Fatigue Assessment of Prestressed Concrete Slab-Between-Girder Bridges
Round 1
Reviewer 1 Report
The paper introduces an interesting study on the fatigue assessment of a bridge in the Netherlands. The Van Brienenoord bridge is prestressed concrete slab-between-girder, which was the common of bridges in Europe during a certain period of time. The study is decent, clear, and the paper is well-written. The authors came up with valuable results. The paper is ready for publication with considering the below comments:
Line 56: "One mechanisms ..", remove "s"
Line 58: "if progressive cracking and damage accumulation accumulation affects .. ”, remove “s” in affects.
Line 152: “The clear cover to the reinforcement is 7 mm.”, this clear cover looks small. Is this according to a standard?.
Author Response
Reviewer 1
The paper introduces an interesting study on the fatigue assessment of a bridge in the Netherlands. The Van Brienenoord bridge is prestressed concrete slab-between-girder, which was the common of bridges in Europe during a certain period of time. The study is decent, clear, and the paper is well-written. The authors came up with valuable results. The paper is ready for publication with considering the below comments:
Thank you for your comments and review. We have prepared a revised version of the manuscript based on your comments and the comments of the other reviewers.
Line 56: "One mechanisms ..", remove "s"
Thank you, I have corrected this.
Line 58: "if progressive cracking and damage accumulation accumulation affects .. ”, remove “s” in affects.
Thank you, I have corrected this.
Line 152: “The clear cover to the reinforcement is 7 mm.”, this clear cover looks small. Is this according to a standard?.
The cover of the Van Brienenoord Bridge is assumed as 30 mm (not given on the reinforcement drawings), and for the scale model we used 7 mm for practical purposes.
Reviewer 2 Report
The paper addresses fatigue performance of prestressed concrete slab between girder bridges with a particular focus in the Netherlands. In general, the reviewer identifies too much overlap between this paper and the authors' previous work. A few major concerns and comments are given below:
1- Eurocodes referred in the paper as "recently introduced" dates back to more than 10-15 years ago. The reviewer would expect some level of maturity in the research addressing bridges built before this time interval. However, large-scale laboratory tests conducted in this study still propose high value.
2- However, the test results presented in this study has already been published in two ACI Structural Journal papers ([34, 37]). As the authors acknowledge in numerous figures, current manuscript is preoccupied by former findings related to these papers. Please explain if that is not the case.
3- The only novel material in this manuscript corresponds to the Van Brienenoord Bridge, as far as I understand. In that sense, the reviewer expects justification stating that the paper presents more material than a case study since this is a research article.
4- Probably, it must have been stated in [34,37], but it would be nice to know how the experimental setup geometry is determined.
In summary, it is a very clearly written paper with good scientific method, however, there needs to be a valid improvement and significant novel contribution compared with the formerly published studies. Otherwise, current version of the paper seems to be restricted to a case study.
Editorial:
-Well written text, almost no typos.
-Line 56: replace "mechanisms" with "mechanism"
-Line 183: replace "gives" with "give"
Author Response
Reviewer 2
The paper addresses fatigue performance of prestressed concrete slab between girder bridges with a particular focus in the Netherlands. In general, the reviewer identifies too much overlap between this paper and the authors' previous work.
Thank you for your comments. We have addressed these comments as indicated below, and we have prepared a revised version of the manuscript based on your comments and the comments of the other two reviewers.
As for the overlap with previous work, the two papers that are accepted for publication in the ACI Structural Journal only deal with the experimental part of the research. They report the experiments and all details, and only contain the analysis in terms of the developed S-N diagrams. In this paper, the practical value of these experiments is given. The new and unique contributions of this manuscript are the following:
- Introducing the case study of the Van Brienenoord Bridge
- Methodology for converting test results of scale models in tool for assessment of punching shear and punching fatigue capacity of slab-between-girder bridges
- Outcome of assessment of slab-between-girder bridges: the analysis of the results of the scale model show that these bridges fulfil the code requirements.
A few major concerns and comments are given below:
1- Eurocodes referred in the paper as "recently introduced" dates back to more than 10-15 years ago. The reviewer would expect some level of maturity in the research addressing bridges built before this time interval. However, large-scale laboratory tests conducted in this study still propose high value.
In the Netherlands, the Eurocodes were introduced in 2012. While the Ministry of Transportation has carried out assessments based on the Eurocodes (and Eurocode proposals) for the past 15 years, a number of research projects are still ongoing related to the topic of assessment of the existing bridges. Open questions also still remain.
2- However, the test results presented in this study has already been published in two ACI Structural Journal papers ([34, 37]). As the authors acknowledge in numerous figures, current manuscript is preoccupied by former findings related to these papers. Please explain if that is not the case.
The experiments themselves have been published previously. However, the focus of the paper under consideration is the application of the outcome of these experiments to the assessment of slab-between-girder bridges. As indicated above, a number of elements of novelty and original research are introduced in this paper which distinguish is from the ACI SJ papers.
3- The only novel material in this manuscript corresponds to the Van Brienenoord Bridge, as far as I understand. In that sense, the reviewer expects justification stating that the paper presents more material than a case study since this is a research article.
As indicated above, there are three elements of novelty in this manuscript. The main novelty is the methodology for scaling up the test results to a method for the assessment of slab-between-girder bridges. For this purpose, we had to extend the procedures from the Eurocode for design by testing.
4- Probably, it must have been stated in [34,37], but it would be nice to know how the experimental setup geometry is determined.
The following elements played a role in determining the geometry of the test setup:
- Overall, the geometry is taken as 1:2 of the geometry of the Van Brienenoord Bridge, the most critical slab-between-girder bridge.
- The geometry of the end girders was determined to have the same (yet 1:2 scale) moment of inertia and torsional stiffness as the end girders in the case study bridge.
- The overall size of the test setup was limited by the available space in the laboratory.
The following information is added to the manuscript:
The choice for the size of the scale model and number of girders was determined as a function of the available test floor space in the laboratory.
In summary, it is a very clearly written paper with good scientific method, however, there needs to be a valid improvement and significant novel contribution compared with the formerly published studies. Otherwise, current version of the paper seems to be restricted to a case study.
Thank you for your comments. I hope the clarification of the novel contributions of this paper as compared to the ACI SJ publications (which are currently in press and not available yet, which makes comparison a bit harder) is clear.
Should you consider that this paper is more suitable as a case study, I would like to ask the handling editor to evaluate the manuscript and see if it should be changed from “original research” to “case study”. We think “original research” is the correct category, since the paper uses the case of the Van Brienenoord Bridge to evaluate all slab-between-girder bridges in the Netherlands.
Editorial:
-Well written text, almost no typos.
Thank you.
-Line 56: replace "mechanisms" with "mechanism"
Thank you , this typo has been corrected.
-Line 183: replace "gives" with "give"
I corrected this from “the tables gives” to “the table gives”.
Reviewer 3 Report
The present work consists reports about an experimental study in which the fatigue properties of reduced scale models of an existing bridge with varying parameters assessed. Typically, two-parameter and three-parameter Weibull distribution models are used to fit the experimental results which leads to ascertaining the predictive fatigue life assessment Arguments for selecting two-parameter and three-parameter Weibull distributions when modeling fatigue problems are already discussed in various published findings [1–4], where a justification is provided for the regression model proposed that has to fulfil the compatibility conditions between the cumulative distribution functions, namely F(f_max;N) =F(N;f_max) over the whole S-N field. This compatibility requirement is also claimed by Bolotin [5]. I highly suggest to cite these adequate references for enhanced clarity and readability.
The experimental design is adequate, results are properly reported and discussed, whereas the reference list is not exhaustive proving the weak knowledge of the authors on the subject, and conclusions do not properly summarize adequately the achieved goals of the paper.
Accordingly, a major revision of the paper is proposed taking into account the comments provided before the paper is published.
1. Figure 1 is taken from a book, I guess. Please seek the permission of the copyright holder.
2. F/Pmin is nor mentioned anywhere, does the lower level of fatigue loading was zero?
3. Wouldn’t a goodness-of-fir useful to corroborate the obtained results?
4. Pmax given in Table 2 has been experimentally determined, and no data reliability index is shown (standard deviation or error)> Also, the testing ages are varying greatly. Is there any particular reason for that.
5. Please provide the experimental setup picture (photograph).
References
[1] E. Castillo, A. Fernendez-Canteli, A Unified Statistical Methodology for Modeling Fatigue Damage, Springer Netherlands, 2009. doi:10.1007/978-1-4020-9182-7.
[2] A. Hanif, M. Usman, Z. Lu, Y. Cheng, Z. Li, Flexural Fatigue Behaviour of Thin Laminated Cementitious Composites Incorporating Cenosphere Fillers, Materials & Design. 140 (2018) 267–277. doi:10.1016/j.matdes.2017.12.003.
[3] A. Hanif, Y. Kim, C. Park, Numerical Validation of Two-Parameter Weibull Model for Assessing Failure Fatigue Lives of Laminated Cementitious Composites—Comparative Assessment of Modeling Approaches, Materials. 12 (2018) 110. doi:10.3390/ma12010110.
[4] C. Lu, B. Dong, J. Pan, Q. Shan, A. Hanif, W. Yin, An investigation on the behavior of a new connection for precast structures under reverse cyclic loading, Engineering Structures. 169 (2018) 131–140. doi:10.1016/j.engstruct.2018.05.041.
[5] V. V Bolotin, Statistical methods in structural fatigue, ASCE Proceedings. 96 (1969) 1201–1219.
Author Response
Reviewer 3
The present work consists reports about an experimental study in which the fatigue properties of reduced scale models of an existing bridge with varying parameters assessed.
Thank you for your comments. We have addressed your comments on an item-by-item basis as indicated below, and have prepared a revised version of our manuscript based on your comments and the comments of the two other reviewers.
Typically, two-parameter and three-parameter Weibull distribution models are used to fit the experimental results which leads to ascertaining the predictive fatigue life assessment. Arguments for selecting two-parameter and three-parameter Weibull distributions when modeling fatigue problems are already discussed in various published findings [1–4], where a justification is provided for the regression model proposed that has to fulfil the compatibility conditions between the cumulative distribution functions, namely F(f_max;N) =F(N;f_max) over the whole S-N field. This compatibility requirement is also claimed by Bolotin [5]. I highly suggest to cite these adequate references for enhanced clarity and readability.
Thank you for discussing the two- and three-parameter Weibull distribution models. The information is relevant and interesting for our work. We have included these references, except for reference 5. For reference 5, I have found the following article from the ASCE Journal of the Structural Division: “Statistical Methods in Structural Fatigue” from 1970 (not 1969?) and with Paul Wirsching and James Yao as authors (not V. Bolotin) here: https://cedb.asce.org/CEDBsearch/record.jsp?dockey=0016978.
The information is added as follows:
The current approach is based on a linear fit for the Wöhler curve. For an improved approach, two- and three-parameter Weibull distribution models could be used [38]. In [39-41], the methodology for selecting the Weibull distribution models and the compatibility requirements over the whole S-N field are given. This approach may be suitable to ascertain the predictive fatigue life assessment.
The experimental design is adequate,
Thank you.
results are properly reported and discussed,
Thank you.
whereas the reference list is not exhaustive proving the weak knowledge of the authors on the subject,
We have included references that discuss the two- and three-parameter Weibull distribution for modelling the S-N curve.
and conclusions do not properly summarize adequately the achieved goals of the paper.
The conclusions focused on three elements:
- Justification of the experiments and why we needed to use experiments for the assessment of slab-between-girder bridges
- Outcome of assessment for static punching and assessment for punching fatigue for the Van Brienenoord Bridge based on the scale model test results.
- What the outcome of the assessments implies for the assessment of the existing slab-between-girder bridges.
Since the focus of this paper was on using the experimental results for the assessment of the existing slab-between-girder bridges in the Netherlands, the contents of the conclusions seems to be properly reflecting our paper. We have, indeed, focused less on the actual fatigue testing in the conclusions – as this was a smaller part of our study and the focus of this paper is on bridge assessment. Please let us know if you disagree with this approach.
Accordingly, a major revision of the paper is proposed taking into account the comments provided before the paper is published.
We have prepared a revised version of the manuscript for your evaluation.
1. Figure 1 is taken from a book, I guess. Please seek the permission of the copyright holder.
Figure 1 is taken from the original plans of the Van Brienenoord Bridge.
2. F/Pmin is nor mentioned anywhere, does the lower level of fatigue loading was zero?
The original manuscript mentioned in words: with the lower limit 10% of the upper limit
For clarity, this is replaced with with the lower limit Fmin 10% of the upper limit
And Fmin has been added to the list of notations.
3. Wouldn’t a goodness-of-fir useful to corroborate the obtained results?
We calculated the goodness-of-fit with the chi-squared test, and added the following results to the manuscript:
For the current approach, the goodness-of-fit is calculated using the chi-squared test. For all fatigue tests (datapoints in Figure 5), the value equals 1. As such, the approach is considered satisfactory for purpose.
4. Pmax given in Table 2 has been experimentally determined, and no data reliability index is shown (standard deviation or error).
The value of Pmax is generally taken from a representative static test. As such, there is a one-on-one relation between the fatigue testing and static testing.
5. Also, the testing ages are varying greatly. Is there any particular reason for that.
The reason for the variation in testing ages is twofold:
- Since all experiments were carried out on the bridge mockup in the laboratory, we carried out the experiments consecutively. So, you can see that the age of the deck increases for increasing test number for specimen 1, specimen 1 with the new deck, and specimen 2.
- For the fatigue tests were a large number of cycles were carried out, the range of ages that was given includes the age of specimen at start and age of specimen at end of the fatigue test. Because of the large number of cycles, the fatigue tests lasted several days.
5. Please provide the experimental setup picture (photograph).
Figure 5 shows a photograph of the test setup.
References
[1] E. Castillo, A. Fernendez-Canteli, A Unified Statistical Methodology for Modeling Fatigue Damage, Springer Netherlands, 2009. doi:10.1007/978-1-4020-9182-7.
[2] A. Hanif, M. Usman, Z. Lu, Y. Cheng, Z. Li, Flexural Fatigue Behaviour of Thin Laminated Cementitious Composites Incorporating Cenosphere Fillers, Materials & Design. 140 (2018) 267–277. doi:10.1016/j.matdes.2017.12.003.
[3] A. Hanif, Y. Kim, C. Park, Numerical Validation of Two-Parameter Weibull Model for Assessing Failure Fatigue Lives of Laminated Cementitious Composites—Comparative Assessment of Modeling Approaches, Materials. 12 (2018) 110. doi:10.3390/ma12010110.
[4] C. Lu, B. Dong, J. Pan, Q. Shan, A. Hanif, W. Yin, An investigation on the behavior of a new connection for precast structures under reverse cyclic loading, Engineering Structures. 169 (2018) 131–140. doi:10.1016/j.engstruct.2018.05.041.
[5] V. V Bolotin, Statistical methods in structural fatigue, ASCE Proceedings. 96 (1969) 1201–1219.
Round 2
Reviewer 2 Report
The authors have properly addressed the questions by the reviewer. Given that the three merits are clearly stated in the manuscript as well, it would distinguish the new material compared with ACI Structural Journal.
Regarding the article type, this paper can still be considered as a research article in the form it extends laboratory work in to the field through the Van Brienenoord Bridge findings.
Reviewer 3 Report
The Authors have satisfactorily addressed the Reviewers' comments. I recommend the paper to be accepted now.
