Review Reports - Study on Penetration Sensing Method Based on Pool Oscillation and Arc Voltage during Pulsed GMAW

Round 1

Reviewer 1 Report

In principle, this paper is well-written and describes an experimental analysis of PGMAW using primarily optical methods to analyze weld pool oscillations.

The acquisition and analysis of the data are described in an appropriate manner. The conclusions drawn appear valid and reasonable, but I believe some additional discussion is necessary.

Hence, I would like the authors to address the following points before final agreement can be given:

Please discuss if your side-on view will always deliver the correct data used for analysis. In many welding investigations it has been shown that the arc itself, when centered on the bead, pushes the molten material to the outside of the liquid pool, which in your case may change your measured distances between wire tip and molten material.
From your paper I deduce that you are evaluating an oscillation in one direction (up/down), but you will have a superposition of oscillations in longitudinal and vertical direction as well. How does this affect your experiments?
I would like to see a detailed current/trace for the slowest and fastest welding speed (one pulse only) to see if the welding machine really provides you with comparable welding parameters. (we all know that the internal feedback control systems more often than not mess up comparability)
Please discuss the possible influence of metal vapor creation on your results, especially in the case of “slow welding”. It is well known that the occurrence of metal vapor can produce additional forces on the weld pool as well as the measured voltages. How do you make sure that your measurements are not affected by that?
Please re-check the grammar, some minor errors in the paper

Author Response

Dear reviewer,

We deeply appreciate the time and effort you have spent in reviewing our manuscript. Based on your comments and suggestions, we have made a modification on the manuscript and the revisions were highlighted in red in the revised manuscript. Here below are our descriptions of the revisions according to your comments.

Point 1: Please discuss if your side-on view will always deliver the correct data used for analysis. In many welding investigations, it has been shown that the arc itself, when centred on the bead, pushes the molten material to the outside of the liquid pool, which in your case may change your measured distances between wire tip and molten material.

Response 1: Thanks very much for your careful review of our manuscript. Existing research shows [1] that the arc pressure has a negligible effect on the weld pool surface and the molten material of P-GMAW when welding current within the range of 60-200A. The long arc length of the GMA weld pool may be the factors.

The arc pressure can be expressed as:

P_arc=μ₀IJ/4π

where μ₀ is the magnetic permeability of free space, I the welding current and J the current density.

The arc length influences the current density. The shorter the arc length the higher the current density and, therefore, the higher the arc pressure on the weld pool. It is expected that the reduction of the current density with increasing arc length is due to the increased contact area between the arc and the weld pool. While the longer arc in P-GMA welding is required to prevent short-circuiting of the electrode and the workpiece, this will reduce the arc pressure on the weld pool surface. As shown in Paper [1] page 103-104 figure 5.11 and 5.12, a deep surface depression due to the current pulse, as observed on the GTA weld pool in Paper [1], page 19, figure 2.7, is not visible on P-GMA weld pools. All of these indicate that the side-on view images can deliver the correct data with the welding parameters used in our manuscript.

When the welding current is greater than 200A, the arc pressure may have a significant effect on the weld pool surface, and hardly get the exact distances between wire tip and molten material. It is not accurate to judge the measurement distance only from the side view, and the voltage signal under larger current parameters is heavily disturbed by more factors, such as metal vapour in high concentration and unstable transition. The penetration sensing based on the voltage signal under large current condition needs further research.

Point 2: From your paper, I deduce that you are evaluating an oscillation in one direction (up/down), but you will have a superposition of oscillations in the longitudinal and vertical direction as well. How does this affect your experiments?

Response 2: Thanks very much for your careful review of our manuscript. The back and forth motion of the liquid metal in longitudinal in the weld pool can be seen from Paper [1], page 102, paragraph 3. Normally, the travelling wave of the weld pool triggered by the droplet moves to the back end of the pool and reflected from the back edge of the pool to the front edge. While in the case of 40 Hz pulse frequency used in our manuscript a complete back and forth wave motion cannot be observed because when the wave reaches the weld pool trailing edge, the subsequent pulse is applied and another droplet is transferred triggering another liquid wave. A similar phenomenon can be observed in paper 1, page 105. 53Hz pulse frequency was used in paper 1. And the wave speed was found to be almost constant.

All of these indicate that the effect of the surface travelling wave triggered in the last period on the molten pool behaviour in the current period can be ignored, and the superposition of oscillations in the longitudinal and vertical direction has no obvious effect on the experimental results in this paper.

Point 3: I would like to see a detailed current/trace for the slowest and fastest welding speed (one pulse only) to see if the welding machine really provides you with comparable welding parameters. (we all know that the internal feedback control systems more often than not mess up comparability)

Response 3: Thanks very much for your careful review of our manuscript. The detailed current/voltage signals for the slowest and fastest welding speed were shown in the attachment.

Point 4: Please discuss the possible influence of metal vapour creation on your results, especially in the case of “slow welding”. It is well known that the occurrence of metal vapour can produce additional forces on the weld pool as well as the measured voltages. How do you make sure that your measurements are not affected by that?

Response 4: Thanks very much for your careful review of our manuscript. It is known that the presence of metal vapour leads to an increase of the electrical conductivity and the emission coefficient of the plasma. These phenomena have different effects on the voltage. In response to the experimental findings presented in this work. Lowke [2] explained that on the one hand, the increase of electrical conductivity due to the metal vapour tends to reduce the voltage; on the other hand, the increase of the radiation emission tends to cool the arc, reducing the electrical conductivity and increasing the voltage. The markedly increased radiation of the arc can override the effect of increasing conductivity. As an implication, more energy is required to compensate the energy loss so that the voltage increases.

In our manuscript, at the beginning of the base current period, the voltage signal showed an obvious upward trend when the minimum distances between wire tip and molten material showed a downward trend or unchanged, as shown in Figure 14(a)(b)(d)(e) of our manuscript. It is contrary to the trend that arc length is proportional to voltage. The likely explanation of the unusual voltage fluctuation observed in our manuscript is based on the presence of metal vapour in the arc. The metal vapour is excited when the droplet impinges on the molten pool. Then the concentration of the metal vapour gradually decreases in the rest time of the base period, which results in the gradual decline of the electric conductivity of the arc. And the voltage signal trend and the arc length variation trend cannot correspond.

It is found from the results of our manuscript that under the condition of no spatter transition, the amplitude of voltage signal caused by arc length variation is much larger than that caused by metal vapour. While the influence of metal vapour on the voltage signal under other welding parameters and other welding material conditions still needs further study.

We re-checked the whole article and fixed the grammar mistakes.

We hope you will be satisfied with the revisions for the resubmitted manuscript. If you have any queries and suggestions, please do not hesitate to contact me.

Many thanks for your time and consideration.

Best regards

Tao Chen

taocmsc@nuaa.edu.cn

Reference:

[1] Yudodibroto, B.Y.B. Liquid Metal Oscillation and Arc Behaviour during Welding. PhD. Thesis, Delft University of Technology, Delft, The Netherlands, 25 January 2010.

[2] Lowke, J. J., Tanaka, M., and Murphy, A. B., Metal vapour in MIG arcs can cause (1) minima in central arc temperature and (2) increased arc voltages, IIW Doc. 212-1139-09, 2009.

Author Response File: Author Response.docx

Reviewer 2 Report

Although many researchers have focused on the jointing quality by analyzing with various tests after welding, time-dependent variation of melt pool would not be hardly discussed. From this viewpoint, this paper would provide beneficial results for researchers or engineers who want to know the fluid-dynamics behavior of melt pool.

As a disadvantage of this paper, the necessity of experiments is not emphasized enough, though the relative researches are sufficiently introduced. The authors should clarify how to apply their results to welding process, e.g., to enhance the joint strength, to ensure the shape accuracy, or to increase the efficiency. If the purpose of study does not related to any clear merits, the significance of this paper would not be understood by readers. The purpose needs to be more clarified in "1. Introduction."

I would like to summarized other revision points as follows.

Figure 2: The names of each part should be added. If possible, it should be colored to make each part easy to distinguish.
2.3. Penetration Status Definition: I am sure that the difference between the critical and partial penetrations is not obvious. Although both of them are categorized into Fig. 9(a), I could not understand why you could separate the situation into two categories. Is there any concrete criteria in this classification?
Figure 9: The "external force" is a strange expression in this case. If this arrow express the "force," the vector needs to be started from the acting point. Furthermore, it is strange that the same arrow are used for expressing various different factor, force, flow and welding direction.
Figure 9(b): The horizontal flow would be generated in the melt pool as (a) too.
Figure 11: The blue lines are difficult to distinguish from the green back screen.
General comment: Many grammatical mistakes are found in the text. I strongly recommend you to receive a check from an native English speaker.

Author Response

Dear reviewer,

Point 1: If the purpose of the study is not related to any clear merits, the significance of this paper would not be understood by readers. The purpose needs to be more clarified in "1. Introduction."

Response 1: Thanks very much for your careful review of our manuscript. In order to emphasize the necessity of experiments, the sentence “the researches about the relation among the weld pool oscillation, the penetration status and the voltage signals are of great significance to realize P-GMAW weld penetration sensing based on voltage signal with low cost and high reliability.” was added in Page 2, Line 80-83 to illustrate the significance of this paper.

Point 2: Figure 2: The names of each part should be added. If possible, it should be coloured to make each part easy to distinguish.

Response 2: Thanks very much for your careful review of our manuscript. we redrew Figure 2. The names of each part were added.

Point 3: 2.3. Penetration Status Definition: I am sure that the difference between critical and partial penetrations is not obvious. Although both of them are categorized into Fig. 9(a), I could not understand why you could separate the situation into two categories. Are there any concrete criteria in this classification?

Response 3: Thanks very much for your careful review of our manuscript. It is impossible to conduct a metallographic analysis of the weld on the workpiece in real production. Generally, the status of penetration is determined by observing the back of the weld. The back of the partial penetration weld is the continuous unfused groove, while the shape of the bead on the back of the critical penetration weld is discontinuous and constructed by discrete fusion points. We redrew Figure 3 to emphasis the difference between the critical and partial penetrations.

Point 4: Figure 9: The "external force" is a strange expression in this case. If this arrow expresses the "force," the vector needs to be started from the acting point. Furthermore, it is strange that the same arrow is used for expressing various different factor, force, flow and welding direction. Figure 9(b): The horizontal flow would be generated in the melt pool as (a) too.

Response 4: Thanks very much for your careful review of our manuscript. We redrew Figure 9 and fixed the errors. “the force” was substituted by “Direction of arc pressure and impact of the droplet”. Arrows with different colours were used in this Figure to express various different factor.

Point 5: Figure 11: The blue lines are difficult to distinguish from the greenback screen.

Response 5: Thanks very much for your careful review of our manuscript. The coefficient matrix was added in Figure 11, The rendering colours of different locations of scale vector contour map depend on the ψ_a,b(t). The black contour is useful information to calculate the signal real-time frequency.

We re-checked the whole article and fixed the grammar mistakes.

We hope you will be satisfied with the revisions for the resubmitted manuscript. If you have any queries and suggestions, please do not hesitate to contact me.

Many thanks for your time and consideration.

Best regards

Tao Chen

taocmsc@nuaa.edu.cn

Reviewer 3 Report

This paper discusses a high-speed camera and image analysis system that used to extract the oscillation characteristics of the molten pool. The intrinsic relations between the oscillation characteristics of the molten pool, the voltage signal curve and the state of the penetration depth of the weld inP-GTAW were studied. The results show that as the penetration depth is proportional to the amplitude of the molten pool oscillation. With the transition from partial penetration to full penetration, the frequency component and amplitude of the molten pool oscillation suddenly changed. On the other hand, due to the curvature of the welding pool surface, the voltage signal curve loses the oscillation frequency characteristics of the welding pool. Although similar to the amplitude of the oscillation, the fluctuation of the voltage signal caused by the oscillation of the molten pool reflects the penetration of the molten pool. In the basic duration, the sudden change in the amplitude of the voltage signal from partial penetration to full penetration can be used to detect the penetration of the weld pool in real time.

In summary, this paper describes very interesting and important results for both academics and industry, thus it should be a good paper for the Journal “Applied Sciences”.

However, the authors need to improve their writing (including grammar/syntax) from head to toe for the whole paper. I can find more than several dozens of grammar mistakes only for page 1st. So please rewrite the whole paper with good care.

Author Response

Dear reviewer,

We hope you will be satisfied with the revisions for the resubmitted manuscript. If you have any queries and suggestions, please do not hesitate to contact me.

Many thanks for your time and consideration.

Best regards

Tao Chen

taocmsc@nuaa.edu.cn

Round 2

Reviewer 3 Report

Please consider to edit the whole paper with care. You should understand that English without ambiguity and grammatical errors is very important. And bad English will have a fatal blow to the reputation of the entire journal. In the attachment please find my humble suggestions to the 1st paragraph.

Comments for author File: Comments.pdf

Author Response

Dear editor and reviewer

We have substantially revised our manuscript after reading your kind advices and

the comments provided by the reviewers. All the revisions have been highlighted

in the revised manuscript. We sincerely hope this manuscript will be finally

acceptable to be published on Applied Science. Thank you very much for all your

help and looking forward to hearing from you soon.

Point: Please consider to edit the whole paper with care. You should understand

that English without ambiguity and grammatical errors is very important. And

bad English will have a fatal blow to the reputation of the entire journal. In the

attachment please find my humble suggestions to the 1st paragraph.

Response: Thanks for your precise comments. We are very sorry that our

revision did not make your satisfaction. We revised the 1st paragraph according

to your suggestion, and We have checked all spellings and grammatical mistakes

carefully. We have used the "Track Changes" function in Microsoft Word so that

the details of the revisions are easily visible to the editors and reviewers.

We hope you will be satisfied with the revisions for the resubmitted manuscript.

If you have any queries and suggestions, please do not hesitate to contact me.

We deeply appreciate the time and effort you have spent in reviewing our

manuscript. Your comments and suggestions are all valuable and very helpful

for revising and improving our paper.

Once again, many thanks for your time and consideration.

Best regards

Tao Chen

taocmsc@nuaa.edu.cn

Author Response File: Author Response.pdf