Common practise in concrete design engineering is the sectional design approach. The model should... more Common practise in concrete design engineering is the sectional design approach. The model should be the most simple choice in this approach and will be often a 1D design model. The design process delivers the generalised moments, forces and displacements. The checks to the actual checking code is on a sectional base, where the section evaluation will be made nonlinear. In a more advantage design process the design model will be updated to a 2D design model. The 2D model results are distributed moments and forces, which give an overview of the results over the width of a 2D model. The distributed moments, forces can be integrated to the generalised moments and forces, so a check to the 1D model can be given too. Both models, the 1D and 2D models have a linear distribution over the height of the structure. The full 3D design model isn’t a common model in practices engineering, but by integrating stresses or strains over the height of the 3D model, the similar results coming from 1D and 2D can be generated. Today checking codes like Eurocode2 and ModelCode2010 allow the designer to use nonlinear analysis as additional tool to the linear approach. For existing structures these nonlinear analysis can help by looking to the remaining capacity and the remaining lifetime of a concrete structure. In this way it is possible to keep ‘old’ structures in operation on the actual infrastructural network. For new structures it is obvious that at this moment nonlinear design steps of a concrete structure aren’t acceptable. It is a too new process in the actual civil engineering practise, where there are no direct failures to the common accepted linear design process. But this nonlinear accepted design process could be moved to a closed concrete design process, where the nonlinear analysis gets the status of controlling the linear design process. A control function could avoid failures in designing the reinforcement or prestressed reinforcement. The designer will be warned already in the concrete design stage for failures or mistakes. However implementing a full nonlinear analysis isn’t an option, but the so-called sequential linear analysis, a quasi nonlinear analysis, is very robust and powerful to minimize the elapse time of this kind of control analysis. While the design process is setup in full 3D, BIM, inspection regime and maintenance are coming more together in an overall process in the near future.
In the Netherlands, the partnership between the Ministry of Infrastructure and the Environment an... more In the Netherlands, the partnership between the Ministry of Infrastructure and the Environment and the market has led to the new method CRIAM, which stands for 'Structural Risk Index Weighing Model'. With this model, an inspector and engineer can objectively determine whether a Unity Check, further material research or a complete re-examination is necessary. Many hours of research and calculations of infrastructural objects were saved with this objective weighing model. This process has led to considerable savings in both time and social costs. In addition, this method objectively determines a score which provides the urgency in which further research or reexamination is required. Meanwhile over 800 objects have been classified by urgency in The Netherlands. The uniform objective CRIAM score was successfully validated by the Ministry of Infrastructure and the Environment using historical figures and refined re-examination of objects, which took already place in the past.
The paper describes the comparison between three types of reliability methods: code type level I ... more The paper describes the comparison between three types of reliability methods: code type level I used by a designer, full level I and a level III method. Two cases that are typical for civil engineering practise, a cable-stayed subjected to traffic load and the installation of a soil retaining sheet pile wall, have been analysed with the three reliability methods and conclusions are drawn with respect to accuracy and computational effort. As expected the level III method uses more computational effort than a code type level I method. It is however shown that in practise, the effort for the level III reliability method is only slightly more than used in a full level I method. For new kind of structures and/or materials, where there is only little experience, a level III methods is then to be preferred above the level I method because it gives direct insight into the structural reliability and decisive parameters. Besides the structures will either be safer or more economically build in comparison to structures designed using the level I safety factors
This paper gives the results of a study into the (semi) long-term effects of wet and outdoor cond... more This paper gives the results of a study into the (semi) long-term effects of wet and outdoor condition on the material properties of Glass Fibre Reinforced Polyester (GFRP). In 2001, 6 GFRP test panels have been placed above, below and on the water level at the lock-heads of the “Spieringsluis” lock in the Netherlands. After 13 years of being exposed to these outdoor conditions, the panels were removed and material tests were carried out to determine the remaining strength and stiffness. In this paper, test results are compared to test results from panels that have been stored indoors during the same period and to test results from tests that were carried out in 1999. The reduction of strength and stiffness properties is compared to the reduction factors for temperature and humidity effects according to the Dutch CUR Recommendation 96 for the design of FRP structures.
Many prestressed concrete bridges with spans between 100 and 160 m have been built in the Netherl... more Many prestressed concrete bridges with spans between 100 and 160 m have been built in the Netherlands. The method of construction is the so-called in situ free cantilever construction. The first bridge executed according to this method was built in 1966. Starting from that date deflections were measured yearly by the Ministry of Transport, Public Works and Water Management. The observation of the actual performance of bridges revealed that in general the deflections were much larger than expected. Furthermore, no limit value of the actual deflection was reached even after 25 years. The deflections were mainly caused by shrinkage and creep of the concrete. A second reason for the deflections is the construction method: cantilever bridges, which are isostatic during the construction period are made continuous by keying the different beams. Starting from the end of the 70's a computer program was used to predict the deflections. The time-related losses were based on the CEB-FIP Model Code 1978. The computer predictions were compared with the field measurements. It appeared that these predictions underestimated the actual deflections by at least 20 percent. A further study on the effect of the time-related losses, if the CEB-FIP Model Code 1990 is used, is in progress. Although, this investigation is still in an early stage, some conclusions are discussed.
New codes for the design of concrete structures in The Netherlands give designers the opportunity... more New codes for the design of concrete structures in The Netherlands give designers the opportunity to utilize the possibilities of partial prestressed concrete. It will be possible in bridge construction to combine prestressing and reinforcement so that the result will be optimal; technically and economically. In order to learn about the behaviour of a concrete box girder bridge when it is overloaded to the ultimate limit state, non-linear calculations were made using DIANA. Two different models were made: one model as a combination of shell and solid elements and one model of beam elements. The results of the calculations are similar.
The use of Externally Bonded Reinforcement based on Fiber Reinforced Polymers (EBR FRP) has becom... more The use of Externally Bonded Reinforcement based on Fiber Reinforced Polymers (EBR FRP) has become increasingly popular in the construction industry lately. Strengthening of existing concrete structures with FRP materials has proven to be a cost-effective and durable strengthening technique. The extensive amount of experimental, numerical and analytical research into this relatively new strengthening technique confirms the wide interest. By now, guidelines have become available for the construction industry [1,2] and numerous applications have been realized worldwide. Despite these developments, some questions have still remained unanswered. Debonding of the FRP, which generally governs the failure of a strengthened structure, has still not been fully understood and new developments, like anchoring and prestressing of FRP, are continuously going on. One of the questions that so far has received only little attention is the effect of the ambient temperature on the debonding behavior of externally bonded FRP [3]. Temperature changes will induce additional thermal stresses, due to the significant difference in the Coefficient of Thermal Expansion (CTE) between concrete (ac ˜ 10 × 10-6 /°C) and e.g. CFRP (af ˜ -1 × 10-6 /°C in the fiber direction). The material properties of the concrete, adhesive and CFRP and the bond between these materials will also change with changing temperature. This particularly applies for the adhesive, as the strength and stiffness will drop significantly when the glass-transition temperature is reached (Tg ˜ 45°C – 80°C for most cold curing epoxies). Both the difference in the coefficient of thermal expansion and the changing material properties may affect the load level at which debonding occurs.
The experimental results of shear tests on two prefabricated large prestressed concrete T-beams a... more The experimental results of shear tests on two prefabricated large prestressed concrete T-beams are presented and discussed. The beams were used in previous experiments after which they remained undamaged. The beams are a 1:2 scale model of the approach bridge of the Van Brienenoord bridge in Rotterdam (the Netherlands). However, the reinforcement and prestressing is not an exact scale of the real bridge beams and was designed with requirements of the previous experiments and can be considered as over dimensioned. The beams have a length of 12 m and a depth of 1.3 m. The depth can be considered quite high for laboratory tests especially in terms of the equipment needed to load the beam until failure. Also, when considering a 1:1 scale structure, a beam with this depth could easily have a span of 30 m. Empty ducts in the top flange, used in the previous experiments for transverse prestressing, are filled with high strength cement mortar to prevent a premature flexural failure of the compression zone (crushing). The four shear tests consist of a single point load at a distance of 2.1d from the support. In previous tests a distance of 2.7d was used resulting in a close to shear but ultimately flexural type of failure (Ensink 2015). However, in the previous tests the flexural type of failure was also triggered by the empty ducts. Finally, the results of the shear tests are compared to non-linear 3D finite element analysis and Eurocode design formula using real material properties.
During the last decade, major developments have been made to include durability of reinforced con... more During the last decade, major developments have been made to include durability of reinforced concrete structures into the design process. 1-3 As corrosion of reinforcement steel is considered the major threat, special attention has been given to models for chloride ingress into concrete. However, mostly analytical expressions based on Fick's second law of diffusion 4 are used and consequently this introduces major limitations. In this paper the genometrical constraints imposed by embedded reinforcing bars is addressed. To this end, numerical calculations have been performed to quantify the effect of physical hindrance of chloride ingress on the corrosion initiation time. Based on the findings, it is concluded that the common approach assuming a unilaterial undisturbed ingress presents too optimistic results. Consequently, it is advocated to include the reinforcement geometry into service life calculations so as to achieve a more realistic assessment of the performance of reinforced concrete structures over time.
The Construction Simulation project will first of all be explained as it has taken shape in the C... more The Construction Simulation project will first of all be explained as it has taken shape in the Civil Engineering Division of the Directorate-General for Public Works and Water Management (Rijkswaterstaat) in cooperation with TNO-Bouw, the Building and Construction Department of TNO, the Dutch Organization for Applied Scientific Research. Next some practical applications, two bridges over the river Ussel, will be discussed as a part of the area of application.
Reinforcement in a concrete structure is often determined based on linear elastic stresses. This ... more Reinforcement in a concrete structure is often determined based on linear elastic stresses. This paper considers computation of the required reinforcement when these stresses have been determined by the finite element method with volume elements. Included are both tension reinforcement and compression reinforcement, multiple load combinations and crack control in the serviceability limit state. Results are presented of seventeen stress state examples.
A method is proposed for determining the required reinforcement based on stresses that have been ... more A method is proposed for determining the required reinforcement based on stresses that have been computed by the finite element method using volume elements. Included are, multiple load combinations, compression reinforcement, confinement reinforcement and crack control. The method is illustrated by several stress examples and a structural example.
Common practise in concrete design engineering is the sectional design approach. The model should... more Common practise in concrete design engineering is the sectional design approach. The model should be the most simple choice in this approach and will be often a 1D design model. The design process delivers the generalised moments, forces and displacements. The checks to the actual checking code is on a sectional base, where the section evaluation will be made nonlinear. In a more advantage design process the design model will be updated to a 2D design model. The 2D model results are distributed moments and forces, which give an overview of the results over the width of a 2D model. The distributed moments, forces can be integrated to the generalised moments and forces, so a check to the 1D model can be given too. Both models, the 1D and 2D models have a linear distribution over the height of the structure. The full 3D design model isn’t a common model in practices engineering, but by integrating stresses or strains over the height of the 3D model, the similar results coming from 1D and 2D can be generated. Today checking codes like Eurocode2 and ModelCode2010 allow the designer to use nonlinear analysis as additional tool to the linear approach. For existing structures these nonlinear analysis can help by looking to the remaining capacity and the remaining lifetime of a concrete structure. In this way it is possible to keep ‘old’ structures in operation on the actual infrastructural network. For new structures it is obvious that at this moment nonlinear design steps of a concrete structure aren’t acceptable. It is a too new process in the actual civil engineering practise, where there are no direct failures to the common accepted linear design process. But this nonlinear accepted design process could be moved to a closed concrete design process, where the nonlinear analysis gets the status of controlling the linear design process. A control function could avoid failures in designing the reinforcement or prestressed reinforcement. The designer will be warned already in the concrete design stage for failures or mistakes. However implementing a full nonlinear analysis isn’t an option, but the so-called sequential linear analysis, a quasi nonlinear analysis, is very robust and powerful to minimize the elapse time of this kind of control analysis. While the design process is setup in full 3D, BIM, inspection regime and maintenance are coming more together in an overall process in the near future.
In the Netherlands, the partnership between the Ministry of Infrastructure and the Environment an... more In the Netherlands, the partnership between the Ministry of Infrastructure and the Environment and the market has led to the new method CRIAM, which stands for 'Structural Risk Index Weighing Model'. With this model, an inspector and engineer can objectively determine whether a Unity Check, further material research or a complete re-examination is necessary. Many hours of research and calculations of infrastructural objects were saved with this objective weighing model. This process has led to considerable savings in both time and social costs. In addition, this method objectively determines a score which provides the urgency in which further research or reexamination is required. Meanwhile over 800 objects have been classified by urgency in The Netherlands. The uniform objective CRIAM score was successfully validated by the Ministry of Infrastructure and the Environment using historical figures and refined re-examination of objects, which took already place in the past.
The paper describes the comparison between three types of reliability methods: code type level I ... more The paper describes the comparison between three types of reliability methods: code type level I used by a designer, full level I and a level III method. Two cases that are typical for civil engineering practise, a cable-stayed subjected to traffic load and the installation of a soil retaining sheet pile wall, have been analysed with the three reliability methods and conclusions are drawn with respect to accuracy and computational effort. As expected the level III method uses more computational effort than a code type level I method. It is however shown that in practise, the effort for the level III reliability method is only slightly more than used in a full level I method. For new kind of structures and/or materials, where there is only little experience, a level III methods is then to be preferred above the level I method because it gives direct insight into the structural reliability and decisive parameters. Besides the structures will either be safer or more economically build in comparison to structures designed using the level I safety factors
This paper gives the results of a study into the (semi) long-term effects of wet and outdoor cond... more This paper gives the results of a study into the (semi) long-term effects of wet and outdoor condition on the material properties of Glass Fibre Reinforced Polyester (GFRP). In 2001, 6 GFRP test panels have been placed above, below and on the water level at the lock-heads of the “Spieringsluis” lock in the Netherlands. After 13 years of being exposed to these outdoor conditions, the panels were removed and material tests were carried out to determine the remaining strength and stiffness. In this paper, test results are compared to test results from panels that have been stored indoors during the same period and to test results from tests that were carried out in 1999. The reduction of strength and stiffness properties is compared to the reduction factors for temperature and humidity effects according to the Dutch CUR Recommendation 96 for the design of FRP structures.
Many prestressed concrete bridges with spans between 100 and 160 m have been built in the Netherl... more Many prestressed concrete bridges with spans between 100 and 160 m have been built in the Netherlands. The method of construction is the so-called in situ free cantilever construction. The first bridge executed according to this method was built in 1966. Starting from that date deflections were measured yearly by the Ministry of Transport, Public Works and Water Management. The observation of the actual performance of bridges revealed that in general the deflections were much larger than expected. Furthermore, no limit value of the actual deflection was reached even after 25 years. The deflections were mainly caused by shrinkage and creep of the concrete. A second reason for the deflections is the construction method: cantilever bridges, which are isostatic during the construction period are made continuous by keying the different beams. Starting from the end of the 70's a computer program was used to predict the deflections. The time-related losses were based on the CEB-FIP Model Code 1978. The computer predictions were compared with the field measurements. It appeared that these predictions underestimated the actual deflections by at least 20 percent. A further study on the effect of the time-related losses, if the CEB-FIP Model Code 1990 is used, is in progress. Although, this investigation is still in an early stage, some conclusions are discussed.
New codes for the design of concrete structures in The Netherlands give designers the opportunity... more New codes for the design of concrete structures in The Netherlands give designers the opportunity to utilize the possibilities of partial prestressed concrete. It will be possible in bridge construction to combine prestressing and reinforcement so that the result will be optimal; technically and economically. In order to learn about the behaviour of a concrete box girder bridge when it is overloaded to the ultimate limit state, non-linear calculations were made using DIANA. Two different models were made: one model as a combination of shell and solid elements and one model of beam elements. The results of the calculations are similar.
The use of Externally Bonded Reinforcement based on Fiber Reinforced Polymers (EBR FRP) has becom... more The use of Externally Bonded Reinforcement based on Fiber Reinforced Polymers (EBR FRP) has become increasingly popular in the construction industry lately. Strengthening of existing concrete structures with FRP materials has proven to be a cost-effective and durable strengthening technique. The extensive amount of experimental, numerical and analytical research into this relatively new strengthening technique confirms the wide interest. By now, guidelines have become available for the construction industry [1,2] and numerous applications have been realized worldwide. Despite these developments, some questions have still remained unanswered. Debonding of the FRP, which generally governs the failure of a strengthened structure, has still not been fully understood and new developments, like anchoring and prestressing of FRP, are continuously going on. One of the questions that so far has received only little attention is the effect of the ambient temperature on the debonding behavior of externally bonded FRP [3]. Temperature changes will induce additional thermal stresses, due to the significant difference in the Coefficient of Thermal Expansion (CTE) between concrete (ac ˜ 10 × 10-6 /°C) and e.g. CFRP (af ˜ -1 × 10-6 /°C in the fiber direction). The material properties of the concrete, adhesive and CFRP and the bond between these materials will also change with changing temperature. This particularly applies for the adhesive, as the strength and stiffness will drop significantly when the glass-transition temperature is reached (Tg ˜ 45°C – 80°C for most cold curing epoxies). Both the difference in the coefficient of thermal expansion and the changing material properties may affect the load level at which debonding occurs.
The experimental results of shear tests on two prefabricated large prestressed concrete T-beams a... more The experimental results of shear tests on two prefabricated large prestressed concrete T-beams are presented and discussed. The beams were used in previous experiments after which they remained undamaged. The beams are a 1:2 scale model of the approach bridge of the Van Brienenoord bridge in Rotterdam (the Netherlands). However, the reinforcement and prestressing is not an exact scale of the real bridge beams and was designed with requirements of the previous experiments and can be considered as over dimensioned. The beams have a length of 12 m and a depth of 1.3 m. The depth can be considered quite high for laboratory tests especially in terms of the equipment needed to load the beam until failure. Also, when considering a 1:1 scale structure, a beam with this depth could easily have a span of 30 m. Empty ducts in the top flange, used in the previous experiments for transverse prestressing, are filled with high strength cement mortar to prevent a premature flexural failure of the compression zone (crushing). The four shear tests consist of a single point load at a distance of 2.1d from the support. In previous tests a distance of 2.7d was used resulting in a close to shear but ultimately flexural type of failure (Ensink 2015). However, in the previous tests the flexural type of failure was also triggered by the empty ducts. Finally, the results of the shear tests are compared to non-linear 3D finite element analysis and Eurocode design formula using real material properties.
During the last decade, major developments have been made to include durability of reinforced con... more During the last decade, major developments have been made to include durability of reinforced concrete structures into the design process. 1-3 As corrosion of reinforcement steel is considered the major threat, special attention has been given to models for chloride ingress into concrete. However, mostly analytical expressions based on Fick's second law of diffusion 4 are used and consequently this introduces major limitations. In this paper the genometrical constraints imposed by embedded reinforcing bars is addressed. To this end, numerical calculations have been performed to quantify the effect of physical hindrance of chloride ingress on the corrosion initiation time. Based on the findings, it is concluded that the common approach assuming a unilaterial undisturbed ingress presents too optimistic results. Consequently, it is advocated to include the reinforcement geometry into service life calculations so as to achieve a more realistic assessment of the performance of reinforced concrete structures over time.
The Construction Simulation project will first of all be explained as it has taken shape in the C... more The Construction Simulation project will first of all be explained as it has taken shape in the Civil Engineering Division of the Directorate-General for Public Works and Water Management (Rijkswaterstaat) in cooperation with TNO-Bouw, the Building and Construction Department of TNO, the Dutch Organization for Applied Scientific Research. Next some practical applications, two bridges over the river Ussel, will be discussed as a part of the area of application.
Reinforcement in a concrete structure is often determined based on linear elastic stresses. This ... more Reinforcement in a concrete structure is often determined based on linear elastic stresses. This paper considers computation of the required reinforcement when these stresses have been determined by the finite element method with volume elements. Included are both tension reinforcement and compression reinforcement, multiple load combinations and crack control in the serviceability limit state. Results are presented of seventeen stress state examples.
A method is proposed for determining the required reinforcement based on stresses that have been ... more A method is proposed for determining the required reinforcement based on stresses that have been computed by the finite element method using volume elements. Included are, multiple load combinations, compression reinforcement, confinement reinforcement and crack control. The method is illustrated by several stress examples and a structural example.
Uploads
Papers by Ane de Boer