Material Model Fidelity Comparison for the Efficacy of Predicting Residual Stresses in L-PBF Additively Manufactured IN718 Components

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The work deals with simulation processes for the determination of residual stresses for IN718 processed by LPBF. These models can be used for components with complex shapes.

The work reports interesting parts, but the numerical models should be compared to at least one experimental technique.

Moreover, there are several studies on residual stress evaluation for IN718 processed by LPBF related to the use of experimental techniques. A comparison should be done in order to improve the discussion.

Here I reported the main points that need to be revised for possible consideration for publication:

1)      The work is related to different simulation techniques. However, at least one experimental technique should be used to compare the results with the simulations.

 2) Please report more in detail the negative aspects of the high residual stresses in the component in the introduction. 

3)      I would recommend that the authors discuss whether the current results are different or similar to the other studies where the residual stresses are determined for the IN718 alloy processed by LPBF. This will support the numerical models.

4)      For Figures 8 and 12: it would be better to provide also the standard deviation in the residual stresses determination.

 5)      4. Results. I guess that it is an error and it should be 4. Conclusions

 

 

6)      Overall, the results should be compared with the experimental results in the literature in order to compare it with other scientific work.  

7) The determined trend of residual stresses reported in the manuscript should be discussed (like for Figures 8 and 12). 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Your paper introduces two approaches for residual stress modeling: EPP, which is more computationally efficient, and EMMI, which is less computationally efficient but higher fidelity. For surface C2, EMMI predicted larger residual stresses, particularly near free surfaces, and the calculated stiffness for both models was greater than the experimental observation. For C3, the EPP approach outperformed the EMMI approach for some regions, with the rationale that the EMMI approach has more physics built in but is highly sensitive to its calibration. This is a really interesting conclusion, but it seems to conflict with your statements near the end of the introduction stating that “EMMI answers the demand for more accurate residual stress predictions that are physically motivated”- it seems like there was no case where EMMI actually outperformed the EPP approach due to the EMMI’s sensitivity to calibration. Without some feature that the EMMI approach was able to capture better than the EPP approach, it is hard to claim that the ISV modeling improved accuracy enough to justify the increased computational time. I think your paper should either add more details about a situation where the EMMI’s accuracy definitively outperformed EPP (or was otherwise able to capture some physical behavior that EPP did not have the physics to model) – or it should be reframed around your findings that a higher fidelity model may not necessarily be more accuracy than a lower fidelity one if it is more sensitive to unknown inputs.

 

Some additional minor points of clarification and suggested edits:

 

Page 1, line 42: The transitional phrase “However, the…” doesn’t really make much sense here since you’re not making a counterpoint to anything. Was there a missing sentence about MBAM advantages before this one, since this sentence introduces the disadvantage of high residual stresses?

 

Page 2, lines 45-46: The phrase “the surface between the two layers begins to interact” is a little vague here, can you clarify? The residual stress comes from the remelting and resolidification of material and the associated expansion and contraction, right? I feel like spending an extra sentence giving a more precise explanation about the source of these stresses would be good considering they’re the main focus of the paper.

 

Page 2, lines 54-56: This sentence “One of the most common method employed…” has a grammatical error

 

Page 2: lines 65-69: This introduction to ISV models seems a little bit out of the blue, I feel like this sentence would be better suited for the subsequent paragraph as you haven’t yet explained why using ISV models to capture MBAM thermomechanical effects could be advantageous (which comes on lines 73-77).

 

Some of the section and subsection headings are incorrect: you have three different sections labeled “2.1. Mesh”, there are two sections labeled “3.1.”, and sections 3 and 4 are both labeled “Results”. There’s also some strange formatting going on, the font and font size for some section headings is inconsistent with the others, and the font used for variables from page 8, line 295 through page 9, line 309 is larger than that used elsewhere. Additionally, some reference numbers (such as [26] on page 5, line 181) are italicized while most others are not.

 

Figures 7 through 11: Have you considered plotting a “difference in predicted stress” image alongside any of these? It might be easier to see the difference between the two modeling approaches, though it might not be as useful as I think since it’ll just make the figures larger.

 

Table 3: Your paper is on the efficiency vs fidelity comparison of the two thermomechanical models, but from this table it seems that the thermal model runtime is by far the larger limiting factor. Why is the heat transport model so much slower, and why was it run on 32 threads instead of 60? It doesn’t seem like the element count is that much larger. Is there anything that can be done to accelerate it? Can you include more explicit detail in the methods section on things like the layer height, number of time steps used, etc?

 

The very top of Figure 4 seems to be cut off

Comments on the Quality of English Language

There are a few minor grammatical errors but otherwise the english language quality is good.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors improved the quality of the work.

However, it was time-consuming not having a version that highlighted the modifications. 

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for addressing my comments