A large number of bridges in the Netherlands have transversely post tensioned deck slabs cast in-situ between flanges of precast girders and were found to be critical in shear when evaluated by Eurocode 2. To investigate the bearing (punching shear) capacity of such bridges, a 1:2 scale bridge model was constructed in the laboratory and static tests were performed by varying the transverse prestressing level (TPL). A 3D solid, 1:2 scale model of the real bridge, similar to the experimental model, was developed in the finite element software DIANA and several nonlinear analyses were carried out. It was observed that the experimental and numerical ultimate load carrying capacity was much higher than predicted by the governing codes due to lack of consideration of compressive membrane action (CMA). In order to incorporate CMA in the Model Code 2010 (fib 2012) punching shear provisions for prestressed slabs, numerical and theoretical approaches were combined. As a result, sufficient factor of safety was observed when the real bridge design capacity was compared with the design wheel load of Eurocode 1. It was concluded that the existing bridges still had sufficient residual bearing capacity with no problems of serviceability and structural safety.

@en

Load testing and in some cases failure (or collapse) testing of bridges is a method to learn more about the behaviour of full-scale bridges in site conditions. Since such experiments, especially failure tests, are expensive, an extensive preparation of these tests is important. This paper addresses the question of when a bridge is a good candidate for a load test or a failure test. To answer this question, a multi-level assessment methodology is developed. The proposed method includes a decision tree that helps users decide which method should be used to reach the desired level of accuracy. These procedures are followed to carry out an assessment based on the load and resistance models and factors from the code, as well as to estimate the maximum required load in a collapse test based on average values and a single tandem. The procedures are illustrated with the case of the Nieuwklap Bridge in the province Groningen, the Netherlands. The multi-level analysis showed that testing the Nieuwklap bridge would most likely not result in a shear failure, and thus the test would not meet the goals of a collapse test in shear, which would provide valuable research insights. On a more abstract level, the result of this research is the development of a multi-level decision-making procedure that can be used to evaluate if a field test should be planned and can meet the identified goals.@en

When the shear resistance of prestressed beams with stirrups is determined with the current Eurocode, no distinction is made between regions with and without flexural cracks. This while it may be expected that a region without flexural cracks will have a higher shear resistance. This is due to the lower longitudinal strains and the narrow crack widths, resulting in a higher contribution of aggregate interlock. Also, the Eurocode does not take into account that in regions without flexural cracks, a significant part of the shear force is transferred through the uncracked flanges. This article proposes therefore a shear resistance model, based on Modified Compression Field Theory (MCFT), that does consider the low longitudinal strains and shear transfer through the uncracked flanges. From a comparison it was found that the proposed model can determine shear resistance as accurately as the most comprehensive level III approach of the Model Code 2010. However, the proposed model was found to be much easier to use in engineering practice as no iterations are necessary.

@en

Diagonal tension cracking is the governing failure mode for bridge girders with a thin web that are highly prestressed and contain little shear reinforcement. When assessing existing bridge girders using the Eurocode 2 [1], it often turns out that it is not possible to demonstrate sufficient resistance to diagonal tension cracking. This paper evaluates the method to determine the maximum principal tensile stresses as used in the Eurocode 2 [1] and investigates how flexural cracks affect the principle tensile stresses in the regions without flexural cracks. This paper also investigates how the tensile strength of the web is affected by the presence of compressive stresses and by the size of the area subjected to high tensile stresses. Based on the results of these investigations, an improved model is proposed to determine the resistance to diagonal tension cracking.

@en

Proof load testing can be an interesting method to assess existing bridges for which analytical methods are unable to provide an accurate assessment. In a proof load test, a load representative of the factored live load is applied to the bridge. If the bridge can carry this load without distress, the proof load test is success-ful, and the bridge proves it fulfils the code requirements. Since large loads are applied, the structure or element that is tested needs to be carefully monitored during the test. This paper reviews the literature on reported load tests and the measurement techniques used during these tests. It also includes the test goals these techniques can address, and the advantages and disadvantages of the contact and non-contact techniques. The result of this re-view is guidance for the selection of appropriate monitoring and measurement techniques during load tests. This practical recommendation can serve engineers during the preparation of a load test, and will be extended in the future with stop criteria validated with experimental results.@en

In The Netherlands, existing slab-between-girder bridges with prestressed girders and thin transversely prestressed concrete decks require assessment. The punching capacity was studied in a previous series of experiments, showing a higher capacity thanks to compressive membrane action in the deck. Then, concerns were raised with regard to fatigue loading. To address this, two series of large-scale experiments were carried out, varying the number of loads (single wheel print versus double wheel print), the loading sequence (constant amplitude versus variable amplitude, and different loading sequences for variable amplitude), and the distance between the prestressing ducts. An S-N curve is developed for the assessment of slab-between-girder bridges. The experiments showed that compressive membrane actions enhances the capacity of thin transversely prestressed decks subjected to fatigue loading.@en

The Netherlands has a large number of thin, transversely prestressed concrete bridge decks, cast in-situ between flanges of prestressed concrete girders dating back to the 1960s and 1970s. These bridges are critical in shear when analyzed using EN 1992-1-1:2005; however, in reality, they show no significant signs of distress, possibly because of residual bearing (punching shear) capacity arising from compressive membrane action. Since these bridges are old, it is an astute approach to check whether they can be used for a few more decades, provided they are safe and reliable against modern traffic loads. The results could then be applied to a wider range of structures, especially in developing countries facing economic constraints. A prototype bridge was selected and experimental, numerical and theoretical approaches were used to investigate its bearing capacity. Respective coefficients of variation of 11% and 9% were obtained when the experimental and the finite element analysis punching loads were compared with the theoretical results. This led to the conclusion that the existing transversely prestressed concrete bridge decks still have sufficient bearing capacity and considerable cost savings can be made if compressive membrane action is considered in the analysis.@en

It is widely known that as the structure of the size increases, its nominal strength decreases. In this paper, the effect of size on punching shear has been quantified for transversely post-Tensioned deck slabs cast between flanges of precast concrete girders. A 1:2 scaled model of the bridge was constructed in the laboratory, and experimental and numerical analyses were carried out. However, in order to apply these results on a real bridge, simply using the geometrical scale factors is not sufficient and a structural size effect has to be taken into account. Since a full-scale experimental study was not possible due to the costs involved, a numerical approach using finite element analysis software package TNO DIANA was used to model both the prototype and the real bridge, and a comparison was made to estimate the effect of size on the bearing capacity. It was found that increasing the transverse prestressing level had a positive effect on the punching shear strength of the deck slab. Furthermore, a lower size effect was observed with higher transverse prestressing levels. It is concluded that if a suitable size factor is used, either numerical or small-scale experimental studies can be reasonably used to investigate existing structures.@en

Most of the Dutch bridges were built around middle of the last century and it is vital for designers to find out if these bridges can still be considered safe for the traffic of modern times. The capacity in shear is especially critical as it was not considered in design recommendations before 1976. Therefore, experiments on a 1:2 scale model of a transversely prestressed concrete bridge deck cast between concrete girders were carried out to investigate the bearing (punching shear) capacity. The scale was selected based on the space available in the laboratory and the expected failure loads that would have to be applied. Also, a three-dimensional, solid, nonlinear finite element model was developed in the Finite Element Analysis software package TNO DIANA to study the structural behavior of deck slabs and is the focus of this paper. The results of the experimental and numerical analyses leads to the conclusion that existing bridges still have significant residual strength due to the presence of transverse prestressing and the membrane forces, and nonlinear finite element models can predict the load carrying capacity quite accurately.@en

In the Netherlands, existing bridges are being assessed to investigate whether they are still capable to resist current and future traffic loads. Bridges that are compiled of single span prestressed girders, appear to have insufficient resistance to diagonal tension cracking. This concerns bridges that do not contain sufficient stirrups. Consequently, diagonal tension cracking could result in an abrupt brittle failure. However, the assessments are performed using the Eurocode model and there is doubt about its accuracy. In this research the accuracy of the Eurocode model is determined by comparing predicted resistances with experimentally found resistances. Moreover the stress distribution according to the Eurocode model is compared with the stress distribution of a linear elastic finite element analysis. Based on the comparison, an alternative model is suggested, that predicts the resistance to diagonal tension cracking more accurately.@en

Proof load testing of existing reinforced concrete bridges is becoming increasingly important as the current bridge stock is aging. In a proof load test, a load that corresponds to the factored live load is applied to a bridge structure, to directly demonstrate that a bridge fulfills the code requirements. To optimize the procedures used in proof load tests, it can be interesting to combine field testing and finite element modeling. Finite element models can for example be used to assess a tested structure after the test when the critical position could not be loaded. In this paper, the case of viaduct De Beek, a four-span reinforced concrete slab bridge, is studied. Upon assessment, it was found that the requirements for bending moment are not fulfilled for this structure. This viaduct was proof load tested in the end span. However, the middle spans are the critical spans of this structure. The initial assessment of this viaduct was carried out with increasingly refined linear finite element models. To further study the behavior of this bridge, a non-linear finite element model is used. The data from the field test (measured strains on the bottom of the concrete cross-section, as well as measured deflection profiles) are used to update the non-linear finite element model for the end span, and to improve the modeling and assessment of the critical middle spans of the structure. Similarly, an improved assessment based on a linear finite element model is carried out. The approaches shown for viaduct De Beek should be applied for other case studies before recommendations for practice can be formulated. Eventually, an optimized combination of field testing and finite element modeling will result in an approach that potentially reduces the cost of field testing.@en

In the Netherlands, existing bridges are being assessed to investigate whether they are still capable to resist current and future traffic loads. A part of these bridges consists of prestressed I-and T-shaped girders with a low shear reinforcement ratio. This ratio is low because the principle stress criterion was used to verify sufficient shear resistance at the time of engineering. These bridges are now assessed with the present Eurocode. It appears that it is difficult to demonstrate sufficient shear tension resistance. In this paper it is investigated whether existing models can accurately predict shear tension resistance for girders with a low shear reinforcement ratio. The model used in the CSA was found to predict this resistance conservatively and consistent. However, for single span girders with low shear reinforcement ratios, it was also found that it is difficult to demonstrate additional capacity compared to the resistance to diagonal tension cracking.@en

This chapter describes the proof load testing of viaduct Zijlweg at a position that is critical for bending moment and at a position that is critical for shear. The viaduct Zijlweg has cracking caused by alkali-silica reaction, and the effect of material degradation on the capacity is uncertain. Therefore, the assessment of this viaduct was carried out with a proof load test. This chapter details the preparation, execution, and evaluation of viaduct Zijlweg. The outcome of the proof load test is, according to the currently used methods for proof load testing, that the viaduct fulfills the code requirements, and that strengthening or posting is not required.@en

Previous research showed that the capacity of existing slab-between-girder bridges is larger than expected based on the punching shear capacity prescribed by the governing codes, as a result of compressive membrane action. A first series of fatigue tests confirmed that compressive membrane action also acts under cycles of loading. However, a single experiment in which first a number of cycles with a higher load level and then with a lower load level were applied, seemed to indicate that this loading sequence shortens the fatigue life. This topic was further investigated in a second series of fatigue tests with three static tests and ten fatigue tests. The parameters that were varied are the sequence of loading and the effect of a single or a double wheel print. The results show that the sequence of load levels does not influence the fatigue life. @en

In a proof load test, a load representative of the factored live load is applied to the bridge. Since the applied load is large, stop criteria are important. Stop criteria for shear and flexure are proposed based on existing codes and guidelines, laboratory experiments, and theoretical considerations. This proposal is verified with the results from pilot proof load tests. The result of this comparison is that the stop criteria are never exceeded, or that they are exceeded only in the last load step. The proposed stop criteria are thus not overly conservative for application to field testing. However, information about the available margin of safety is not always available, especially for shear failures, and will need further experimental validation.@en

In the Netherlands many concrete bridges that were built in the sixties and seventies are still in service. One subset of these bridges consists of prestressed concrete T-beams with cast-in-between slabs, cross-beams and transverse prestressing. Upon (re)assessment, the strength of these bridges is often too low. However, for this type of bridge several mechanism, that could possibly contribute to a higher load bearing capacity, are usually not taken into account. One of these mechanism is compressive membrane action (CMA), in transverse direction, in the concrete deck slab. Another potential mechanism is arch action, in longitudinal direction, in the T-beam. To investigate these mechanism or the so called ‘system behaviour’, and the ultimate load capacity, a T-beam bridge from 1962 was tested in seven full size collapse tests. In four tests the deck was sawn in longitudinal direction so that it became possible to test the load capacity of the individual beams. The individual beam tests are analyzed using a 3D non-linear finite element model. In the experiments, the load was placed at two different positions from the support. It was found that the nonlinear analysis shows good agreement with the load placed at 2.25 m from the support. However, with the load at 4.00 m from the support, the non-linear analysis shows an overestimation of ∼15%. Ultimately, this research aims to improve the calculation methods for the existing T-beam bridges in the Netherlands.@en

In the Netherlands, slab-between-girder bridges with prestressed girders and transversely prestressed decks in between the girders require assessment. Static testing showed that compressive membrane action increases the capacity of these structures and that the decks fail in punching shear. The next question is if compressive membrane action also increases the capacity of these decks under repeated loads. Therefore, the same half-scale bridge structure as used for the static tests was subjected to repeated loads at different fractions of the maximum static load, different loading sequences, and for single and double concentrated loads. A relationship between the load level and number of cycles at failure (S-N curve) for the assessment of these bridges is proposed, but the influence of the loading sequence was not successfully quantified yet. The conclusion of the experiments is that compressive membrane action enhances the punching capacity of transversely prestressed thin decks subjected to repeated loads. @en

In the Netherlands, the assessment of existing prestressed concrete slab-between-girder bridges has revealed that the thin, transversely prestressed slabs may be critical for static and fatigue punching when evaluated using the recently introduced Eurocodes. On the other hand, compressive membrane action increases the capacity of these slabs, and it changes the failure mode from bending to punching shear. To improve the assessment of the existing prestressed slab-between-girder bridges in the Netherlands, two 1:2 scale models of an existing bridge, i.e., the Van Brienenoord Bridge, were built in the laboratory and tested monotonically, as well as under cycles of loading. The result of these experiments revealed: (1) the static strength of the decks, which showed that compressive membrane action significantly enhanced the punching capacity, and (2) the Wöhler curve of the decks, showed that the compressive membrane action remains under fatigue loading. The experimental results could then be used in the assessment of the most critical existing slab-between-girder bridges. The outcome was that the bridge had sufficient punching capacity for static and fatigue loads and, therefore, the existing slab-between-girder bridges in the Netherlands fulfilled the code requirements for static and fatigue punching.@en

Existing bridges with large uncertainties can be assessed with a proof load test. In a proof load test, a load representative of the factored live load is applied to the bridge at the critical position. If the bridge can carry this load without distress, the proof load test shows experimentally that the bridge fulfills the requirements of the code. Because large loads are applied during proof load tests, the structure or element that is tested needs to be carefully monitored during the test. The monitored structural responses are interpreted in terms of stop criteria. Existing stop criteria for flexure in reinforced concrete can be extended with theoretical considerations. These proposed stop criteria are then verified with experimental results: reinforced concrete beams failing in flexure and tested in the laboratory, a collapse test on an existing reinforced concrete slab bridge that reached flexural distress, and the pilot proof load tests that were carried out in the Netherlands and in which no distress was observed. The tests in which failure was obtained are used to evaluate the margin of safety provided by the proposed stop criteria. The available pilot proof load tests are analyzed to see if the proposed stop criteria are not overly conservative. The result of this comparison is that the stop criteria are never exceeded. Therefore, the proposed stop criteria can be used for proof load tests for the failure mode of bending moment in reinforced concrete structures.@en

Most Dutch bridges were constructed around the middle of the twentieth century and considering the fact that traffic has increased exponentially since, it is important to find out if these bridges are still safe for use. Experiments on a 1:2 scale were carried out in the laboratory of the Delft University of Technology to investigate the bearing capacity of transversely prestressed concrete bridge decks subjected to concentrated wheel loads. All the tests showed failure in punching shear at loads much higher than expected from the current codes. This paper presents the results of the experimental parametric study including the effect of the transverse prestressing level (TPL), location of the load, number of loads, size of the loading area (wheel print/loading plate), and influence of previous damage to the deck slab panel, on the bearing capacity.@en