to 1813 N/mm2 by applying Eq. (17) with Di = 15.7 mm in order to account for the deviation of the tendons in the saddle area.

Acknowledgements Financial support from the Austrian Research Promotion Agency (FFG) and the Federal Ministry for Transport, Innovation & Technology (BMVIT) is gratefully acknowledged. The authors would also like to express their gratitude to VSL (Schweiz) AG and wbr Rohr- und Bauelemente GmbH for providing 5-55 anchor heads and corrugated metal duct samples in accordance with EN 523.

Notation Rmin R Δl dp deq Di Ap Ap,tendon M I Fpk Fpm Fpm,exp

minimum radius of curvature radius of curvature difference in length between innermost and outermost strands nominal diameter of strand equivalent diameter of strand inner diameter of duct area of strand area of tendon bending moment moment of inertia characteristic ultimate load on tendon actual ultimate load on tendon ultimate load on tendon in experiment

Fpm,model ultimate load on tendon according to analytical model Ptendon axial force in tendon ptendon transverse force on tendon per unit length pstrand transverse force on single strand per unit length Ep modulus of elasticity of strand fpk characteristic ultimate strength of strand fpm actual ultimate strength of strand fp0,01 0.01 % proof stress of strand fp0,2 0.2 % proof stress of strand Agt actual elongation of strand at maximum load [%] faxial axial stress in strand fradial,av average radial stress in strand fradial radial stress in strand α factor for reduction in efficiency β factor accounting for increase of transverse pressure k cable factor according to [13] Ν number of strands in one tendon C constant factor for a curved tendon References 1. Kollegger, J., Blail, S.: Balanced Lift Method for Bridge Construction. Structural Engineering International, 2008, pp. 283–289. 2. Kollegger, J., Gmainer, S., Wimmer, D.: Maintaining Balance: Full-Scale Test of Balanced Lift Method. Bridge Design and Engineering, No. 61, 2010, pp. 30–31. 3. Kollegger, J.: Tilt-lift Method for Erecting a Bridge. World Intellectual Property Organization, WO 2008/022359. 4. Gmainer, S.: Brückenklappverfahren – Untersuchungen zur Entwicklung eines praxistauglichen Bauverfahrens. PhD thesis, Vienna University of Technology, 2011. 5. Rucabado, R., Moreno, A., Calvo, M., Taboas, P.: Lowering Manoeuvre on Concrete Semi Arches for the Deza Viaduct in Spain. 3rd fib Congress, Washington, 2010. 6. Walraven, J. et al.: fib Model Code. 1st complete draft, vol. 1, 2010. 7. VSL Post-Tensioning System, European Technical Approval ETA-06/006, 31 Jul 2006. 8. Lehner, K.: Großversuche an umgelenkten Spanngliedern zur Weiterentwicklung des Brückenklappverfahrens, Master’s thesis, Vienna University of Technology, Oct 2010. 9. European Organization for Technical Approvals: Guideline for European Technical Approval of Post-Tensioning Kits for Prestressing of Structures. ETAG 013, Jun 2002. 10. Ambro, S. Z., Kollegger J.: Freie Spanngliedlage für Spannglieder mit nach-träglichem Verbund. Report 02/2005, Institute for Structural Engineering, Vienna University of Technology, 2005. 11. Cable Stays – Recommendation of French Interministerial Commission on Prestressing: Service d’Études Techniques des Routes et Autoroute, Jun 2002. 12. Rostasy, F., Holzenkämpfer D.: Auswirkungen der zulässigen Spannstahlspannungen von EC 2, Teil 1 auf die Zulassung von Spannverfahren. Research Report, IBMB, TU Braunschweig, 1994. 13. Weiher, H., Specht, E., Pfeiffer, B., Klamroth, K., Zilch, K.: Determination of the Cable Factor for Deviated Tendon Bundles. Structural Engineering International, 2008, pp. 88–94. 14. Mehlhorn, G., Kollegger, J.: Anwendung der Finite Elemente Methode im Stahlbetonbau. Der Ingenieurbau, 1995.

Structural Concrete 13 (2012), No. 1

J. Kollegger, S. Gmainer, K. Lehner, J. Simader · Ultimate strength of curved strand tendons

Prof. Dr.-Ing. Johann Kollegger TU Vienna – Institute for Structural Engineering Karlsplatz 13/212-2, Vienna 1040 Austria johann.kollegger@tuwien.ac.at

Dipl.-Ing. Dr. Susanne Gmainer TU Vienna – Institute for Structural Engineering Karlsplatz 13/212-2, Vienna 1040 Austria susanne.gmainer@tuwien.ac.at

Dipl.-Ing. Klaus Lehner TU Vienna – Institute for Structural Engineering Karlsplatz 13/212-2, Vienna 1040 Austria betonbau@tuwien.ac.at

Dipl.-Ing. Josef Simader Grund- Pfahl- und Sonderbau GmbH Industriestraße 27a, Himberg 2325 Austria josef.simader@gps-bau.com

Structural Concrete 13 (2012), No. 1

Articles Carl-Alexander Graubner Erik Boska Christoph Motzko Tilo Proske Frank Dehn*

DOI: 10.1002/suco.201100012

Formwork pressure induced by highly flowable concretes – design approach and transfer into practice An analytical model for the calculation of the pressure of concrete on vertical formwork has been developed on the basis of experimental tests on highly flowable concretes in the fresh state (see companion paper “Material investigations and large-scale tests” [1]). The model takes into account the time-dependent material parameters of the fresh concrete, the specific properties of the highly flowable vibrated concretes and self-compacting concretes (SCC) as well as operational aspects. A proposal for the design of formwork was developed based on the experimental tests and the semi-probabilistic safety concept. It was found that the design load is often lower than the hydrostatic concrete pressure – even for highly flowable concretes. On construction sites, the pressure can be best controlled by limiting the casting rate. Insufficient experience of personnel in the handling of highly flowable concretes increases the uncertainties with respect to the design values and the safety risk. Further investigations into construction management-related aspects regarding the use of highly flowable concretes cover the risk assessment during concreting, design of the processes on the construction site and the development of the basis for a documentation system. Keywords: formwork pressure, friction, fresh concrete properties, setting

Introduction

The construction industry is using more and more concretes with a higher workability as well as self-compacting concretes (SCC) in order to build slender and heavily reinforced concrete building elements with sufficient quality. The use of effective superplasticizers and the rheological properties attained have called into question the existing knowledge about the pressure of fresh concrete based on normal vibrated concretes and existing design concepts such as DIN 18218:1980-09 [2] and ACI 347-04 [3]. Furthermore, the reduction in or absence of vibration when using SCC generally invalidates the existing models. The topic of fresh concrete pressure has been very intensively discussed recently. A number of measurements on SCC in the laboratory and on site (e.g. Billberg & Österberg [4], Assaad & Khayat [5], Gregori et al. [6],

* Corresponding author: dehn@mfpa-leipzig.de Submitted for review: 18 February 2011 Revised: 11 August 2011 Accepted for publication: 11 August 2011

Graubner & Proske [7]) have revealed new knowledge about the correlation between different influencing factors and the pressure on formwork. Different models [5–13] were developed based on the experimental tests, which often consider the material properties of the concrete merely under static conditions and neglect the loaddependent part of the yield strength (inner friction angle) as well as the construction operations aspects. Moreover, determining the model parameters in practice, e.g. the thixotropy or the friction parameters, is often problematic. Test facilities are not standardized, are comparatively expensive and the personnel of the concrete producer are usually not appropriately qualified. A state of the art report [14] gives an overview of the existing models for calculating the pressure of fresh concrete in general and identifies five categories of influencing parameters. These are: the fresh concrete properties, the formwork and reinforcement, the interface between concrete and formwork as well as concrete and reinforcement, processing and external influences. This publication presents the results of the research project “Schalungsbelastung durch Hochleistungsbetone mit fließfähiger Konsistenz” (pressure of fresh concrete asserted by highly flowable concretes) [15]. The studies regarding the design model for pressure of fresh concrete are based on experimental tests, which are presented in the companion paper [1]. The workability of concretes is classified in DIN EN 206-1:2008-08 depending on the spread of the concrete a in the flow-table test. Vibrated concretes with consistency F5 (560 ≤ a ≤ 620 mm) and F6 (630 ≤ a < 700 mm) as well as SCC (a ≥ 700 mm) were investigated. By including the results of the large-scale tests and measurements documented in the technical literature, it was possible to develop a proposal for the practical calculation of the pressure of fresh concrete on formwork. It tries to encompass both the reality and practicability to a large extent and considers the requirements for the safety and reliability of temporary structures. The new German standard DIN 18218:2010-01 [16] for the design of formwork was recently issued and includes a simplified concept for the realistic calculation of the pressure on formwork based on the studies presented. When using highly flowable concretes, the implementation processes of the structural work change considerably compared with the use of concretes of consistency classes F1 to F4. In particular, due to the changed structure of pro-

C.-A. Graubner/E. Boska/Ch. Motzko/T. Proske/F. Dehn · Formwork pressure induced by highly flowable concretes – design approach and transfer into practice

cedures and the requirements regarding safety at work, rearrangement of the working systems in the on-site realization becomes clear. Owing to this initial situation, construction management-related aspects were also examined within the scope of the aforementioned research project. The objectives were to develop practical assistance for the creation of a risk assessment during concreting, to draft recommendations for organizing the processes on the construction site, with a focus on the concreting work, to develop the basis for a documentation system for the concreting process and to select criteria for suitable formwork.

Modelling the pressure of fresh concrete

The typical development of the pressure of fresh concrete on vertical formwork during the casting of a member is shown in Fig. 1. The casting rate is assumed to be constant and the concrete production continuous. For the design of vertical formwork using two opposite formwork panels, the maximum pressure of fresh concrete is of particular interest. If the formwork panels are supported on only one side of the element without formwork ties, the pressure distribution over the height of the formwork is also of interest. Initially, hydrostatic pressure σh,hydro must be assumed in section H due to the low inner friction of the fresh concrete plus the dynamic impact during casting. The chemical-physical processes in the fresh concrete increase the internal friction of the concrete. This friction can be characterized by the yield strength of the concrete, which depends on the total vertical stress in addition to the age of the concrete. The material parameters of the inner friction are – in a similar way to soil mechanics – the cohesion and the angle of internal friction. These parameters control the ratio between the horizontal and vertical pressure in the formwork and hence the maximum formwork pressure. The increase in the pressure slows down as casting progresses and with the concrete height h. The maximum pressure σh,max is located at height hmax. The subsequent decrease in the pressure of fresh concrete is caused by the structural build-up of the concrete as well as the decrease in the pore water content due to the thixotropic build-up at rest and the onset of hydration. A reduction in the inner strain condition (the strain is

Fig. 1. Typical distribution of pressure of fresh concrete during casting

Structural Concrete 13 (2012), No. 1

a result of the elastic deformation of the formwork caused by the maximum pressure) must be realized through shrinkage of the concrete, a horizontal deformation or by releasing the formwork [8]. Once the concrete is in the state of final setting (with setting time tE), no support by the formwork is necessary anymore. This behaviour was confirmed in the present research project [7], using stiff and soft formwork in the large-scale tests. In general, for the design of formwork and falsework, only the loads in the section hE = v ⋅ tE need to be considered because of the ductile behaviour of the construction. The development of the concrete level during the concreting implies different load cases. However, the maximum pressure σh,max is of primary importance. In the following, a new analytical model for the calculation of the maximum pressure of fresh concrete σh,max asserted by highly flowable concretes will be presented. Assuming a constant casting rate v, the normalized maximum pressure of fresh concrete σ–h,E,max is:

m h, E,max =

m h,max v · tE · l c u g

m h,max v · tE · a c

m h,max m h,max (1) = hE · a c m h, E,hydro

The value σ–h,E,max represents the maximal horizontal pressure of fresh concrete σh,max divided by the maximum hydrostatic pressure at height hE = v ⋅ tE using the specific concrete weight γc = ρc ⋅ g, with density ρc and gravity constant g. The setting time tE is based on the Vicat penetration test according to DIN EN 480-2:2006-11 and is equal to the initial setting time according to ASTM C403/C403M-05. A more practical alternative is the setting time tE,KB according to the setting-bag test in DIN 18218:2010-01 [16]. The final setting state occurs when the indentation of a thumb in the concrete is < 1.0 mm (with a force of 50 N), which corresponds to a compressive strength < 50 kPa. The correlation between tE,KB and tE is: tE 5 1.25 u tE, KB

(2)

If the value σ–h,E,max is given, the maximum pressure of fresh concrete can be calculated according to Eq. (3), with the casting rate v and the setting time tE:

C.-A. Graubner/E. Boska/Ch. Motzko/T. Proske/F. Dehn · Formwork pressure induced by highly flowable concretes – design approach and transfer into practice

Fig. 2. Development of the normalized maximum pressure σ–h,E,max asserted by highly flowable concretes

(3)

1 ) 1 hE in [m] hE

)

(4)

0.8 ) 1 hE in [m] hE

(5)

(

SCC:

m h, E,max = 0.16 +

(

F5 + F6 model SCC - samples, static

0.8

F6 - samples, static

0.7

F5 - samples, static SCC - walls

0.6

F6 - walls F5 - walls

0.5

SCC - literature

0.4

F6 - literature

0.3 0.2 0.1 0.0 0

hE = v ⋅ tE [m]

Fig. 3. Proposed normalized pressure and measured normalized pressure asserted by highly flowable concrete

Consistency classes F5 and F6:

m h, E,max = 0.18 +

SCC model

hydrostatic until tE (upper limit)

0.9

σh,E,max [-]

A proposal for σ–h,E,max was developed based on experimental tests on small concrete samples and full-scale tests (see [1]). For each consistency class, the small-scale tests showed that the value σ–h,E,max is to a large extent independent of the concrete design, assuming the same boundary conditions such as formwork stiffness and vibration [1]. Frequent vibrations have a significant influence on σ–h,E,max. Therefore, for each consistency, the value σ–h,E,max is not assumed to be constant in the model (if concreting from the top), but depends on the height hE = v ⋅ tE and consequently considers the influence of the compacting technology and external vibrations on the pressure of fresh concrete – see Fig. 2 and Eqs. (4) and (5):

1.0

normalised maximum pressure

m h, max = m h, E, max u v u tE u a c

)

The lower the value hE, the lower the distance between the position of the dynamic impact caused by the compacting technology (hv) and the position of the maximum pressure of fresh concrete (hmax), provided the vibration depth hv does not exceed 1 m. Because of the reduction in the inner friction resistance close to the maximum pressure, the normalized pressure of concrete σ–h,E,max increases as hE decreases. By contrast, a larger hE value will decrease the influence of vibration and hence reduce the value σ–h,E,max. For consistency classes F5 and F6, no significant differences between σ–h,E,max values were measured in the material test. Hence, the model was chosen similarly. If height hE is approx. 2.5 m, vibration must be assumed to exert a significant influence. In this case the value σ–h,E,max = 0.58, measured in the material test with direct vibration, is implemented in the model. If hE ≈ 4 m, a high

influence is still expected and σ–h,E,max = 0.41 is assumed in the model. The influence of compaction can be neglected if hE = 10 m. If the age of the concrete exceeds tE, the pressure of fresh concrete cannot be increased, even with strong and constant vibration. Therefore, the upper limit of the normalized pressure is σ–h,E,max = 1.0. For SCC, the influence of vibration – as a result of the casting process – on the pressure of fresh concrete is much lower compared with the vibrated concretes. Hence, the σ–h,E,max values of SCC were found to be lower – see Eq. (5). The friction between concrete and formwork or reinforcement can be reduced significantly by vibration. As concretes with consistencies F5 and F6 are compacted with intense vibration, the pressure-reducing influence of the silo effect is not considered explicitly in the analytical model. Moreover, for SCC, the risk of vibrations resulting from construction operations cannot be excluded. However, positive effects can be introduced in the calculation of the model (un)certainties, which are used for calibrating the design values. In Fig. 3, calculated σ–h,E,max values are compared with the results of different measurements based on the maximum pressure of concrete; in general,

Structural Concrete 13 (2012), No. 1

C.-A. Graubner/E. Boska/Ch. Motzko/T. Proske/F. Dehn · Formwork pressure induced by highly flowable concretes – design approach and transfer into practice

lowing variation in the operational process, e.g. a high local casting rate vlocal compared with the calculated mean casting rate v. This minimum value should be considered with 25 kN/m2 for concrete consistencies F5 and F6 and 20 kN/m2 for SCC, assuming for the latter a local impact down to a depth of 0.8 m. If vibration is excluded, Fig. 4 shows good correlation between the analytical model and the test results. The model is verified for a casting rate of up to 8 m/h, a casting height of up to 20 m and a setting time of between 5 and 20 h. Assuming, for example, a casting rate of 2 m/h and a setting time of 7 h, the calculated maximum formwork pressure for SCC is 76 kN/m2. The actual pressure on the formwork for narrower walls and columns containing dense reinforcement could be significantly lower than the model values due to the silo effect. An analytical model explicitly incorporating the friction between concrete, formwork and reinforcement is presented by Graubner & Proske [8, 11, 17]. Based on this model, a higher casting rate would be possible in some cases.

250 γc = 25 kN/m³

maximum concrete pressure σh,max [kN/m²]

225 200 175

hydrostatic until tE SCC model

150

F5 + F6

125

F5 + F6 model SCC - samples, static

SCC

F6 - samples, static

100

F5 - samples, static SCC - walls

F6 - walls 50

F5 - walls SCC - literature

F6 - literature 0 0

hE = v ⋅ tE [m]

Fig. 4. Pressure of fresh concrete σh,max according to the analytical models and measured pressure

there is a good fit. However, the test results are often considerably lower than the calculated values, especially if the loss of workability of the concretes was high (not appropriate for practical applications). It must be considered that some test results would be higher if the concrete level had increased further. That the model results were exceeded can be explained by unusually high vibration during the casting process. Fig. 4 presents the maximum horizontal pressure σh,max, inserting Eqs. (4) and (5) in Eq. (3). The pressure of fresh concrete varies linearly with the casting rate v and the setting time tE. Assuming γc = 25 kN/m3 leads to Eqs. (6) and (7):

The bilinear pressure distribution in line with DIN 18218: 1980-09 (applicable with DIN EN 12812:2004-09) was chosen for the design of the formwork and is presented in Fig. 5. Accordingly, the pressure of fresh concrete must be assumed to be hydrostatic until the maximum horizontal pressure σh,max and the respective height hs is reached. Further, the horizontal pressure is constant in the remaining section of hE = v ⋅ tE. Once the concrete reaches an age of tE, no pressure need be considered anymore. This simplified distribution has the advantage of applying the maximum pressure over a large distance. In reality, the position of the maximum pressure of fresh concrete hmax is influenced by a number of parameters, e.g. formwork stiffness or early shrinkage, and cannot be predicted exactly. A significantly lower pressure than σh,max occurs below the section of hmax. However, for the design of formwork, the “safe” bilinear distribution is advantageous, considering the variation in setting time tE, casting rate v and height hE (see Fig. 5). Assuming hydrostatic pressure up to hs, un-

consistency classes F5 and F6:

m h,max = 25 kN/m 2 + 4.5 u v u tE ) m h, E,hydro

(6)

SCC:

m h,max = 20 kN/m 2 + 4.0 u v u tE ) m h, E,hydro

Proposed formwork design approach

(7)

The hydrostatic concrete pressure σh,E,hydro = v ⋅ tE ⋅ γc is a theoretical upper limit value. However, for practical applications it should be considered as a minimum value, al-

hydrostatic concrete pressure

γF x

h = v ⋅ tE

hydrostatic concrete pressure mean value of the max. concrete pressure

fresh concrete

hmax

concrete level

hardened concrete

characteristic max. concrete pressure

σ hd,max = γF ⋅ σ hk,max σ h,max

σ hk,max

Fig. 5. Distribution of the horizontal pressure of fresh concrete on formwork

Structural Concrete 13 (2012), No. 1

design value of the max. concrete pressure

σh

C.-A. Graubner/E. Boska/Ch. Motzko/T. Proske/F. Dehn · Formwork pressure induced by highly flowable concretes – design approach and transfer into practice

(8)

m hd = m hk u a F

The characteristic value and the partial safety factor take into account the variation in the model parameters (e.g. casting rate v) and uncertainties in the calculation model and hence limit the probability of failure of the construction. Like the analytical model (approx. mean value), the distribution of the characteristic pressure σhk and the design value σhd is chosen to be bilinear (see Fig. 5). The characteristic value of the maximum pressure σhk,max was calculated using the full probabilistic method (level III) according to DIN 1055-100:2001-03, Appendix B, and the Monte Carlo method. The calculation of σhk,max was based on the analytical models (see companion paper [1]) as well as the large-scale tests. The partial safety factor for the load in the ultimate limit state was defined as γF = 1.5. The limit state function for the calibration of the characteristic pressure of fresh concrete was developed from the relation between impact E and resistance R. Compared with the collapse of a multi-storey building or a nuclear power plant, the negative effects of a form-

200 SCC-walls F6-walls

175

R2 (walls and literature) = 0.74

σh [kN/m²]

F5-walls

measured concrete pressure

certainties due to discontinuous casting rate and vibrations can be considered. The calculated pressure of concrete σh,max based on the bilinear distribution and Eqs. (6) and (7) is compared with measured values in Fig. 6. The good correlation between the test results and the calculations can be seen. The design of formwork and falsework must also fulfil requirements regarding the safety and reliability of the construction. According to the semi-probabilistic safety concept, the design value of the pressure of fresh concrete σhd can be calculated with Eq. (8). The characteristic value σhk must be multiplied by the partial safety coefficient γ F:

SCC-samples

150

F6-samples F5-samples

125

SCC-literature F6-literature

100

measured = calculated 75 50 25 0 0

100

125

150

175

200

σh according to the model [kN/m²]

Fig. 6. Correlation between the pressure of fresh concrete according to the analytical model and measured pressure

work collapse for persons and property are relatively moderate. Hence, the failure probability in the ultimate limit state was defined as Pf = 10−4. The basic variables of the load are summarized in Table 1. The statistical parameters were derived from the experimental tests and values found in the literature. The casting rate v has the biggest influence on the variation, with a variation coefficient of V = 0.25, assuming a low casting rate of 1 m/h and a standard derivation of 0.25 m/h. The reason for this is mainly insufficient control of the casting process. The variation coefficient will decrease at higher casting rates. For v = 2 m/h, a variation coefficient of V = 0.20 was assumed and gives a standard derivation of 0.4 m/h. A high variation is seen in the setting time tE. In particular, the variation of the mix constituents using high-performance additives increases the variation coefficient of tE. Furthermore, the uncertainties caused by the setting tests cannot be neglected. The variation coefficient for SCC is higher than

Table 1. Basic variables for calibrating the characteristic pressure of fresh concrete

Basic variable SCC, consistencies F6 and F5

Mean value m

Standard deviation S

Variation coefficient V [–]

Specific concrete weight Casting rate

γc v

[kN/m3]

Internal force variable

SCC Setting time of the concrete Model uncertainties Additional factor

[–]

23.5 1.0 2.0 1.0

0.5 0.25 0.40 0.10

0.021 0.25 0.20 0.10

tE θE u

[h] [–] [–]

7.0 0.77 1.15

1.75 0.21 0

0.25 0.27 0

Consistency F6 Setting time of the concrete Model uncertainties Additional factor

tE θE u

[h] [–] [–]

7.0 0.96 1.00

1.4 0.19 0

0.20 0.20 0

Consistency F5 Setting time of the concrete Model uncertainties Additional factor

tE θE u

[h] [–] [–]

7.0 0.75 1.10

1.4 0.10 0

0.20 0.14 0

[m/h]

Structural Concrete 13 (2012), No. 1

C.-A. Graubner/E. Boska/Ch. Motzko/T. Proske/F. Dehn · Formwork pressure induced by highly flowable concretes – design approach and transfer into practice 250

175

SCC F5

vibration depth 2 m, maximum 0.65 tE

SCC - vibrations excluded

150 125

SCC - samples, static

hydrostatic until tE

100

σh [kN/m²]

γc = 25 kN/m³ 225

characteristic value for the design σ hk,max

200

concrete pressure

250

γc = 25 kN/m³

F6 - samples, static F5 - samples, static

SCC - walls F6 - walls

F5 - walls SCC - literature

maximum concrete pressure

model value σ h,max

225

σ hk,max

F6 hydrostatic until tE

200

SCC σ hd,max

175

F6 SCC - design value, ULS

SCC

150

F6 - design value, ULS

125

SCC - characteristic value F6 - characteristic value

100

vibration depth 2 m, max. until 0.65tE

SCC - samples, vibration

F6 - samples, vibration + compaction SCC - walls, summarised

F6 - walls, summarised

F6 - literature 0

hE = v ⋅ tE [m]

Fig. 7. Characteristic pressure of fresh concrete for the design σhk,max, measured values and analytical model

Fig. 8. Design values of pressure of fresh concrete σhd,max and characteristic values σhk,max

for consistencies F5 and F6 because of the higher sensitivity of SCC. The model uncertainties θE were calculated based on the measured and calculated pressures and depending on the consistency class. Besides the basic variables, e.g. specific concrete weight γc and the calculation of the internal force variables, an additional factor u was introduced. This factor considers other uncertainties such as unintentional vibrations during casting, the geometry of the member or the mix design. The impact of strong and constant vibration is an exceptional load and not considered in the calibration initially. Eqs. (9) to (11) were determined for calculating the maximum characteristic pressure:

lower spread of tE. Using SCC and neglecting the influence of unintended vibrations, the pressure of SCC is lower than the load of the vibrated concrete. In this case the higher workability of SCC has less influence on the pressure of fresh concrete than the mechanical compaction. The experience on construction sites shows that a vibration depth of 2 m cannot be excluded. However, the material tests showed that a change in the concrete consistency due to vibration can be achieved up to a concrete age of approx. 0.65 ⋅ tE and a depth of 0.65 ⋅ hE. Fig. 7 shows that the pressures accounting for this additional requirement are mostly included in the characteristic equation. Further, how the pressure of fresh concrete is influenced by any impacts of the concrete skip or the pump hoses or continuous external vibration can be seen as special loads. Assuming a maximum failure probability of Pf = 10−2 in this situation, an additional factor u according to the probabilistic calculation is 1.65 for SCC (replacing u = 1.15). Hence, in case of a special load, a pressure increase of 45 % can be accepted. For consistencies F6 and F5, an increase of 40 and 50 % respectively is acceptable. For SCC and consistency F6, the design values of the maximum pressure σhd,max (calculated with γF = 1.5 and the mean values of the input parameters) are presented in Fig. 8. The test results with intense vibration do not exceed the design vales significantly. A formwork collapse is also unlikely in this case because the actual load must be compared with the formwork resistance, which is enhanced by the corresponding partial safety factor. Therefore, such special loads are considered adequately in the design model.

SCC:

(

)

(9)

(

)

(10)

(

)

(11)

m hk,max = hs u a c = 1 m + 0.26 u v u tE u a c consistency F6:

m hk,max = hs u a c = 1 m + 0.30 u v u tE u a c consistency F5:

m hk,max = hs u a c = 1 m + 0.24 u v u tE u a c

Assuming γc = 25 kN/m3 leads to Eqs. (12), (13) and (14): SCC:

m hk,max = 25 kN/m 2 + 6.5 u v u tE

(12)

consistency F6:

m hk,max = 25 kN/m 2 + 7.5 u v u tE

(13)

consistency F5:

m hk,max = 25 kN/m 2 + 6.0 u v u tE

(14)

According to the analytical model, the characteristic values for SCC are lower than for consistency F6. The values for consistency F5 are lower than those of both SCC and F6 because of the lower model uncertainties of F5 and the

Structural Concrete 13 (2012), No. 1

Construction management-related aspects

The construction management-related aspects were selectively directed at the subject of the cost-effectiveness regarding the execution of the construction work as well as towards the matter of safety at work. The obligations of employer and employee with regard to safety at work have changed decisively as a result of Directive 89/391/EEC [18], issued by the Council of the European Community, coming into force. Regarding the creation of a risk assessment, which in Germany is compulsory for all employers in accordance with cl. 5 of the Labour Protection Act (Arbeitsschutzgesetz, ArbSchG [19]) and which must be available on every construction

C.-A. Graubner/E. Boska/Ch. Motzko/T. Proske/F. Dehn · Formwork pressure induced by highly flowable concretes – design approach and transfer into practice

Fig. 9. Basic measured values for risk assessment during casting process

site, risk factors within the scope of building site investigations during the concreting process were recorded and assessed by the Institut für Baubetrieb (Institute of Construction Technologies & Management). The use of standard concretes with consistencies F1 to F4 and the use of highly flowable concretes were examined. The risk assessment was carried out in accordance with the proven procedure according to Nohl which is recommended by BG Bau (Employers’ Liability Insurance Association for the Construction Industry). Based on the results of building site investigations, integrated basic measured values of the risk evaluation for concreting were developed in a comparison of work systems in the area of standard concretes with consistencies F1 to F4 and highly flowable concretes, which are shown in Fig. 9. The arrows in Fig. 9 show the level of risk using workable concretes compared with conventional concretes. The results form a valuable basis for the creation of a risk assessment for construction sites whose individual steps can be taken from “Recommendations for the compilation of a risk assessment with the use of formwork” [20]. Apart from the aspects relevant to safety at work, the processes were examined in great detail with regard to their cost-effectiveness. The working time study for this purpose was carried out according to REFA [21] methods. Regarding the use of highly flowable concretes, it can be said that for actual concreting procedures, many construction companies currently have little or no data available from post-calculation analyses or working time studies [22]. Evaluation of the working time studies carried out allows a weak-point analysis of the concreting procedures to be made. What is particularly interesting here is the result when using SCC. The following weak points were determined: – Too many workers in the construction crew: for the concreting procedure, the building contractor used the

same amount of personnel required for vibrated concrete. The potential for reductions was not exploited. – Inadequate monitoring of the mean casting rate: for regulating the pressure of fresh concrete, the mean casting rate was not monitored. Casting breaks that might have been required during concreting were not heeded. – Unsatisfactory coordination of the so-called supply and consumption rhythm: the call-off order of the fresh concrete and the entire delivery structure were insufficiently coordinated with the concrete suppliers and there was also poor coordination during concreting itself. – Unsatisfactory formwork construction: in one case, the formwork construction (door block-out) was not designed to withstand the higher concrete pressure. This resulted in fresh concrete escaping from the formwork, which caused substantial damage (see Fig. 10).

Fig. 10. Damage on a construction site: the formwork was inadequately designed and constructed and so did not bear up to the pressure of fresh concrete!

Structural Concrete 13 (2012), No. 1

C.-A. Graubner/E. Boska/Ch. Motzko/T. Proske/F. Dehn · Formwork pressure induced by highly flowable concretes – design approach and transfer into practice

The construction management-related investigations have revealed a lack of data on construction site experience concerning the use of highly workable concrete. Among other things, this is the reason for the scatter of the basic variables for the calibration of the characteristic pressure of fresh concrete (see Table 1). In particular, the casting rate during the casting process is subject to high fluctuations. A precise control and regulation system (see Fig. 11) has the potential to minimize the scatter of the basic variables. These construction management measures contribute to reducing the rather high design value for fresh concrete pressure (see Fig. 8).

Fig. 11. Basic structure of a database system for documenting the casting process

6 SCC

hydrostatic until t E

125

5 F6

100

4 F3

tE = 5 h

25 consistency classification according to DIN EN 206-1:2001-07

hydrostatic concrete level hs [m]

maximum concrete pressure

σ hk,max [ kN/m²]

150

γc = 25 kN/m³

0 0

casting rate v [m/h]

Fig. 12. Maximum lateral pressure of fresh concrete σhk,max according to DIN 18218:2010-01; setting time of concrete: 5 h

Furthermore, in the context of the research project [15], the basic structure of a database system was developed for documenting the concreting procedure. A permanent target–actual comparison of different parameters should be possible here (see Fig. 11). The “mean casting rate” as well as “position of installation” values were revealed. The poly-sensory systems, e.g. image processing systems, already developed and currently being developed at the Institut für Baubetrieb show the potential for supplying the relevant actual data both reliably and promptly. The controlling elements implemented should send immediate corresponding warning messages to the construction site management in the case of any inadmissible deviations so that process control takes place in time.

Structural Concrete 13 (2012), No. 1

Pressure of fresh concrete to DIN 18218:2010-01

At the request of the construction industry, the standard DIN 18218 “Pressure of fresh concrete on vertical formwork” was improved by DIN committee NA 005-07-11 AA. The intention was to integrate concrete with consistencies F5, F6 and SCC into the standard. The provisions for consistency classes F1 to F4 were not modified significantly, compared with the former standard. The proposal presented here for the calculation of the pressure of fresh concrete was implemented in DIN 18218:2010-01 [16] mostly unchanged. In Fig. 12 the maximum pressure σhk,max according to DIN 18218:2010-01 is plotted against the mean casting rate for the final setting time tE = 5 h, γc = 25 kN/m3 and concrete placed from the top. The absolute minimum pressure is defined as 30 kN/m2. To determine the final setting time tE, the concrete must include all chemical additives. The temperature of the concrete sample may not exceed the expected concrete temperature on a construction site. For practical applications, determining the setting time with the settingbag test according to DIN 18218:2010-01 is recommended. Compared with the pressure for concretes with consistencies F1 to F4, the pressure for highly flowable concretes is considerably higher. In addition to the different rheological behaviour connected with the development of the inner friction, the higher sensitivity of the concrete and the consideration of higher uncertainties are responsible for this difference. If SCC is placed from below by pumping, the design pressure of fresh concrete must be assumed to fully hydrostatic, multiplied by γF = 1.5 according to DIN 18218:2010-01. The material tests showed at most 15 % higher values compared with the hydrostatic pressure. However, the committee decided to apply the relatively high factor γF = 1.5 to the hydrostatic pressure in order to consider all casting situations and processes with the same safety factor.

Concluding remarks

The pressure of highly flowable concretes is mainly influenced by the casting rate, the setting behaviour of the concrete, the specific concrete weight and the member height. In addition, the pressure of fresh concrete can be increased significantly by the dynamic impact of the concreting plant and external vibration as a result of the construction work. The friction between the concrete, the

C.-A. Graubner/E. Boska/Ch. Motzko/T. Proske/F. Dehn · Formwork pressure induced by highly flowable concretes – design approach and transfer into practice

formwork surface and reinforcement can reduce the pressure of fresh concrete. According to the proposal for the calculations, there is generally no need to calculate using the hydrostatic pressure, especially if the building element is comparatively tall. Vibrated concrete with consistency F6 applies the highest maximum pressure σhk,max of all consistencies. The most important parameter controlling the pressure of fresh concrete on construction sites is the casting rate. Furthermore, a reduction in the pressure of fresh concrete can be achieved by reducing the setting time. Considering the concrete technology, this parameter is mainly influenced by type of cement, water-cement ratio and chemical admixtures. It should be noted that the setting time cannot capture the development of the inner friction exactly and hence the pressure of fresh concrete. From the point of view of construction practice, it must be further said that the complex theme of risk assessment regarding the use of F1 to F6 consistencies and SCC has to be carried out more intensively, both by construction companies and on building sites. In the context of the research projects, new insights were gained concerning the risk factors during concreting work in the differentiation between standard concretes with consistencies F1 to F4 and highly flowable concretes. In particular, it was determined that due to insufficient experience in handling highly flowable concretes, the psychological strain on construction site personnel rose considerably. Furthermore, clear deficits have been determined with regard to structural and procedural organization in construction practice when using highly flowable concretes. For this reason, careful working processes are necessary in the context of the planning procedures for the construction projects. A more exact statement about the spread of different influencing variables on the pressure of fresh concrete regarding existing uncertainties on construction sites, e.g. mean casting rate, can be made with continuing investigations in the process configuration of working time studies. For a defined failure probability, the design value of the pressure of fresh concrete is primarily influenced by the variation of the model parameters casting rate and setting time. The insufficient experience of the personnel in the handling of highly flowable concretes significantly increases the uncertainties, too. Further research into this subject is necessary. The results of this research project were incorporated in the improved standard DIN 18218:2010-01.

Acknowledgements

The research presented in this publication received financial support from the German Federal Ministry of Transport, Building & Urban Development (BMVBS), Güteschutzverband Betonschalungen e.V., Bilfinger Berger AG, Wayss & Freytag Ingenieurbau AG, Max Bögl Bauunternehmen GmbH and RSB Schalungstechnik GmbH. Materials were generously provided by MEVA Schalungssysteme GmbH, ELBA-WERK MaschinenGesellschaft mbH, Benno Drössler GmbH & Co. Bauunternehmung KG and further producers of formwork and building materials.

Notation a g h hE

hS H tE v V γF ρc σh

σh,max σh,max – σ h,E,max

spread in flow-table test [mm] gravity constant [m/s2] distance from the concrete level to a certain location [m] distance from the concrete level to the location where the concrete has achieved the final setting [m] hydrostatic height corresponding to the maximum pressure [m] height of casting section [m] final setting time of the concrete [h] mean casting rate [m/h] variation coefficient [–] partial safety factor [–] concrete density [g/cm3] horizontal (lateral) pressure of fresh concrete [kN/m2] maximum horizontal pressure of fresh concrete [kN/m2] maximum horizontal pressure of fresh concrete [kN/m2] normalized maximum pressure of fresh concrete [–]

References 1. Brameshuber, W., Beitzel, H., Beitzel, M., Bohnemann, C., Dehn, F., Graubner, C.-A., König, A., Lingemann, J., Müller, H. S., Pistol, K., Proske, T., Stettner, C., Zilch, K.: Formwork pressure induced by highly flowable concretes – Material investigations and large-scale tests. Structural Concrete 4/2011. 2. DIN 18218:1980-09: Pressure of fresh concrete on vertical formwork, Beuth Verlag, 1980. 3. ACI 347-04: Guide to Formwork for Concrete. American Concrete Institute, 2004. 4. Billberg, P., Österberg, T.: Självkompakterande betong – användingsteknik. Cement och Betong Institutet, CBI rapport, vol. 2, 2002, Stockholm. 5. Khayat, K. H., Assaad, J.: Use of Thixotropy-Enhancing Agent to Reduce Formwork Pressure Exerted by Self-Consolidating Concrete. ACI Material Journal, vol. 105, No. 1, JanFeb 2008, pp. 88–96. 6. Gregori, A., Ferron, R. P., Sun, Z., Shah, S. P.: Experimental Simulation of Self-Consolidating Concrete Formwork Pressure. ACI Material Journal, vol. 105, No. 1, Jan-Feb 2008, pp. 97–104. 7. Proske, T., Graubner, C.-A.: Pressure on Formwork using SCC – Experimental Studies and Modelling. Proceedings of the 5th RILEM International Symposium on Self-Compacting Concrete, Ghent, 2007, pp. 473–478. 8. Graubner, C.-A., Proske, T.: Formwork Pressure of Concretes with High Workability. In: Advances in Construction Materials 2007, Stuttgart, Springer-Verlag, 2007. 9. Vanhove, Y. et al.: Study of Self-Compacting Concrete Pressure on Formwork. In: Proceedings of 2nd International Symposium of Self-Compacting Concrete, COMS Eng. Corporation, Fukui, Japan, 2001. 10. Ovarlez, G., Roussel, N.: A physical model for the prediction of lateral stress exerted by self-compacting concrete on formwork. Materials and Structures, RILEM Publications SARL, vol. 39, 2006. 11. Proske, T.: Frischbetondruck bei Verwendung von Selbstverdichtendem Beton – Ein wirklichkeitsnahes Modell zu

Structural Concrete 13 (2012), No. 1

C.-A. Graubner/E. Boska/Ch. Motzko/T. Proske/F. Dehn · Formwork pressure induced by highly flowable concretes – design approach and transfer into practice

12.

13.

14.

15.

16. 17.

18.

19.

20.

21.

22.

Bestimmung der Einwirkungen auf Schalung und Rüstung, Dissertation, Technische Universität Darmstadt, 2007, 310 pp. Kim, Beacraft, Kwon, Shah: Simple Analytical Model for Formwork Design of Self-Consolidating Concrete. ACI Materials Journal, vol. 108, No. 1, Jan-Feb 2011, pp. 38–45. Khayat, K. H.: Final Report to the Ready-Mix Concrete Research Foundation American Concrete Institute – Concrete Research and Education Foundation on Self-Consolidating Concrete Formwork Pressure. 1 July 2006 – 25 May 2009, Université de Sherbrooke, 2009. Graubner, C.-A., Beitzel, H., Beitzel, M., Brameshuber, W., Brunner, M., Dehn, F., Glowienka, S., Hertle, R., Huth, J., Leitzbach, O., Meyer, L., Motzko, C., Müller, H. S., Proske, T., Rathfelder, M., Schuon, M., Uebachs, S.: DAfStb-Sachstandsbericht, Frischbetondruck fließfähiger Betone (State of the Art Report – Formwork Pressure using Concretes with high Workability), Schriftenreihe des Deutschen Ausschusses für Stahlbeton, vol. 567, Beuth-Verlag, Berlin, 2006, 75 pp. Graubner, C.-A., Beitzel, H., Beitzel, M., Bohnemann, C., Boska, E., Brameshuber, W., Dehn, F., König, A., Lingemann, J., Motzko, C., Müller, H. S., Pistol, K., Proske, T., Stettner, C., Zilch, K.: Schalungsbelastung durch Hochleistungsbetone mit fließfähiger Konsistenz – Ein Gemeinschaftsprojekt deutscher Forschungseinrichtungen. Final report F09-72008, TU Darmstadt, Institut für Massivbau, Nov 2008. DIN 18218:2010-01: Pressure of fresh concrete on vertical formwork, Beuth Verlag, 2010. Graubner, C.-A., Proske, T.: Frischbetondruck bei Verwendung von Selbstverdichtendem Beton. Beton- und Stahlbetonbau, Verlag Ernst & Sohn, No. 2, 2009. Council Directive 89/391/EEC of 12 June 1989 on the introduction of measures to encourage improvements in the safety and health of workers at work, 1989. Gesetz über die Durchführung von Maßnahmen des Arbeitsschutzes zur Verbesserung der Sicherheit und des Gesundheitsschutzes der Beschäftigten bei der Arbeit (Arbeitsschutzgesetz, ArbSchG), 21 Aug 1996. Empfehlungen zur Anfertigung einer Gefährdungsbeurteilung bei der Anwendung von Schalungen, Güteschutzverband Betonschalungen e.V., Ratingen/Darmstadt, 2007. REFA in der Baupraxis – Teil 1 Grundlagen. REFA Verband für Arbeitsstudien und Betriebsorganisation e.V. Darmstadt – Fachausschuss Bauwesen, Frankfurt am Main: ztv-Verlag, 1984. Huth, J.: Baubetriebliche Analyse von Selbstverdichtendem Beton. Dissertation, Institut für Baubetrieb, TU Darmstadt, Mensch & Buch Verlag, 2005.

Univ.-Prof. Dr.-Ing. Frank Dehn Full Professor University of Leipzig Institute for Mineralogy, Crystallography & Material Science (IMKM) Leipzig Germany dehn@mfpa-leipzig.de

Structural Concrete 13 (2012), No. 1

Dipl.-Ing. Erik Boska Research Assistant Institute of Construction Technologies & Management Technische Universität Darmstadt (IfB) Darmstadt Germany e.boska@baubetrieb.tu-darmstadt.de

Univ.-Prof. Dr.-Ing. Carl-Alexander Graubner Full Professor Technische Universität Darmstadt (IfM) Fachgebiet Massivbau Darmstadt Germany graubner@massivbau.tu-darmstadt.de

Univ.-Prof. Dr.-Ing. Christoph Motzko Full Professor Institute of Construction Technologies & Management Technische Universität Darmstadt (IfB) Darmstadt Germany c.motzko@baubetrieb.tu-darmstadt.de

Dr.-Ing. Tilo Proske Research Assistant Technische Universität Darmstadt (IfM) Fachgebiet Massivbau Darmstadt Germany proske@massivbau.tu-darmstadt.de

fib-news fib-news is produced as an integral part of the fib Journal Structural Concrete.

fib Symposium 2013 in Tel Aviv, Israel: Call for papers Contents

Issue 1 (2012)

fib Symposium 2013, Tel Aviv: Call for papers

ICCS13: Call for papers

fib Symposium Stockholm

ICI-fib Workshop, New Delhi

60th Birthday of Harald Müller

65th Birthday of Joost Walraven

Recent activities of fib Commission 8

Honorary doctorate to M. Curbach

fib Bulletins

Congresses and Symposia

Acknowledgement

Tel Aviv, venue of the fib symposium, seen from Jaffa (photo credit: Rémi Jouan via Wikimedia Commons)

Abstracts are now being accepted for the upcoming fib Symposium in Tel Aviv, which will take place from 19 to 24 April 2013. The deadline is April 2nd 2012; abstracts of up to 200 words should be submitted through the symposium website, www.fib2013tel-aviv.co.il. The symposium theme, “Engineering a Concrete Future: Technology, Modeling and Construction”, is directed at innovative aspects of concrete engineering in its various stages. Each topic includes several distinct areas of interest.

Symposium topics 1. Advanced and innovative cementitious materials and concrete 2. Constitutive modeling of cementitious and composite materials 3. Design concepts and structural modeling

4. Punching and shear in RC and in PC (prestressed concrete) 5. Challenges in bridge engineering 6. Advances in precast and PC engineering 7. Concrete structures under seismic and extreme loads 8. Pioneering structures and construction methods 9. Structural aspects of tunnel construction and design The 2013 fib symposium will be an excellent opportunity to gain professional state-of-the-art knowledge on recent developments and technologies as well as interesting current projects, share information and ideas and discuss them with experts, form international networks, and renew old acquaintances.

which is located about 2 km from the business center of the city and has state-of-the-art conference facilities. Between meetings, visitors can take in sea views, relax at the beach, or enjoy the spa facilities. Ben Gurion International Airport is less than a 15-kilometer drive away. Tel Aviv is the second largest city in Israel and considered to be the country’s commercial and cultural capital. In 2009 it celebrated its 100th anniversary; in 2003 it was proclaimed a World Cultural Heritage site by UNESCO, for its special “Bauhaus” architectural style. A special blend of Mediterranean ambience, seaside resort, modern facade, and many possibilities for dining and entertainment makes the city uniquely appealing.

About the venue The symposium will be held at the beachside Hilton Tel Aviv Hotel,

For more information about the event and venue, visit www.fib2013tel-aviv.co.il. Structural Concrete 13 (2012), No. 1

fib-news

ICCS13, Tokyo: Call for papers

fib Symposium Stockholm Registration is now open for the 2012 fib Symposium in Stockholm, taking place on 11-14 June. Early bird rates are available until 31 March 2012. For more information, visit www.fibstockholm2012.se.

Tokyo, venue for ICCS13: view of Shinjuku skyscrapers and Mount Fuji (photo credit: Morio via Wikimedia

The first fib Symposium in Sweden is approaching rapidly. The Swedish Concrete Association and KTH are happy to invite the large fib family to the Swedish capital Stockholm at the time of the year when there is sunlight – we cannot guarantee sunshine – almost 18 hours a day and when the trees are still bright green. The Swedish Concrete Association was founded in 1912 so this Symposium is very suitable as a part of the centennial celebration.

Commons)

Abstracts are being accepted until 31 May 2012 for the First International Conference on Concrete Sustainability (ICCS13), which will take place from 27 to 29 May 2013 in Tokyo, Japan.

Submission of abstracts Abstracts are now being accepted for the following topics: 1) Environmental impact reduction technologies

About the conference Sustainability is an issue of global importance in the 21st century. It has become essential for all industries around the world to take action toward sustainable development. The concrete industry is no exception, as it uses an enormous amount of resources and energy. However, concrete is also essential for the creation of infrastructure and buildings that form the basis of human socioeconomic activities. Future concrete use should promote the development of innovative concrete technologies and systems from the perspective of sustainability. The First International Conference on Concrete Sustainability will provide a forum to encourage such developments and to discuss the future of concrete by exchanging the latest information, technologies, and ideas.

2) Sustainability aspects in durability 3) Environmental design, evaluation, and systems 4) Social & economic aspects 5) Case studies of sustainable concrete materials and structures 6) Other related topics

About the venue Tokyo is the capital and economic, political, and entertainment center of Japan. With a population of over 13 million in the Tokyo metropolitan area alone, development and construction in Tokyo has produced one of the most land- and energy- efficient cities in the world. For more information, visit www.jci-iccs13.jp

Structural Concrete 13 (2012), No. 1

This summer will mark another centennial celebration because it has been 100 years since Stockholm hosted the 5th Olympic Games. The Stockholm Olympic Stadium was designed by the architect Torben Grut for the Olympic Games and this stadium is still used both for national soccer games and international track & field events. The Stockholm Stadium has been the arena for 83 world records through the years since 1912. Despite that the visible parts of the Stadium is structural masonry, concrete was also an important material for the Stadium structure and the continuous use is one evidence of sustainability which, as the reader already knows, is the theme of the Stockholm Symposium. Currently, the review process of the 180 papers is ongoing. The various topics have received different numbers of submissions. The most popular topics are the following: – Designing concrete structures for durability & sustainability – Durability – Repair, renovation, and upgrading for improved sustainability – Sustainable concrete materials – Sustainable concrete structures

fib-news

SLD of Concret Structures: ICI-fib Workshop, 14-15 November 2011 Other topics that have attracted a number of submissions include alternative binders, case studies, LCC & LCA, recycling, and sustainable concrete pavements, but the topics are not limited to those. The papers will be presented on three parallel sessions totalling 21 sessions including the opening and closing sessions and six keynote lectures. There will be no poster session, but papers describing ideas, current projects, preliminary results and minor studies will be presented at six special sessions with shorter presentation times in order to provide a platform for these authors who are generally younger and at an earlier stage of their career. The Organizing Committee has selected the largest Swedish technical university – KTH, the Royal Institute of Technology – as the symposium venue. KTH is nicely located within walking distance from Downtown Stockholm and is neighbouring a green area that also surrounds the Stockholm Stadium. The Thursday after the three-day symposium will contain an optional technical tour. A great deal of construction work is currently on-going in Stockholm. A 16-km long motorway mostly in tunnels is planned through the west suburban. Through downtown Stockholm a new railway tunnel – Citybanan – for commuting trains is under construction. It contains both rock tunnels, concrete tunnels, and submerged tunnels. The technical visit will focus on Citybanan. The social programme contains a welcoming reception at the famous Vasa Ship museum and a banquet starting with a boat trip in the Stockholm Archipelago.

Johan Silfwerbrand, Chair of the Organizing Committee and Deputy Chair of the Scientific Committee

From left to right: Dr. Harshavardhan Subbarao, Dr. Stuart Matthews, Mr. Steinar Helland, Mr. Jose Kurian and Prof. Mahesh Tandon

A very successful workshop on ‘Service Life Design of Concrete Structures’ was held on 14-15 November 2011 in New Delhi. The workshop was jointly organized by the Indian Concrete Institute (ICI) and fib under the leadership of ICI President Mr. Jose Kurian. The expert speakers from fib were Stuart L. Matthews (Building Research Establishment Ltd.) and Steinar Helland (Skanska Norge AS) and the material was largely drawn from fib Bulletins 34 and 53, International Standards ISO 22966 and 16204, and papers from the 2001 “Duranet” workshop and the fib Structural Concrete Journal. Stuart and Steinar provided the delegates with thorough understanding of Service Life Design and durability along with case studies drawn from their experience. The concept of through-life management, birth and rebirth certificates of a structure was also presented by Harshavardhan Subbarao of fib Commission 5. Prof. Mahesh Tandon, Past President of ICI, gave a talk on Indian practice and durability-related standards.

life design. The material included verification of the design service life according to a limit state and reliability-based methodology, guidance on writing specifications and standards on those subjects, behaviour of concrete structures and their interaction with the environment, measures to improve durability performance and reliability of concrete structures, guidance on the assessment, maintenance and extension of life of existing concrete structures. Case studies covered buildings and offshore marine structures subjected to carbonation and chloride deterioration respectively. Condition control – planned through-life structure management and care with monitoring of durability and performance – was also addressed.

The workshop was attended by consultants, engineers, manufacturers, owner representatives, contractors and academics. The focus was on the complex set of phenomena governing durability and long-term performance of concrete structures and how this forms a basis for service

Well done, Stuart Matthews, Steinar Helland and the organizers of the Workshop!

Interaction between the experts and audience was lively; this was to be expected as it was the first time that all these concepts were presented at an event in India. The event went a long way in promoting fib in India in partnership with the ICI.

Harshavardhan Subbarao, Construma Consultancy fib Commission 5

Structural Concrete 13 (2012), No. 1

fib-news

60th birthday of Harald Müller turned to Karlsruhe to take over the chair of his teacher and doctoral adviser Hubert Hilsdorf. He was also appointed head of the Construction Materials Technology Department at the Institute of Concrete, Masonry & Construction Materials Technology and director of the Materials Testing & Research Institute. Prof. Dr.-Ing. Harald S. Müller celebrated his 60th birthday on 16 December 2011. Born in 1951 in Osterburken, he was a Baden boy surrounded by the influences of Hesse, Franconia and Swabia. Construction materials caught the attention of Harald Müller very early on in his engineering studies. While still a student, he worked as an assistant at the Karlsruhe Institute of Concrete, Masonry & Construction Materials Technology under Hubert Hilsdorf, and later as a scientific assistant and doctoral candidate. Harald Müller’s obvious skills in the exact application of mathematical methods and mechanical principles led Hubert Hilsdorf to involve him in his research work at an early stage, and also to his discourse with Zdenek Bazant concerning creep in concrete, which appeared in numerous publications. Harald Müller quickly advanced to become an international expert on creep; the prediction of creep in concrete was the subject of his highly praised 1986 dissertation, which was awarded the “Honorary Senator Huber Prize”.

It was not long before Harald Müller emerged from the shadow of his highly respected predecessor and established himself as an expert on the science of construction materials in Germany and abroad. To this day he has remained faithful to his original scientific interest: describing the properties of concrete in constitutive laws. Testing methods for concrete and masonry became another focus of his research, along with the rheology of fresh concrete, the microstructure and durability of mineral building materials, plus life cycle management issues and the upgrading of structures. So far, that has added up to about 250 publications and some 40 dissertations for which he was either principal advisor or co-advisor. Harald Müller’s scientific work has always been characterised by high aspirations to achieve a system in terms of content and method

Structural Concrete 13 (2012), No. 1

Harald Müller is also a devoted teacher. In terms of their didactic concept, the rigorous structure of his courses is not unlike that of his research work, from the mathematical and physical principles to practical application. The construction engineering course in Karlsruhe also bears his hallmark, as he has held the post of Dean of Student Affairs for more than 10 years. On the occasion of his 60th birthday, it was a pleasure to be able to wish Harald Müller many happy returns and the health, creativity and enthusiasm needed to keep accomplishing all his activities. Harald Budelmann, TU Braunschweig

Festkolloquium in honour of Harald S. Müller

This author’s friendship with Harald Müller goes back to those early years. I can remember long telephone calls between Karlsruhe and Braunschweig, exchanging technical details about creep models and personal details about the quirks of our bosses and doctoral advisers! Harald Müller left Karlsruhe in 1989 to take charge of the Construction Materials Group at the Federal Institute for Materials Research & Testing in Berlin. In 1995 he re-

throughout, by clearly defined experimental work, materials models free from contradictions and with easily understandable physics, and mostly highly complex numerical analyses at component level, always prepared bearing in mind practical usability. It almost goes without saying that Harald Müller is involved in many national and international bodies; particularly worthy of note are his membership of the board of the German Reinforced Concrete Committee (DAfStb) and the Presidium of fib.

At the celebratory colloquium in his honour, Harald Müller was delighted to receive a bound compendium of his work since the beginning of his academic career.

On Thursday 19 January 2012, a celebratory colloquium (“Festkolloquium”) was held at the Karlsruhe Institute of Technology (KIT) in honour of the 60th birthday of Prof. Harald S. Müller. The programme, which was organized and moderated by Dr. Michael Haist and Dr. Nico Hermann, began with welcome addresses by Prof. Detlef Löhe, Vice President of Research and Information at KIT, and Prof. Bernhard Heck, Dean of Civil Engineering, Geo and Environmental Sciences.

fib-news

A laudatory speech (“Laudatio”) was then given by Prof. Franz Nestmann from KIT, which made clear that Harald Müller is not only prolific and successful in his research but also in his personal relationships with friends and colleagues, Also highlighted was his development project in Indonesia. Three 30minute technical presentations were then given: Prof. György L. Balázs (Budapest University of Technology and Economics/President of fib),

“fib Model Code 2010 as basis of codes for future concrete structures”; Prof. Manfred Curbach (TU Dresden), “Wie Baustoffe von heute das Bauen von morgen beeinflussen”; and Prof. Harald Budelmann (TU Braunschweig), “Baustofftechnologie in Lehre und Forschung”. A reception accompanied by music made for a festive closing to this pleasant and convivial event.

65th birthday of Joost Walraven

On 6 February 2012, Professor Dr.Ir. Joost C. Walraven turned 65. He was Professor of Concrete Structures at the Delft University of Technology in the Netherlands until the end of 2011. Through his teaching activites at the Technical University of Darmstadt from 1985 to 1989, his numerous lectures, for example at the Ulmer BetonTage, through his involvement in the Eurocodes, but especially through his many years volunteering for fib, Joost Walraven has become widely recognized beyond the borders of his home country as a researcher as well as as an engineer. Joost Walraven was born in 's-Hertogenbosch, Netherlands. He studied Civil Engineering at the Technical Institute of Delft (later renamed Delft University of Technology) at the time, where he earned his diploma in 1972. His former teacher, Professor A. Bruggeling, recognized Joost Walraven's skills and em-

ployed him as a scientific assistant. From 1981 to 1985 he was at Corsmit, an engineering office in The Hague, to gain practical experience. He then worked as a professor for concrete technology at the Technical University of Darmstadt until 1989. In 1989, he became professor at the Delft University of Technology, which was remained his professional home until his retirement. In addition, he volunteered in various associations and committees, especially fib as already mentioned. The scientific achievements of Joost Walraven cannot be described in detail in this short piece. But we can say that he has helped shape modern concrete construction significantly, especially when it comes to new models, new materials and new processes. This journey began with his doctoral thesis at TU Delft on aggregate interlock in concrete, which gained international attention in the concrete community. Other topics that are closely associated with his name are related to highstrength concrete, fiber reinforced concrete, ultra high performance concrete, self compacting concrete and the precast units. His goal was always to make these new technologies for practical applications. His nearly 300 publications and his

many lectures show that he has achieved this goal. He received many honors, including the Swedish Concrete Award in 1991, the fib Medal of Merit in 1998, and an honorary doctorate at the University of Kassel in July 2009 (see the September 2009 issue of fib-news). The undersigned has the honor to be working with Joost Walraven in fib since 1998, especially in the Presidium in the years 1998 to 2006. Joost Walraven has achieved several milestones in fib, the results of which are published in a number of fib Bulletins. Of outstanding importance is the 2010 fib Model Code for Structural Concrete, which was prepared under his leadership in recent years and in October 2011 was approved by the General Assembly of fib. One can assume that this Model Code as well as the provisions of previous editions in 1978 and 1990 has had a lasting influence on national and international regulations for concrete structures, such as the Eurocodes. Joost Walraven played a major role in this. The portrait of Joost would be incomplete without a glimpse of his personality. He is characterized by a strong interest in new developments, by a desire to be practical, fair, open to his colleagues, as well as by infinite diligence. His colleagues also admire his calm; even in critical situations he does not seem to lose either his wonderful sense of humor or his positive spirit. Surely these qualities are also appreciated by wife, Rose, who has always supported his commitments. To us, his friends and colleagues, remains the duty and pleasure of thanking Joost Walraven for his great professional commitment and to wish him, his wife Rose and his family all the best for the next stage in his life, especially good health. Hans-Ulrich Litzner Structural Concrete 13 (2012), No. 1

fib-news

Recent activities of fib Commission 8 fib Commission 8, Concrete, currently consists of the following active Task Groups: – TG 8.3 “Fibre-reinforced concrete” (Convener: Lucie Vandewalle, KU Leuven, Belgium) – TG 8.6 “Ultra-high performance fibre-reinforced concrete” (Convener: Joost Walraven, TU Delft, The Netherlands) – TG 8.7 “Code-type models for concrete behaviour” (Convener: Harald S. Müller, Karlsruhe Institute of Technology, Germany) – TG 8.8 “Structural design with flowable concrete” (Convener: Steffen Grünewald, TU Delft, The Netherlands & Liberato Ferrara, Politecnico di Milano, Italy) – TG 8.9 “Aesthetics of concrete surfaces” (Convener: Ludger Lohaus, Leibniz Univ. Hannover, Germany) – TG 8.10 “Performance-based specifications for concrete” (Convener: Hans Beushausen, UCT, South Africa & Frank Dehn, MFPA Leipzig, Germany) – TG 8.11 “Fire resistant concretes and cementitious composites for tunnel construction” (Convener: Frank Dehn, MFPA Leipzig) – TG 8.12 “Constitutive laws for concretes with supplementary cementitious materials” (Convener: Tor Arne Martius-Hammer, SINTEF, Norway & Harald Justnes, SINTEF, Norway) New Task Groups 8.11 and 8.12 were approved at the 2011 fib Technical Council meeting in Prague. The scopes of these Task Groups are briefly summarized below.

Task Group 8.11 In research and practice – both in the private and public sector – a great requirement is recognized to have information about concretes and cementitious composites which show a significant resistance against extremely high temperatures and temperature gradients as they can 66

Structural Concrete 13 (2012), No. 1

occur during tunnel fire scenarios. Therefore, the new TG 8.11, Fire resistant concretes and cementitious composites for tunnel construction, intends to prepare, for publiction as a technical report, a collection of available mix designs for concretes and cementitious composites in terms of their performances in fire, based on experimental studies and/or real-life experiences. Task Group 8.11 will cooperate intensively with fib Commission 1 and WG 4.3.5 on “Fire Design of Concrete Tunnels”.

Task Group 8.12 The use of Supplementary Cementitious Materials (SCM) a as binder in concrete is increasing, mainly driven by the need of the concrete industry to make concrete more environmentally friendly and in particular to meet official requirements for lower CO2 emissions. Today’s codes for concrete and concrete structures allow the use of SCM but limit which materials can be used and the amount of SCM (e.g. fly ash, silica fume, blast furnace slag, natural pozzolans, fine powders, etc.). Hence, the new Task Group intends to prepare the basis for an extension of fib MC 2010 which includes the assessment of lesser-known SCM and to verify, validate and extend the constitutive laws/models when SCM are used, e.g. on mechanical and durability properties. The launch meeting of Task Group 8.12 was held in Leipzig, Germany, in November 2011. In the future, it will strongly liaise with RILEM TC 238-SCM “Hydration and microstructure of concrete with supplementary cementitious materials (Chair: Nele de Belie, Gent University, Belgium).

Contribution of Commission 8 to MC2010, and Bulletins in progress One of the major targets during recent months was the preparation and finalization of chapters 5.1 “Concrete”, 5.6 “Fibres and fibre-reinforced concrete” and 7.7 “Verification of safety and serviceability of FRC structures” for the new fib Model Code 2010 (MC2010). This very important and voluminous work was accomplished by Task Group 8.3 “Fibre-reinforced concrete” and Task Group 8.6 “Codetype models for concrete behaviour”. Each Task Group is now preparing a bulletin which will provide background information and explanations for the content of the above-mentioned MC2010 chapters. The drafts of both bulletins will be are targeted for completion by the end of 2012. Also by the end of 2012, Task Group 8.7 will finalise its work on the design of “Ultra-high performance fibre-reinforced concrete” (UHPFRC). Also Task Group 8.8 and 8.9 intend to provide their stateof-the-art reports on flowable concretes and SCC – with and without fibres – and on concrete surface aesthetics, respectively.

Short Courses and workshops In June 2011 fib Task Group 8.10 hosted a successful workshop on “Performance-based specifications for concrete”. Intensive discussions amongst the 60 workshop participants together with 35 written contributions published in workshop proceedings established the basis for further Task Group work. Persons interested in purchasing a copy of the workshop proceedings can contact the organisers (www.mfpa-leipzig.de). In March 2012 fib Commission 8, especially Task Group 8.10, will hold a series of one-day workshops on technological issues for con-

fib-news

fib Bulletins cretes exposed to high temperatures and fire. The workshops will take place in South Africa on 5, 7 and 8 March 2012 in Cape Town, Durban, and Johannesburg, respectively. The workshops will provide in-depth information to the local South African technical community and are part of the strong efforts to foster the relationship with South Africa, which is a new National Member Group in fib as of this year.

mance-based specifications for concrete” and “Modern concrete technology” as these courses were successfully hosted by the fib Australian National Member Group and the “Concrete Institute of Australia” in Sydney and Brisbane in January 2011.

Further Commission 8 fib short courses are planned in New Zealand and Croatia during 2012, on “Perfor-

Frank Dehn, MFPA Leipzig Chairman of fib Commission 8 dehn@mfpa-leipzig.de

Further information about the activities of fib Commission 8 Concrete is given on the fib website (www.fibinternational.org).

Honorary doctorate to M. Curbach

Prof. Dr. Dr. h.c. Helmut Schmidt, President of the Technical University of Kaiserslautern (left), and Prof. Dr.-Ing. Dr.-Ing. E. h. Manfred Curbach at the ceremony in Kaiserslautern

On 15 November 2011, Prof. Dr.Ing. Manfred Curbach (Technical University Dresden, Germany) received an honorary doctorate from the Faculty of Civil Engineering at the Technical University of Kaiserslautern, Germany. His outstanding scientific achievements in structural engineering, his services in enforcing the implementation of research results in the construction practice and his exemplary personality were essential criteria for the award of an honorary doctorate to Prof. Manfred Curbach. As the

spokesman of the German Research Foundation’s special research program on textile reinforced concrete, he has contributed significantly to basic research on a new type of construction. He is currently coordinator of the DFG priority program 1542 “Light construction with concrete – Basics for building the future with bionic and mathematical design principles”, substantially initiated by himself, in which over 50 scholars from eleven technical uni versities cooperate in Germany.

In addition, Professor Curbach has held a variety of honorary positions. As Vice-President, he was responsible at the Technical University Dresden for its strategic direction for several years. Since 2004 he is President and CEO of the German Committee for Reinforced Concrete and also head of the German delegation in fib. Jürgen Schnell TU Kaiserslautern

fib Bulletin 61: Design examples for strut-and-tie models. Technical report, September 2011, 220 pages, ISBN 978-2-88394-101-4. Non-member price: 150 CHF. fib Bulletin 61 is a continuation of fib Bulletin 16 (2002). Again the bulletin’s main objective is to demonstrate the application of the FIP Recommendations “Practical Design of Structural Concrete”, and to illustrate the use of strut-and-tie models to design discontinuity regions (Dregions) in concrete structures. Bulletin 61 presents 14 examples, most of which are recently built existing structures. Although some of them can be considered to be quite important and, in some instances, complex, the chosen examples are not intended to be exceptional. The main aim is to look at specific design aspects, by selecting D-regions of the presented structures that are designed and detailed according to the proposed design principles and specifications for the use of strutand-tie models. Two papers deal with the role of concrete tension fields in modelling with strut-and-tie models, and summarize the experiences of the Working Group in applying strut-and-tie models to the examples. It is hoped that fib Bulletin 61 will encourage the use of more consistent design and detailing tools such as strut-and-tie models. To order this and other fib Bulletins: www.fib-international.org/publications. Structural Concrete 13 (2012), No. 1

fib-news

Congresses and symposia Date and location

Event

Main organiser

Contact

14 March 2012 Rio de Janeiro, Brazil

Latin America seminar on conceptual design and applications of precast concrete structures

Abcic fib Commission 6

www.abcic.org.br/ latinamericaseminar

29 May - 1 June 2012 Aix-en-Provence, France

International Conference on Numerical Modeling Strategies for Sustainable Concrete Structures (SSCS 2012)

AFGC

www.sscs2012.com

11â&#x20AC;&#x201C;14 June 2012 Stockholm, Sweden

fib Symposium: Concrete Structures for a Sustainable Community

fib group Sweden

www.fibstockholm2012.se

17-20 June 2012 Brescia, Italy

Bond in Concrete 2012: Bond, anchorage, detailing

University of Brescia, Italy

www.bondinconcrete2012.org

22-25 July 2012 Karlsruhe, Germany

9th fib International PhD Symposium in Civil Engineering

KIT Karlsruhe, Germany

http://fib-phd.imb.kit.edu/

19-21 September 2012 Guimaraes, Portugal

8th International Symposium on Fibre Reinforced Concrete

University of Minho RILEM

www.befib2012.civil.uminho.pt

28-30 January 2013 ta_2013 Johannesburg, S. Africa

Int. Conf. on Advances in

SPIN

www.spin.bam.de/en/acc-

Cement and Concrete Technology in Africa (ACCTA 2013)

20â&#x20AC;&#x201C;24 April 2013 Tel-Aviv, Israel

fib Symposium: Engineering fib group Israel a Concrete Future: Technology, Modeling and Construction

www.fib2013tel-aviv.co.il

27-29 May 2013 Tokyo, Japan

1st International Conference on Concrete Sustainability

JCI

www.jci-iccs13.jp

2-4 October 2013 Marseille, France

2nd International Symposium on UHPFRC

AFGC

www.afgc.asso.fr/

10-14 February 2014 Mumbai, India

4th International fib Congress and Exhibition

fib group India

website and email address to be announced

The calendar lists fib congresses and symposia, co-sponsored events and, if space permits, events supported by fib or organised by one of its National Groups. It reflects the state of information available to the Secretariat at the time of printing; the information given may be subject to change.

Structural Concrete 13 (2012), No. 1

fib-news

Acknowledgement fib – fédération internationale du béton – the International Federation for Structural Concrete – is grateful for the invaluable support of the following National Member Groups and Sponsoring Members, which contributes to the publication of fib technical bulletins, the Structural Concrete Journal, and fib-news.

National Member Groups AAHES - Asociación Argentina del Hormigón Estructural, Argentina CIA - Concrete Institute of Australia ÖVBB - Österr. Vereinigung Für Beton und Bautechnik, Austria GBB - Groupement Belge du Béton, Belgium ABECE - Associação Brasileira de Engenharia e Consultoria Estrutural, Brazil ABCIC - Associação Brasileira da Construção Industrializada de Concreto, Brazil fib Group of Canada CCES - China Civil Engineering Society Hrvatska Ogranak fib-a (HOFIB), Croatia Cyprus University of Technology Ceska betonarska spolecnost, Czech Republic Dansk Betonforening DBF, Denmark Suomen Betoniyhdistys r.y., Finland AFGC - Association Française de Génie Civil, France Deutscher Ausschuss für Stahlbeton e.V., Germany Deutscher Beton- und BautechnikVerein e.V. - DBV, Germany FDB - Fachvereinigung Deutscher Betonfertigteilbau, Germany Technical Chamber of Greece Hungarian Group of fib The Institution of Engineers (India) Technical Executive (Nezam Fanni) Bureau, Iran IACIE - Israeli Association of Construction and Infrastructure Engineers Consiglio Nazionale delle Ricerche, Italy JCI - Japan Concrete Institute

Sponsoring Members PCEA - Prestressed Concrete Engineering Association, Japan Admin. des Ponts et Chaussées, Luxembourg fib Netherlands New Zealand Concrete Society Norsk Betongforening, Norway Committee of Civil Engineering, Poland Polish Academy of Sciences GPBE - Grupo Portugês de Betão Estrutural, Portugal Society for Concrete and Prefab Units of Romania Technical University of Civil Engineering, Romania University of Transilvania Brasov, Romania Association for Structural Concrete (ASC), Russia Association of Structural Engineers, Serbia Slovak Union of Civil Engineers Slovenian Society of Structural Engineers University of Stellenbosch, South Africa KCI - Korean Concrete Institute, South Korea ACHE - Asociacion CientificoTécnica del Hormigon Estructural, Spain Svenska Betongföreningen, Sweden Délégation nationtale suisse de la fib, Switzerland ITU - Istanbul Technical University Research Inst. of Build. Constructions fib UK Group ASBI - American Segmental Bridge Institute, USA PCI - Precast/Prestress. Concrete Institute, USA PTI - Post Tensioning Institute, USA

Preconco Limited, Barbados Liuzhou OVM Machinery Co., Ltd, China Consolis Technology Oy Ab, Finland FBF Betondienst GmbH, Germany FIREP Rebar Technology GmbH, Germany MKT Metall-Kunststoff-Technik GmbH, Germany VBBF, Germany Larsen & Toubro, ECC Division, India Sireg, Italy Fuji P. S. Corporation, Japan Obayashi Corporation, Japan Oriental Shiraishi Corporation, Japan IHI Construction Service Co., Japan SE Corporation, Japan Sumitomo Mitsui Construct. Co., Japan P. S. Mitsubishi Construction Co., Japan Patriot Engineering, Russia BBR VT International, Switzerland VSL International, Switzerland SIKA Services, Switzerland Swiss Macro Polymers, Switzerland China Engineering Consultants, Taiwan (China) PBL Group Ltd, Thailand CCL Stressing Systems, United Kingdom

Structural Concrete 13 (2012), No. 1

Preview

Structural Concrete 2/2012 Eduardo Barros Aguiar, Ellen Bellucio, Mounir El Debs Behavior of grouted dowel used in precast concrete connections Amir M.Alani, Morteza Aboutalebi Analysis of the Subgrade Stiffness Effect on the Behaviour of Ground Supported Concrete Slabs Mathias Flansbjer, Jan Erik Lindqvist, Johan Silfwerbrand, Kamyab Zandi Hanjari Quantitative fracture characteristics in shear load Jirˇí Stráský, Radim Necˇas, Jan Kolácˇek Dynamic response of concrete footbridges

Footbridge across the Delta Pond Expressway, Eugene, Oregon, USA. An experience with the analysis and performance of 15 light concrete footbridges is presented from the point of view of their dynamic response caused by moving people.

Kai Osterminski, Peter Schießl, Bernd Isecke, Matthias Beck, Andreas Burkert, Michael Raupach, Jürgen Warkus, Jörg Harnisch, Wie Tian Deterioration model and input parameters for reinforcement corrosion

Kai Osterminski, Peter Schießl Development of the design model for reinforcement corrosion

Fax +49 (0)30 47031 240

Kyriacos Neocleous, Lluis Torres, Kypros Pilakoutas Design procedure and simplified equations for the flexural capacity of FRP RC sections

Yes, we would like to order Structural Concrete, beginning with the first issue of the current year for one year at first, 4 issues/year. The subscription can be stopped three month before the calendar year ends. If we do not cancel our subscription, please deliver for one additional year.

↓

Please choose:

Journal of the fib

Prices

personal / company

120 b

138 b

personal / company

170 sFr

196 sFr

personal / company

148 $

170 $

institutional / universities

460 b

529 b

institutional / universities

654 sFr

752 sFr

institutional / universities

566 $

651 $

↓

print + online

Please send us a free sample copy/1 issue of the journal Beton- und Stahlbetonbau (published in German language)

Invoice and delivery address: Company

VAT-No.

Titel, Last name, First name

Street/No

Department

Country, Zipcode, City

Phone

We guarantee you the right to revoke this order within two weeks, please mail to Verlag Ernst & Sohn, Wiley-VCH, Boschstr. 12, D-69469 Weinheim

Date

Signature

All prices are net-prices exclusive of VAT but inclusive postage and handling charges. Prices are valid until 31st August 2012.

Q Bauingenieure mögen Fußwegbrücken mäßiger Breite und Traglasten weniger beachten als Straßen- und Eisenbahnbrücken von spektakulärer Spannweite und Konstruktion. Städtebauer und Landschaftsplaner hingegen erachten innerstädtische Rad- und Gehwegbrücken für weitaus entscheidender und veranlassen in der Regel Wettbewerbe. Da es keine einheitlichen Entwurfsrichtlinien gibt, bilden die Erfahrungen aus realisierten Brücken eine wichtige Informationsquelle für Planer. Das vorliegende Buch enthält über 100 Beispiele, die in den letzten zehn Jahren weltweit gebaut wurden: offene Fuß- und Radwegbrücken, Viehtrieb- und Medienbrücken sowie einige geschlossene Verbindungsstege. Die Beispielsammlung ist nach Tragwerkstypen und Spannweiten gegliedert. Zu jeder Brücke gibt es eine kurze Darstellung der KLAUS IDELBERGER Randbedingungen und eine Bauwerksbeschreibung, illustriert anhand von Fotos, Grund- und Fuß- und Radwegbrücken Aufrissen, besonderen Konstruk tionsdetails. SoBeispielsammlung mit bildet es eine Fundgrube für Bauingenieure. 2011. 182 S., 351 Abb., Br. % 49,90* ISBN 978-3-433-02937-4

Q Structural engineers are often more interested in road and rail bridges of spectacular construction and enormous spans than in relatively narrow footbridges built for modest loads. City planners and landscape architects, on the other hand, see inner-city pedestrian and cycle bridges as important architectural elements and generally invite bridge builders to compete to find the winning design. As there are no official guidelines for the design of footbridges, the building techniques and performance of existing bridges are an important source of information for planners. This book contains over 100 examples of bridges built worldwide over the last ten years: open pedestrian and cycle bridges, cattle and utility bridges and some enclosed KLAUS IDELBERGER skyways. The collection is arranged according The World of Footbridges to load-bearing structure. For each bridge there From the Utilitarian to is a brief description of the location and the the Spectacular structural system, including special construction 2011. details, and illustrated by photographs, plans 183 pages, 351 ﬁgures. Softcover. and elevations. “Footbridges“ is a treasure % 69,–* trove for structural engineers. ISBN 978-3-433-02943-5

Online-Order: www.er nst-und-sohn.de

Q When bridges fail, often with loss of human life, those involved may be unwilling to speak openly about the cause. Yet it is possible to learn from mistakes. The lessons gained lead to greater safety and are a source of innovation. This book contains a systematic, unprecedented overview of more than 500 bridge failures assigned to the time of their occurrence in the bridges‘ life cycle and to the releasing events. Primary causes are identified. Many of the cases investigated are published here for the first time and previous interpretations are shown to be incomplete or incorrect. A catalogue of rules that can help to avoid future mistakes in design JOACHIM SCHEER analysis, planning and erection is included. A lifetime‘s work brilliantly compiled and Failed Bridges courageously presented – a wealth of knowledge Case Studies, Causes and experience for every structural engineer. and Consequences 2010. 321 pages. 120 ﬁg. 15 tab. Hardcover. % 79,–* ISBN 978-3-433-02951-0

Ernst & Sohn Verlag für Architektur und technische Wissenschaften GmbH & Co. KG

W i l e y

C o m p a n y

Customer Service: Wiley-VCH Boschstraße 12 D-69469 Weinheim

Tel. +49 (0)6201 606-400 Fax +49 (0)6201 606-184 service@wiley-vch.de

* %-Prices are valid in Germany, exclusively, and subject to alterations. Prices incl. VAT. Books excl. shipping. Journals incl. shipping. 0196300006_dp

Literature for bridge building by Ernst & Sohn

ED.: ECCS

CONVENTION FOR

Design of Steel Structures EC3: Design of Steel Structures. Part 1-1: General Rules and Rules for Buildings.

Fire Design of Steel Structures EC1: Actions on Structures. Part 1-2: Actions on Structures exposed to Fire. EC3: Design of Steel Structures. Part 1-2: Structural Fire Design.

April 2010. 446 pages, 295 ďŹ g., 105 tab., Softcover. % 70,â&#x20AC;&#x201C;*

May 2010. 428 pages, 134 ďŹ g., 21 tab., Softcover. % 70,â&#x20AC;&#x201C;*

ISBN 978-3-433-02973-2

ISBN 978-3-433-02974-9

C O N S T R U C T I O N A L S T E E LW O R K

Q This book introduces the basis design concept of Eurocode 3 for current steel structures, and their practical application. Numerous worked examples will facilitate the acceptance of the code and provide for a smooth transition from earlier national codes to the Eurocode.

Q This publication sets out the design process in a logical manner giving practical and helpful advice and easy to follow worked examples that will allow designers to exploit the benefits of the new approach given in the Eurocodes to fire design. ED.: ECCS

Fatigue Design of Steel and Composite Structures EC3: Design of Steel Structures. Part 1-9: Fatigue. EC4: Design of Composite Steel and Concrete Structures.

ED.: ECCS

Design of Plated Structures EC3: Design of Steel Structures. Part 1-5: Design of Plated Structures.

/CTOBER 2011. 3 pages. 250 ďŹ g. Softcover % 55,â&#x20AC;&#x201C;*

January 2011. 272 pages, 139 ďŹ g., 19 tab., Softcover. % 55,â&#x20AC;&#x201C;*

ISBN 978-3-433-02981-7

ISBN 978-3-433-02980-0

Q This volume addresses the specific subject of fatigue, a subject not familiar to many engineers, but still relevant for proper and good design of numerous steel structures. It explains all issues related to the subject: Basis of fatigue design, reliability and various verification formats, determination of stresses and stress ranges, fatigue strength, application range and limitations. It contains detailed examples of application of the concepts, computation methods and verifications.

Q This design manual provides practical advice to designers of plated structures for correct and efficient application of EN 1993-1-5 design rules and includes numerous design examples.

ED.: ECCS

Design of Connections in Steel and Composite Structures )) ÂŹ1UARTERÂŹ201 . approx. 500 pages, approx. 300 ďŹ g.

) ÂŹ1UARTERÂŹ201 . approx. 512 pages. approx. 300 ďŹ g. Softcover. approx. % 70,â&#x20AC;&#x201C;* ISBN 978-3-433-02979-4

Q The book is concerned with design of cold-formed steel structures in building based on the Eurocode 3 package, particularly on EN 1993-1-3. On this purpose, the book contains the essentials of theoretical background and design rules for cold-formed steel sections and sheeting, members and connections.

Softcover. approx. % 70,â&#x20AC;&#x201C;* ISBN: 978-3-433-02985-5 preliminary

preliminary

Online-Order: www.er nst-und-sohn.de

Design of Cold-formed Steel Structures EC3: Design of Steel Structures. Part 1-3: Design of Cold-formed Steel Structures.

Q This volume elucidates the design rules for connections in steel and composite structures which are set out in Eurocode 3 and 4. Numerous examples illustrate the application of the respective design rule.

Ernst & Sohn Verlag fĂźr Architektur und technische Wissenschaften GmbH & Co. KG

W i l e y

C o m p a n y

Customer Service: Wiley-VCH BoschstraĂ&#x;e 12 D-69469 Weinheim

Tel. +49 (0)6201 606-400 Fax +49 (0)6201 606-184 service@wiley-vch.de

*% Prices are valid in Germany, exclusively, and subject to alterations. Prices incl. VAT. Books excl. shipping. Journals incl. shipping. 0199100006_dp

ED.: ECCS â&#x20AC;&#x201C; EUROPEAN