The Rolling Shear Influence on the Out-of-Plane Behavior of CLT Panels: A Comparative Analysis
Round 1
Reviewer 1 Report
The work is related to a comparative analysis on the rolling shear influence in CLT panels’ behaviour by different methods and H/L scenarios. The results are within the expected and according to existing literature. The work is well structured but there is a need to highlight its innovation. Moreover, there is a lack of information about the numerical model mainly related to its calibration, loading, shell type and connection between shells.
Overall the paper is well organized and within the scope of the journal, but could be improved with a better analysis on the conclusions of the paper and detailing along the paper.
The following comments should be considered in the review process:
General comments:
How was the numerical model validated? Is there an experimental benchmark? The numerical model was made by shells simulating each layer. How were the shells linked between each other (Page 7, line 257)? The work discusses the comparison between different methods and after present their results. At that point it would be important to reference other works that already have done such comparison in order to validate the results of this work. The innovation of the work must be better highlighted.
Specific comments:
There are several typos along the text that should be corrected, as example: Page 5, line 180: “coeffient” should be coefficient; Page 6, line 209: “incluted” should be included; Page 7, line 259: “assigne” should be assigned; Page 7, line 264: “semplicity” should be simplicity; Page 8, line 293: “rersults” should be results; Page 11, line 342: “serivce” should be service; Page 11, line 345: “interal” should be internal. Page 2, line 69: The expression “waste of time” is not suitable as the scope of the paper here is also to analysis shear. Page 3, line 121: Why is the point out of the scope of the paper? The scope of the present paper should be better defined. Page 7, line 260: Why were point loads selected (nodal actions) and not surface loads on the shells? Page 10, figure 9: In the shear analysis, the numerical model consistently presents non-symmetric behaviour for the lower and upper layers. Why is this happening?
Author Response
DETAILED RESPONSES TO REVIEWERS’ COMMENTS
The authors thank the Reviewers for theirs detailed and stimulating comments on the paper. These comments have given us the opportunity of improving the manuscript, as it is in the revised version.
The detailed responses to all the Reviewer’s comments are provided in the following.
Responses to Reviewer 1
Reviewers’ Comment #1 (general comments):
1.1) How was the numerical model validated? Is there an experimental benchmark? The numerical model was made by shells simulating each layer. How were the shells linked between each other (Page 7, line 257)?
1.2) The work discusses the comparison between different methods and after present their results. At that point it would be important to reference other works that already have done such comparison in order to validate the results of this work. The innovation of the work must be better highlighted.
Authors’ Response to comment #1: The reviewer’s observations are correct and stimulating. As regards the point 1.1) it should be noted that the use of finite element models (FEM), here intended as ‘shell elements’, to simulate the elastic behaviour of CLT panels is surely the most diffused approach in literature works (but also in practical applications) for modelling both in-plane and out-of-plane behaviour. Many papers containing the modelling of CLT panels by means of shell elements have been published, especially regarding structural modelling of CLT buildings. To this aim the references of the revised manuscript has been enlarged including citation on this topic.
In particular, the validity FEM has been commonly accepted by researchers to model the in-plane behaviour of CLT panels (provided that the orthotropy of material is taken into account), this also because the rolling shear deformations are negligible and simplify the material schematization. Conversely, in case of out-of-plane behaviour (such as floors) the rolling shear can affect the results of flexural behaviour, thus more attention should be given to develops appropriate material models.
In the present paper a multi-layered shell model has been adopted to simulate the out-of-plane behaviour, in which the layered shell allows any number of layers to be defined in the thickness direction, each with an independent location, thickness, behaviour and material elastic properties. In such a way to each layer is assigned the corresponding elastic moduli. The multi-layered shell model is based on the elastic plate theory in which the membrane deformation within each layer uses a strain projection method (Huges, 2000). While for bending, a Mindlin-Reissner formulation is used which always includes transverse shear deformation, thus the rolling shear transverse deformation are taken into account. With aims of clarify this aspect the following sentence has been included in the manuscript: The multi layered modelling allows any number of layers to be defined in the thickness direction, each with an independent location, thickness, behaviour and material elastic properties. The multi-layered shell model is based on the elastic plate theory where for bending a Mindlin-Reissner formulation is used which always includes transverse shear deformation, thus the rolling shear transverse deformation are taken into account [10].
As regards the validation of the model, it has not been specifically compared with experimental results by the authors. However, the validity of such a kind of modelling approach has been studied in literature (see Sturzenbecher et al. 2010 and Franzoni et al 2017 cited in the manuscript), demonstrating their validity to schematize the out-of-plane flexural behaviour of CLT panels be means of FEM.
In order to better specify this fact, the following sentence has been included in the manuscript on page 8: The effectiveness of 2D finite element models to schematize the in-plane elastic behaviour of CLT panels has been yet demonstrated for both in-plane [22, 23] and out-of-plane flexural behaviour. In particular, in Franzoni et al. (2017) [10] a FEM model has been adopted to perform numerical analyses on CLT panels subjected to out-of-plane loads and its effectiveness validated with reference to experimental results, while in Sturzenbecher et al. (2010) [14] the results derived from FEM have been compared with theoretical ones. In Vilguts et al. (2015) [24] finite element schematization has been also assumed to analyze the rolling shear effect on CLT floors. These comparisons have restituted encouraging results and confirmed the validity of the finite element models. Otherwise, the margin of error which may result from this model is limited as this is however elastic model.
Moreover, the reference of in Vilguts et al. (2015) [24] 2015 has been added because FEM model to schematize CLT panels has been also used.
In the end, concerning the link between the quadrilateral shell elements, it occurs with the function ‘edge constrain’ which connects all joints that are on the edge of the element to adjacent corner joints of the element.
This fact has been underlined in the text adding the following sentence on page 8: These quadrilateral elements have been linked one each other by means of the function ‘edge constrain’ which connects the joint on the edge to adjacent corner joints of the element.
With regards the point 1.2), the reviewer observation is correct again. But it has to be considered that scientific papers concerning comparative analyses on the effect of rolling shear of flexural behaviour of CLT floors under service loads are very few (and that in the authors’ knowledge the most important ones are cited in the manuscript). The major part of the existing papers investigates the experimental behaviour with the aim of obtaining reliable values of rolling shear modulus or strength, while other are focused on numerical modelling, or on the influences of timber species on elastic properties and so on. Instead in our paper the rolling shear influences is studied by a more practical point of view, detecting its influence with reference floor slenderness ratios typically used in real CLT constructions. The paper take inspiration form the paper written by Blass and Fellmoser on 2004 (see citation in the manuscript), in which for the first time the problem of rolling shear deformation on the global deflection of solid wood panels is highlighted. A further deepening of some aspects investigated in that paper are treated in our manuscripts, such as the effect of H/L ratios on the global deflection for a large range of slenderness ratios, the effectiveness of the methodologies used to take into account the rolling shear deformations, the incidence of the rolling shear in practical application, etc.
In order to gives effort to these aspects in the introduction of the paper (on page 2 and 3) it has been added this sentence: The most of existing literature papers mainly deal with experimental results on rolling shear modulus measurement, development of analytical methods or mathematical modelling by advanced plate theory. At date, specific information which could become useful suggestions for the technical codes for constructions are still missing. In the light of this, the present paper investigates the role of rolling shear deformation by a more practical point of view and contributes to cover the gap existing between pure theoretical investigations and concrete applications.
The paper takes inspiration from the work presented on 2004 by Fellmoser and Blass [6], in which the influence of the rolling shear effect on the out-of-plane deflections of cross-layered panels was firstly highlighted, even if indications on the stress state have not been provided. In order to quantify the incidence of rolling shear on floors deflections several aspects have been better deepened while others properly investigated in this paper, such as: (i) the effect of the span-to-length ratios (L/H) ratios on both instantaneous and global deflections at SLS, for a large range of L/H typically used in real buildings; (ii) the modification of the stress state within the panel due to rolling shear.
These aspects have been also studied applying the analytical methods commonly used in literature for taking into account the rolling shear problem and comparing the results provided by them. This gave the possibility of to compare and to verify the effectiveness of the different methods and to compute the weight given to the rolling shear by each method. This have allowed to quantify the role of rolling shear in practical application and to define a reasonable range of H/L in which the phenomenon becomes negligible.
As above mentioned a comparison between the Timoshenko theory and the SAM with respect to the experimental results obtained on 3 and 5-layer CLT panels is reported in Niederwestberg et al. 2018 [18]. As advancement also to the results provided by the modified g-method and by a specific 2D numerical models - properly developed by the authors - have been compared in this paper, thus enlarging the field of available results in the scientific literature.
Anyway, the results obtained by the author can be considered valid because of different approaches have been compared (both numerical and theoretical) providing a good matching among them.
Reviewers’ Comment #2 (specific comments): There are several typos along the text that should be corrected, as example: Page 5, line 180: “coeffient” should be coefficient; Page 6, line 209: “incluted” should be included; Page 7, line 259: “assigne” should be assigned; Page 7, line 264: “semplicity” should be simplicity; Page 8, line 293: “rersults” should be results; Page 11, line 342: “serivce” should be service; Page 11, line 345: “interal” should be internal. Page 2, line 69: The expression “waste of time” is not suitable as the scope of the paper here is also to analysis shear. Page 3, line 121: Why is the point out of the scope of the paper? The scope of the present paper should be better defined. Page 7, line 260: Why were point loads selected (nodal actions) and not surface loads on the shells? Page 10, figure 9: In the shear analysis, the numerical model consistently presents non-symmetric behaviour for the lower and upper layers. Why is this happening?
Authors’ Response to comment #2: According to the reviewer observation, all the grammar errors have been adjusted in the revised manuscript. Furthermore, more sentences have been rewritten in order to clarify the English language.
The question on page 3, line 21 has been better underlined as follows: This topic is beyond the scope of the present paper, because the effect of the rolling shear on the deformative flexural behaviour of CLT panels is mainly investigated here. Whereby, additional information can be found in the reference literature works [5, 12, 14, 16].
Moreover, the question on page 7, line 260 has been better explained, adding the following sentence: The joint of the quadrilateral shell elements have been selected to apply the nodal actions which simulate the uniformly distributed load, this because the used software allows to apply forces in the nodal intersections between shell elements only.
As regards the observation on page 10 - Fig. 9 it has been specified in the manuscript the reason for which the behaviour is not symmetric, as follows: (c) a greater discrepancy of the tangential stress distributions for between analytical and numerical approaches is resulted in case of low H/L ratios (L/H<20), probably due to the fact that the beam theory - on which are essentially based the analytical methods - tends to fall in case of shear-dominated elements (stocky elements). Instead a good agreement among the tangential stress values obtained from the analytical methods, also for low H/L ratios, is resulted.
PS. In the manuscript are highlighted the substantial modifications requested by the reviewer. But, it should be noted that more sentences have been rewritten in order to clarify the English language (not highlighted in the manuscript).
Author Response File: 
Author Response.docx
Reviewer 2 Report
The paper presented by authors compares different models for the rolling shear behavior of cross-laminated timber panels. The authors compare the stress distribution and relative contribution of the shear deformation to the total deformation of the different models.
The authors describe with sufficient accuracy the problem of rolling shear in CLT panels,, the existing literature and their methodology. The presentation of the results is clear.
In its current state, the paper has two main flaws.
(1) It is unclear what the main contribution of the paper is. The authors should more carefully describe the literature gap that this contribution is filling. what is missing from the research literature? What is the contribution of this paper to the field? Why was the comparative analysis performed?
(2) the clarity and style of the language should be revisited.
Author Response
DETAILED RESPONSES TO REVIEWERS’ COMMENTS
The authors thank the Reviewers for theirs detailed and stimulating comments on the paper. These comments have given us the opportunity of improving the manuscript, as it is in the revised version.
The detailed responses to all the Reviewer’s comments are provided in the following.
Responses to Reviewer 2
Reviewers’ Comment #1. In its current state, the paper has two main flaws: 1) It is unclear what the main contribution of the paper is. The authors should more carefully describe the literature gap that this contribution is filling. what is missing from the research literature? What is the contribution of this paper to the field? Why was the comparative analysis performed? 2) the clarity and style of the language should be revisited.
Authors’ Response to comment #1: The observation of the reviewer is correct. It is not so much clear the scope of the paper and its main contribution. Moreover, also the main scope of the comparative analyses has been clarified. To better clarify these facts the following part has been added into the introduction on pages 2 and 3: The most of existing literature papers mainly deal with experimental results on rolling shear modulus measurement, development of analytical methods or mathematical modelling by advanced plate theory. At date, specific information that could become useful suggestions for the technical codes for constructions are still missing. In light of this, the present paper investigates the role of rolling shear deformation by a more practical point of view and contributes to cover the gap existing between pure theoretical investigations and concrete applications.
The paper takes inspiration from the work presented on 2004 by Fellmoser and Blass [6], in which the influence of the rolling shear effect on the out-of-plane deflections of cross-layered panels was firstly highlighted, even if indications on the stress state have not been provided. In order to quantify the incidence of rolling shear on floors deflections several aspects have been better deepened while others properly investigated in this paper, such as: (i) the effect of the span-to-length ratios (L/H) ratios on both instantaneous and global deflections at SLS, for a large range of L/H typically used in real buildings; (ii) the modification of the stress state within the panel due to rolling shear.
These aspects have been also studied applying the analytical methods commonly used in literature for taking into account the rolling shear problem and comparing the results provided by them. This gave the possibility of to compare and to verify the effectiveness of the different methods and to compute the weight given to the rolling shear by each method. This have allowed to quantify the role of rolling shear in practical application and to define a reasonable range of H/L in which the phenomenon becomes negligible.
As above mentioned a comparison between the Timoshenko theory and the SAM with respect to the experimental results obtained on 3 and 5-layer CLT panels is reported in Niederwestberg et al. 2018 [18]. As advancement also to the results provided by the modified g-method and by a specific 2D numerical models - properly developed by the authors - have been compared in this paper, thus enlarging the field of available results in the scientific literature.
Furthermore, according to the reviewer observation, all the grammar errors have been adjusted in the revised manuscript and many sentences rewritten in order to clarify the English language.
PS. In the manuscript are highlighted the substantial modifications requested by the reviewer. But, it should be noted that more sentences have been rewritten in order to clarify the English language (not highlighted in the manuscript).
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The reviewer thanks the authors for improving the paper by considering the reviewer's comments and by implementing the asked information in the updated version of the manuscript. The present version has been significantly improved.
