Fatigue Estimation Using Inverse Stamping

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper deals with fast fatigue life estimation of stamped parts using reverse engineering methods. The research content is rich and has certain research value. Several questions should be modified before final accepted.

1. The authors give too little overview of the study in the abstract. It is recommended to rewrite the summary.

2. What are the research significance and innovation points of this paper?

3. The authors wrote “Figure 6a shows the thickness of this layer for specimen A (the specimens are defined in the following section), Figure 6b shows the thickness of the coating of specimen C.” But the definitions of A and C come later. It is suggested to adjust the order of relevant contents to increase the readability of the paper.

Author Response

The abstract of the article was supplemented with additional information. The innovation of this paper is the use of inverse stamping in conjunction with fatigue life evaluation. It looks for an area of application for such an approach. The description of specimen A and C was improved. We have added links to the text that should help the reader. Thank you for the notice.

Reviewer 2 Report

Comments and Suggestions for Authors

The article concerns the fatigue analysis of stamped elements using reverse engineering. In this work, 4 samples made of two materials were analyzed, some of the samples were covered with a zinc coating. The first two types of samples (A and B) were subjected to three-point bending, and the remaining two types (C and D) were subjected to a tension-compression test. A research algorithm was proposed and three computational approaches were compared. In the first approach, the influence of plastic deformation was completely ignored. Her results were highly conservative. The second approach was based on the MVS method and the results obtained were a bit too optimistic. The final computational approach was based on the MLSS method and the results matched the measured data. On this basis, the authors conclude that the MLSS method seems to be the best option for practical application, because the developed computational procedure is useful in reverse engineering when combined with other reverse engineering techniques, such as 3D scanning.

In the reviewer's opinion, the work is more like a research report than a scientific article, because in many places there are many mental shortcuts, there is no detailed discussion of the research results and dry facts are provided. The authors provide only the most important conclusions, they do not explain what they intend to do to achieve the research results. Moreover, the research shown in this article refers to their own publications or those of other authors, without even a few sentences of commentary. It will be very difficult for readers who are not familiar with this topic to understand the authors' course of action to demonstrate the purpose of the article, i.e. determining the fatigue strength of stamped products. Results information for coated samples is difficult to find. For example: "The number of test pieces that did not have a surface coating varied from three to five at each loading level. The number of tested pieces with surface treatment (Zn-Ni coating) varied from four to six", i.e. it is not actually known how many samples there were. Then, samples with different roughness appear in the text, but it is not known what they mean in the context of the entire article, because at the beginning it mentions samples with coatings. It is also difficult to interpret the results in the charts - in the reviewer's opinion, they require much better commentary for better readability. On what basis do the authors state: "There is good agreement between the MLSS method and the experimental results from samples without surface protection in most cases"??? Only one photo is shown (Figures 19 and 20) and a statement is made that good agreement was achieved. The work would certainly be more interesting and more readable and understandable to other people if the authors showed fewer "dry" results, such as tables and charts, but described it all better. It is recommended that: - explanation of why zinc coatings were used, since there is no such information in the analysis (e.g. in modeling); - they described the experimental tests in more detail, and in particular the fatigue tests, what they involved, etc.; - what exact sample variants did they use and what number of repetitions for each variant; - how did they model samples with coatings in FEM? - they described in more detail the 3 methods they used in their research. - please remove the yellow highlight at the end in the position: "Author Contributions" and "Funding".

Author Response

Above all, we have made changes in the introductory part of the article so that leaps of thought are reduced and the article is clearer for the reader. The number of samples is given in the appendix. We have added a link to this raw data in the text. We improved discussion about the results to explain the sentence: "There is good agreement between the MLSS method and the experimental results from samples without surface protection in most cases".Explanation of zync coating was added to the paper as well as explanation of coating modeling. Yellow highlights were removed from text.

 

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

here are some notions and recomendations:

Line 17-18: Please, indicate here, how significant the decrease

Line 22-23 ....approximately 90% of machine failures are a consequence of fatigue.... I strongly disagree here. Maybe, this is a particular case, but, in general, many parts also fail due to corrosion, wear, and deformations. Please, check this point once again

In the introduction, please, describe in more detail, what parts and what materials are studied here. Different materials have dissimilar fatigue behavior (they have or have no fatigue limit)

Line 142-155. The properties data for materials were taken from one source. I think these data are not reliable enough. Could You please add the cited literature here?

Can the annealing of bent/forged samples reduce the material to equilibrium and remove the impact of deformations on their fatigue performance?

Please, describe the technology of coating deposition. What methods and consumables were used? This is a key to understanding, why they effect the fatigue life.

Please, add the information about chemical composition of used material, and the composition of the coating

Line 206: Please, correct here and in other places: It is not a "surface protection". It is a coating. Do not mix "coatings, surface layers, surface protection, surface treatment". They are different technologies.

Line 210. Why does the coating thickness influence the fatigue life? May it be connected with the duration and depth of heating during coating deposition?

Fig. 17-18. Picture, and fonts on it are too big. Please, improve them. Fig 11 is good

Fig. 19. The fatigue crack can not be distinguished. It looks like it was cut by disk or a band saw.

 

 

Comments on the Quality of English Language

The Authors mix some technologies into one. Generally, English style is good.

Author Response

Line 17-18: Sometimes the effect of the coating is greater (lower stress levels) and sometimes the effect is less (higher stress levels). We didn't quantify it in any way, because it is described by graphs.

Line 22-23 - We removed this statement.  

We described parts and materials in the introction in more detail.

Line 142-155. The reference to the cited literature can be found in the text.

The annealing of bent/forged samples can probably reduce the material to equilibrium and remove the impact of deformations on their fatigue performance. We did not deal with this issue in our article.

Informations about coating technology was added to our paper.

The information about chemical composition of used material, and the composition of the coating was added to our paper.

Line 206: Correction of „surface protection“ was done in our paper.

Line 210. Possible explanation of this issue was added to our paper. We don’t think it is caused by temperature. It may be caused by change of the surface quality.

Fig. 17-18. Picture, and fonts on were changed and improved.

Fig. 19. Detailed picture was added. Also, pictures of others broken parts were added.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors made some changes and corrections in the manuscript, but did not provide detailed and specific answers to the comments in the review. Moreover, in the text of the manuscript they did not mark the corrections made in red or in track changes mode. Unfortunately, it is still not known from the content of the manuscript for what purpose the coated samples were tested. The summary lacks information on utilitarian results, i.e. whether the use of coatings had a positive or negative effect on a given type of sample, because the text shows that in one case the coating had a positive effect on samples subjected to bending, and in the case of samples C and D the situation is the opposite. However, there is a lot of information about calculation methods. In this context, the question arises whether if other materials were used for samples, the situation would not be completely different and what would happen to the proposed calculation methods. Perhaps then a different method would give better results, which in this case did not provide high consistency of results.

The reviewer suggests that the authors refer step by step to the first version of the review (I round) and answer the questions posed there. They improved the summary to include comments on utilitarian conclusions. Additionally, it is necessary to mark any changes resulting from the review in the text of the manuscript - preferably in red, so that you can see the places of changes in the text.

Author Response

Dear sir,

we try to supplement the article according to your requirements. We tried to reduce more the "mental shortcuts" that was mentioned in previous review. We also tried to describe the effect of the coating. We marked the current changes in red and previous changes in yellow.

Best regards

Petr Bernardin

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

I accept the changes introduced and marked by the authors. Thank you.

I have 3 more things to clarify. Please answer in 1-2 sentences for each question:

1. I understand that the coating in modeling is using the surface roughness factor and this is an assumption. The question is whether it is good, both in terms of the thickness/type of coating and, above all, the contact surface. Does this roughness factor take into account the thickness and mechanical properties of the coating?

2. In the case of using a coating, is it unfavorable for fatigue strength, and in the case of a simple load, such as the bending process, does such a coating have a beneficial effect and make cracks smaller? This is clear from the observation of figures 20-21 and 22-23. Please reply.

3. What would the situation be like if the coating was harder than the native material?

Author Response

Dear sir,

here are the answers to the mentioned questions:

1. The roughness factor takes into account only the roughness of the surface, nothing else is considered in it. A "coating factor" (that would take into account the thickness of the coating and its material) seems to be suitable for practical use. However, we are not aware that this issue has been described in the literature.
2. We are not convinced that the coating would prevent the spread of the crack (or make it smaller). Rather, it is decisive for the size of the crack when the critical deflection of the sample was reached and the test had to be stopped. It also happened to us that the sample bent due to the cumulative deformation and subsequently the test was stopped without a crack propagation in it.
3. We think, it would be even worse in terms of longevity. The coating would be more fragile and cracks would easily create and propagate into the base material (substrate). However, we don't have any confirmation for this assumption.

 

Best regards

Petr Bernardin