Method for Controlling Edge Wave Defects of Parts during Roll Forming of High-Strength Steel

Round 1

Reviewer 1 Report

Dear Authors,

The manuscript deals with an interesting technological issue concerning longitudinal cold forming of a profile. The difficulties associated with the production of this type of profiles are related to springback during molding, internal stress in the sheets and the lack of appropriate dimensional accuracy of the finished product.

The authors attempted to evaluate the influence of the process parameters on the edge waviness using numerical modelling.

However, in the paper I found some inaccuracies that should be explained and corrected:

1. Adding the picture of the analysed profile at the beginning of the manuscript would make the article clearer and more understandable.
2. What model of the deformed body was used for the calculations? The applied mathematical model of the deformed body raises doubts, especially in the context of the analysis of the shaping speed. Does the model take into account the strain rate?
3. Were elastic stresses analysed?
4. What is the yield stress and strength of 700BL material?
5. What is the scientific purpose of the analysis and how can the obtained results be applied?

More detailed comments are provided in the attached file.

Yours sincerely,

Reviewer

Comments for author File: Comments.pdf

Author Response

The manuscript deals with an interesting technological issue concerning longitudinal cold forming of a profile. The difficulties associated with the production of this type of profiles are related to springback during molding, internal stress in the sheets and the lack of appropriate dimensional accuracy of the finished product. The authors attempted to evaluate the influence of the process parameters on the edge waviness using numerical modelling. However, in the paper I found some inaccuracies that should be explained and corrected.

1. Adding the picture of the analysed profile at the beginning of the manuscript would make the article clearer and more understandable.

Answer: Thank you for your advice. I put the relevant pictures in the later sections of the article in order to correspond with the text.

2. What model of the deformed body was used for the calculations? The applied mathematical model of the deformed body raises doubts, especially in the context of the analysis of the shaping speed. Does the model take into account the strain rate?

Answer: Consider the overall deformation process, elasto-plastic model is used for the calculations. After changing the forming speed, the strain rate must be considered, the increase of the strain rate has an effect on the hardening behaviour of the material. In the whole process of forming, the strain of deformed body is very complex, It goes through a number of forming pass continuous forming, deformation zone and non- deformation zone strain rate changes have its deformation becomes complex, so this model only analyzes the edge wave phenomenon after forming at different speeds, without detailed analysis of the strain in the process.

3. Were elastic stresses analysed?

Answer: The elastic strain is very important in the forming process, but because this paper mainly analys the edge wave phenomenon after forming，the plastic strain is emphasized in the analysis.

4. What is the yield stress and strength of 700BL material?

Answer: I am very sorry for not providing enough information, the related information is shown in Table1.

5. What is the scientific purpose of the analysis and how can the obtained results be applied?

Answer: The starting point of this paper is to solve the practical production problems, because the current roll bending manufacturing technology is not perfect, related theories are developing gradually, mathematical simulation has become an effective research means. The phenomenon of edge wave in production leads to welding difficulties. This paper discusses the influence of several factors on edge wave by constructing mathematical model. There are more factors that need to be considered in engineering, so the conclusion can be used as a reference to improve the process and promote the production.

6. Other comments marked in the pdf file.

Answer: They have been modified in the pdf file.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper titled “Method for controlling edge wave defects of parts during roll forming of high-strength steel” deals with the design of an innovative manufacturing route for the production of a complex shaped structure using numerical simulation and experimental tests. Despite the topic is of interest, the paper is lacking in some of its part and has to be deeply modified to be reconsidered for the final publications. Language should be double-checked as well as the misuse of punctuation in some parts.

Moreover, the authors are asked to address the following aspects:

1. Equation 1 and Equation 2 are useful to pass from the nominal quantities (stress and strain) to the true ones, but there is no mention about the hardening law that was considered to fit the experimental data and to implement those data in the FE model.
2. The description of the FE model in section 2.2 is poor and must be improved. Some of the useful information are only provided later in the text, thus making the section 2.2 poor. There is no description of the part modelled (the assembly is reported in the second part of the paper) and all the other aspects can only be found later in the manuscript. The information should be better managed within a single-focused section describing the FE model.
3. Points labelled as A, B, C and D in Figure 3a are hardly distinguishable: the font should be enlarged.
4. It is not clear the meaning of the sentence “The actual formation boundary is inconsistent with the theoretical formation boundary”. The authors are asked to better explain such an aspect.
5. What to the indication “100-800 mm” refer to?
6. Figure 3b is difficult to interpret: there’s a profile, presumably the wavy profile, but what is reported on the horizontal line?
7. What do the parameter hi in equation 3 refer to? The maximum height from a single wave?
8. The authors state that the two traditional routes are complicated from the welding point of view; according to the scheme in Figure 1c, the one-time roll forming method requires 2 welding lines, exactly as in the second conventional route. Thus, in what extent the proposed methodology is preferable to the conventional two ones? Moreover, what does it mean when the authors say that the proposed manufacturing route is able to improve the mechanical properties of the component?
9. The authors report that the numerical simulations were carried out only regarding the manufacturing of the “b-shaped” section neglecting the second manufacturing step. How did they estimate that the conclusion from the “b-shaped” section manufacturing could be extended and be representative also for the completion of the component?
10. Acronym DTM is not specified in the text;
11. Titles on the axes in Figure 6b are in Chinese, so absolutely not understandable.
12. The last sentence of section 3.3 is not completely clear: what is the molding process? What about the molding requirements?
13. The assembly of Figure 7 should be better discussed: which of the roller pairs is mentioned in the text (upper roller of 150 mm, lower roller of 100 mm)? Did the numerical simulation regard all the passes to obtain the b-shaped section? How many passes are necessary to obtain the b-shape cross section? As a general consideration, as also anticipated earlier, all the information regarding the FE model should be moved in a dedicated section.
14. Table 2 mentions a friction coefficient of 0.2: is this value taken from literature? If so, the relative paper should be cited in the text.
15. It’s quite difficult to visualize what part of the component is referred as “the flange”: adding a picture showing the roller and a precise indication of the “flange” may help a lot the comprehension of the text, as well as the identification of the top, middle and bottom nodes that are not indicated in any of the reported pictures.
16. Fonts in legends of Figure 9 (a, b, c and d) are too small and have to be enlarged.
17. Lines 213-214 are not clear, probably due to a typo error.
18. It quite difficult to understand the advantage of investigate different flange heights and thickness. It can be assumed that the final components imposes specific geometrical constraint. If so, why three levels of the flange height were investigated even knowing that some of them could be inapplicable?
19. Sentence in line 265-267 is questionable: in principle, the increase of the applied strain has an effect on the hardening behaviour of the material if the process is carried out in warm/hot conditions, which it seems not to be the case of the discussed process. On the other hand, the effect of the strain rate when processing the material at room temperature is negligible, as also described by the well-known hardening law of Hollomon. Another questionable aspect is how the FE model is able to catch the effect of the strain since a single flow curve was implemented obtained at a single strain rate (1E-4 1/s). The authors should better support their considerations.
20. Titles in table 4 are completely wrong.
21. How about the comparison between the numerical and the experimental value?
22. The conclusion regarding the effect of thickness, higher than flange height, higher than the forming speed could be inferred directly from the numerical simulations. In the present form of the manuscript, numerical simulations and experimental tests seem not to be linked in any way: several conditions were simulated and there were no correlation with the experimental campaign, which in turns was based on another set of tests. This consideration leads to the question: what was the added value of the numerical simulation?

Author Response

The paper titled “Method for controlling edge wave defects of parts during roll forming of high-strength steel” deals with the design of an innovative manufacturing route for the production of a complex shaped structure using numerical simulation and experimental tests. Despite the topic is of interest, the paper is lacking in some of its part and has to be deeply modified to be reconsidered for the final publications. Language should be double-checked as well as the misuse of punctuation in some parts. Moreover, the authors are asked to address the following aspects:

1. Equation 1 and Equation 2 are useful to pass from the nominal quantities (stress and strain) to the true ones, but there is no mention about the hardening law that was considered to fit the experimental data and to implement those data in the FE model.

Answer: Thank you for your advice. When defining material properties in the model, isotropic hardening in software is directly selected to fit experimental data and to implement those data. I'm sorry for not expressing clearly. According to the opinion of another reviewer, formulas 1 and 2 have been deleted.

2. The description of the FE model in section 2.2 is poor and must be improved. Some of the useful information are only provided later in the text, thus making the section 2.2 poor. There is no description of the part modelled (the assembly is reported in the second part of the paper) and all the other aspects can only be found later in the manuscript. The information should be better managed within a single-focused section describing the FE model.

Answer: Thank you for your advice. Because section 2.2 mainly introduces the reasons and evaluation methods of edge waves, the finite element model is mainly included in section 3.4.

3. Points labelled as A, B, C and D in Figure 3a are hardly distinguishable: the font should be enlarged.

Answer: Thank you for pointing out the deficiency. I have modified figure3.

4. It is not clear the meaning of the sentence “The actual formation boundary is inconsistent with the theoretical formation boundary”. The authors are asked to better explain such an aspect.

Answer: Thank you for your advice. I have added the following sentence to the paper.Because different parts of the profile have different paths between the two passes, the path length of the side is often greater than the straight line distance between the two passes.

5. What to the indication “100-800 mm” refer to?

Answer: 100-800 mm refers to the analysis of the 100-800 mm edge wave in the forming direction for the 900mm "b" pipe.

6. Figure 3b is difficult to interpret: there’s a profile, presumably the wavy profile, but what is reported on the horizontal line?

Answer: The horizontal line is the longitudinal distance of the sheet.I'm sorry for not expressing clearly, it has been reflected in Figure 3(b).

7. What do the parameter hi in equation 3 refer to? The maximum height from a single wave?

Answer: “hi” represents the height of the i-th wave, I'm sorry for not expressing clearly, it has been reflected in Figure 3(b).

8. The authors state that the two traditional routes are complicated from the welding point of view; according to the scheme in Figure 1c, the one-time roll forming method requires 2 welding lines, exactly as in the second conventional route. Thus, in what extent the proposed methodology is preferable to the conventional two ones? Moreover, what does it mean when the authors say that the proposed manufacturing route is able to improve the mechanical properties of the component?

Answer: You are right, it is the same from the welding lines, but producing according to the traditional method obviously requires more strict welding process. As an automobile protective beam, the workpiece has unique advantages in structure. The one-time roll forming method not only reflects the advantages of roll bending technology, but also can better play the characteristics of the "日"-shaped cross-section. The structure plays a supporting role in space,  improves the overall stability, and can reduce the impact of structural damage.

9. The authors report that the numerical simulations were carried out only regarding the manufacturing of the “b-shaped” section neglecting the second manufacturing step. How did they estimate that the conclusion from the “b-shaped” section manufacturing could be extended and be representative also for the completion of the component?

Answer: Your consideration is very reasonable. This paper does not completely ignore the second manufacturing step, but after discovering the symmetry of the "日"-shaped cross-section, considering the factor of simulation calculation time, carries on the priority analysis of the "b-shaped" section and discusses the factors affecting the edge wave.

10. Acronym DTM is not specified in the text;

Answer: Thank you for pointing out the deficiency. I have modified in the text.

11. Titles on the axes in Figure 6b are in Chinese, so absolutely not understandable.

Answer: Thank you for pointing out the deficiency. I have modified figure6.

12. The last sentence of section 3.3 is not completely clear: what is the molding process? What about the molding requirements?

Answer: The forming process refers to a roll bending "b"-shaped cross-section according to the forming sequence designed above.

The forming requirements mentioned here only take strain into account. It can be seen from Figure 6 that there is no large strain in the forming process and no large strain difference in each part.

13. The assembly of Figure 7 should be better discussed: which of the roller pairs is mentioned in the text (upper roller of 150 mm, lower roller of 100 mm)? Did the numerical simulation regard all the passes to obtain the b-shaped section? How many passes are necessary to obtain the b-shape cross section? As a general consideration, as also anticipated earlier, all the information regarding the FE model should be moved in a dedicated section.

Answer: Thank you for your advice. For the roller pairs in all the passes, upper roller of 150 mm, lower roller of 100 mm.15 passes are necessary to obtain the b-shape cross section, all the information regarding the FE model is placed in section 3.4.

14. Table 2 mentions a friction coefficient of 0.2: is this value taken from literature? If so, the relative paper should be cited in the text.

Answer: Thank you for your advice. I have indicated the source in the paper.

15. It’s quite difficult to visualize what part of the component is referred as “the flange”: adding a picture showing the roller and a precise indication of the “flange” may help a lot the comprehension of the text, as well as the identification of the top, middle and bottom nodes that are not indicated in any of the reported pictures.

Answer: Thank you for pointing out the deficiency. I have modified figure8 and added related pictures.

16. Fonts in legends of Figure 9 (a, b, c and d) are too small and have to be enlarged.

Answer: Thank you for pointing out the deficiency. I have modified figure 9.

17. Lines 213-214 are not clear, probably due to a typo error.

Answer: Thank you for pointing out the deficiency. I'm sorry, the sentences are difficult to understand because of my handwriting problems. I have modified it.

18. It quite difficult to understand the advantage of investigate different flange heights and thickness. It can be assumed that the final components imposes specific geometrical constraint. If so, why three levels of the flange height were investigated even knowing that some of them could be inapplicable?

Answer: This study is intended to provide reference for actual production, so the choice of flange height is mainly based on the actual situation and feasibility consideration.

19. Sentence in line 265-267 is questionable: in principle, the increase of the applied strain has an effect on the hardening behaviour of the material if the process is carried out in warm/hot conditions, which it seems not to be the case of the discussed process. On the other hand, the effect of the strain rate when processing the material at room temperature is negligible, as also described by the well-known hardening law of Hollomon. Another questionable aspect is how the FE model is able to catch the effect of the strain since a single flow curve was implemented obtained at a single strain rate (1E-4 1/s). The authors should better support their considerations.

Answer: Your consideration is very reasonable. The view of this paper is mainly to consider the characteristics of the whole forming process, the sheet is equivalent to being continuously loaded and unloaded, so I think the hardening behavior of the material will be affected by the change of strain rate. I'm sorry, because this model focuses on the analysis of plastic strain after forming, some phenomena in the process of forming may not be reflected in the results, it can be said to be the limitation of this model.

20. Titles in table 4 are completely wrong.

Answer: Thank you for pointing out the deficiency. I have modified it.

21. How about the comparison between the numerical and the experimental value?

Answer: Numerical simulation only considers the influence of several factors on the edge wave, but many engineering parameters will affect the results in actual production, so it is not provided the comparison between the numerical and the experimental value.

22. The conclusion regarding the effect of thickness, higher than flange height, higher than the forming speed could be inferred directly from the numerical simulations. In the present form of the manuscript, numerical simulations and experimental tests seem not to be linked in any way: several conditions were simulated and there were no correlation with the experimental campaign, which in turns was based on another set of tests. This consideration leads to the question: what was the added value of the numerical simulation?

Answer: The starting point of this paper is to solve the practical production problems, because the current roll bending manufacturing technology is not perfect, related theories are developing gradually, mathematical simulation has become an effective research means. The phenomenon of edge wave in production leads to welding difficulties. This paper discusses the influence of several factors on edge wave by constructing mathematical model. There are more factors that need to be considered in engineering, so the conclusion can be used as a reference to improve the process and promote the production.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors corrected some of the shortcomings in the manuscript, however, there are still some comments and fixes that should be clarified and taken into account.
Details are highlighted in the text.

Comments for author File: Comments.pdf

Author Response

Thank you for marking your comments in the paper and I'll sort them out and reply here.

P1-21: Cold forming has been added.

P2-figure 1: Thank you for pointing out the deficiency. I have modified figure1.

Because the data in Table 1 is the average of three sets of data, and Figure 1 is just one set of data, there are some differences. I am sorry for not being clear about this. I have added content to explain it in the paper.

P3-figure 2: I'm sorry that this response did not satisfy you. It is worth noting that roll forming is different from other forming methods. Roll forming is a forming method that continuously deforms while conveying. A certain point on the metal is deformed in the previous process and then transferred to the next mold process to continue  deforming. The relative position of one point to the mold is constantly changing. In other forming methods, the position of the mold is unchanged, and the relative position of the metal material and the mold can establish a mathematical model to simulate the molding process. Figure 5 is a schematic method commonly used in roll forming, showing the deformation process of each pass and the shape of the deformed part. Each deformation corresponds to a set of roll bending molds. If want to focus on the analysis of deformation theory, it can only decompose the continuous roll bending process and analyze the deformation behavior of each set of roll bending mold positions separately. This analysis process is similar to the traditional bending process. But it is inaccurate for this research. This is because there is an obvious transition zone between the two roll bending deformations, and each roll has a different shape, so the deformation process of the transition zone is also irregular, and it is impossible to establish a mathematical model to describe the motion behavior of metal part. Taking into account the actual process of continuous roll forming, there are different degrees of elastic and plastic deformation in each roll bending process and the intermediate transition zone, and the deformation at different positions is also quite different. Roll forming is the ultimate goal of obtaining a good quality part. After the last roll bending process is completed, the part is plastically deformed to meet the forming requirements. There is no discussion on the detailed decomposition of the forming process of each set of roll bending molds in other roll bending references. I hope you can understand this process and approve our response.

P5-figure 5: Thank you for pointing out the deficiency. I have modified figure5.

P6-figure 6: Figure 6 is from COPRA RF 2021 DTM Simulation.

P7-figure 8: Thank you for pointing out the deficiency. I have modified figure8.

P8-figure 9: Thank you for pointing out the deficiency. I have modified figure9.

         Figure 9 can only show the values at a certain time in the forming process.

P9-figure 10:Figure 10a shows the Mises stress (equivalent stress) output of the model after shaping, it is classified as residual stress, and the strain is the total strain accumulated after multiple deformations. So these data do not reflect young's modulus, which defines material properties during modeling.

Author Response File: Author Response.pdf

Reviewer 2 Report

The questions pointed out by the reviewer have been satisfactorily addressed

Author Response

Thank you for your help.

Round 3

Reviewer 1 Report

The Authors provided sufficient answers. There are still some minor inaccuracies in the article that can be corrected and do not require another review.

There are minor notes to the manuscript in the file.

Best regards

Comments for author File: Comments.pdf

Author Response

Thank you for your help in my work, and thank you for your valuable comments.

P6-figure 6:Thank you for pointing out the deficiency. I have modified figure6.

P7-figure 8:Thank you for pointing out the deficiency. I have modified figure8.

P9-figure 10:Thank you very much for your suggestion.

I have modified the description in the figures and the paper.

Author Response File: Author Response.pdf