Mapping Acoustic Frictional Properties of Self-Lubricating Epoxy-Coated Bearing Steel with Acoustic Emissions during Friction Test

Reviewer 1

The paper deals with the stick-slip phenomenon during the sliding motion between solid 17 lubricant-impregnated epoxy polymer-coated steel bars and AISI 52100 steel balls. An acoustic sensor was applied for this purpose. The paper is interesting and it would attract attention of the research community in this field of work. However, serious improvements are needed before any final decision can be made:

We are expressing heartfelt gratitude to the reviewer for the dedicated efforts in meticulously revising our paper. The valuable collaboration and insightful suggestions have significantly enriched the quality of our work. We appreciate the kindness and constructive contributions to the refinement of our research.

1) The whole manuscript needs to be double checked for language, grammar or writing mistakes as well as consistency of writing. For instance, already in the abstract, the authors write “stick-slip” and “stick slip”.

The manuscript has been proof read.

Also, add empty spaces between words and references (e.g. line 44: “manipulation[5]”).

Corrected

Remove empty lines between paragraphs in the introduction.

Corrected

Lines 94-95: “…purchased from local were…” – local what? This is grammatically not correct, and not only that, it should be removed as the information repeats in the very next sentence: “All materials were bought from local vendors in Chennai, India.”

Corrected

When a device is mentioned, no need to write every time who produced it (and no need to write “make”)

Corrected.

Line 151: “Before beginning the testing…” This should be simply: “Before the test…”

Corrected.

Lines 178-179: “Figure 5 shows, the coefficient of friction variation for the obtained result at 10N load and 178 at four different spindle speeds 1Hz,2Hz,3Hzand 4Hz.” There should be no coma after “Figure 5 shows, and many empty places missing here, between 10 and N, between 1 and Hz and so on, after commas, etc.

Corrected.

2) The slip-stick effect has a versatile application and this should be mentioned by the authors. For instance, the application commented in the following paper can be mentioned:

Nguyen, X., & Nguyen, H. (2022). Investigation of influences of fabrication tolerances on operational characteristics of piezo-actuated stick-slip micro-drives. Facta Universitatis-Series Mechanical Engineering, 20(1), 109-126. doi:https://doi.org/10.22190/FUME210311036N

Added a new phrase.

“Stick-slip is an intermittent motion that consists of a stationary phase and a sliding phase and has versatile applications such as microdrives [2], geoscience [3] , dental [4], and mechanical systems [5].”

The reference has been cited.

3) “To the best of the knowledge of the authors, no work has been reported in detecting the stick-slip phenomenon of a semi-liquid cured LY556 80 epoxy coating on bearing steel (EN31) using an acoustic emission technique.” The same is valid for many other material pairs. What is so special about this material pair to make it worth of investigation by this specific technique? Is there anything so special that distinguishes this material pair compared to some other material pairs?

Re-phrased

4) Table 2 is given under the subtitle 2.1, but it is referenced under the subtitle 2.2. Such a structure is not acceptable. Furthermore, some of the Tables are not even referenced in the text, which is also unacceptable.

Corrected

5) Figure 2b is too small for what it aims to show.

The figure intends to show the experimental arrangement

6) Lines after equations, in which the terms are explained, starting with “Where…” should not start with a capital letter, otherwise it would be a question.

Corrected

7) “To reduce the influence of ambient noise, the mean of the first one-minute data was subtracted from the entire 60-minute experimental data.” Are there any significant peaks in the one-minute noise? If yes, this procedure would be quite questionable.

Added the new graph 4b.

Explanations were provided:
To estimate the error induced by the ambient noise, the ambient noise was recorded for 60 s (Figure 4b), the results indicating no significant noise level from surroundings. Post-processing of the acoustic signal insured that the influence of the noise is completely removed from the acquired frictional noise.

To reduce the influence of ambient noise, the mean of the first one-minute data was subtracted from the entire 60-minute experimental data. This ensured that only the noise generated by friction was recorded by the microphone. Once the mean was sub-tracted the sensor values were normalized to bring it to the same scale as that of the coefficient of friction [0 1].

8) “Once the mean was subtracted the sensor values were normalized to bring it to the same scale as that of the coefficient of friction [0 1].” Please. Explain the reasoning behind this in more details. How can it be claimed that the maximum sensor signal corresponds to the coefficient of friction equal to 1?

The normalization was done to compare the levels of the COF and the acoustic sensor mapped, i.e., the maximum, the minimum, and the range of COF vs the maximum of the acoustic sensor. Based on this, a mapping between the COF and the acoustic sensor’s value can be generated to standardize the acoustic sensor’s value in terms of COF.

9) Line 175: “…V=Volumetric loss, k=Specific wear rate, S= sliding distance, N=normal load”. Do not use this form. Write sentences, just as you did in places prior to this one. Similarly, rewrite and reformulate lines 198 and 199.

Corrected

10) General rule: define any abbreviation before using it for the first time. For instance, you use COF without defining it (although most readers would easily recognize the meaning).

Corrected

11) Figure 8 is referenced already in page 8, but it is given for the first time in page 10 with several figures referenced after it, but appearing before it. This is unacceptable.

Corrected

12) Lines 240-245 – I have a feeling that the same claim repeats several time in this text.

Corrected

13) “From Figure 7, it can be seen that the acoustic sensor's output is consistent with COF, particularly in Figure 7 (b) and Figure 7 (d). Although the acoustic sensor's range and COF values vary, the acoustic sensor exhibits the same pattern as COF.” Looking at Fig. 7, I could not recognize the correctness of the author claim “that the acoustic sensor's output is consistent with COF”. For instance, in Fig. 7a, the acoustic sensor signal is nearly constant, while COF changes significantly over the same range. How is that consistent? Similar questions may be asked regarding the remaining diagrams.

This synchronization cannot be observed in C1 due to the use of highly complex solid lubricants (Table 2). This, in turn, has reduced significant stick-slip and wear and tear (Fig 6 (b)). The flatness in the acoustic sensor in Fig 7(a) is also due to the averaging effect of the resistor-capacitive components in the acoustic sensor’s amplifier circuit.

Added in the text.

14) The section Conclusions is not correctly numbered

Corrected

15) Conclusions should include limitations of the work done and some clear directions of the future work. Also, explain: “however, a piece of expert advice would always be helpful as the graphs may exhibit slightly different results during tribe-pair interactions.” What kind of conclusion is this?

Corrected

The authors are grateful to the reviewer for these comments. We have tried to answer all the comments. The authors would like to inform the reviewer that the aim of the manuscript was to establish the relationship between the COF and sound generated during the tests. The lubrication mechanisms and parametric influence on the wear and friction has not been the focus of the work. At the end it was observed that the COF and acoustics can be corelated with careful experimental analysis.

1

Please explain the election of table 2 values. Why not using a design of experiments DOE?

The current paper was intended to prove a concept, that the acoustic emission can be successfully applied to monitor the stick-slip phenomena during sliding motion between solid lubricant-impregnated epoxy polymer-coated steel bars and AISI 52100 steel balls. We plan to carry out more complex research in the future and to predict the COF values versus input parameters for a DOE of 25 by using artificial neural networks, ANN (please see our past works: https://doi.org/10.1177/1350650120965754 ). Anyway, we appreciate your useful comments, and any suggestion is welcome.

2

Please define the geometry of the bars 50x20xthickness? What are the circles shown in figure 1?

Thank you for the careful revision. We denoted the dimensions of the cylindrical samples in the revised manuscript as: 50 mm x 20 mm x 10mm. We also mentioned that the tribological test is ball on flat surface type. “This work aims to detect the presence of the stick-slip phenomenon between the solid lubricant-reinforced polymer-coated steel (LY556) and plain steel interaction under sliding conditions at different speeds.”

The circles are the slots for the screws which are required to attach the specimen to the tribometer.

3

Why are the samples C1 to C4 tested at different frequencies? This could be understood form table 3. This should be explained in detail. With a proper DOE you could check if there is an influence of coating or frequency

The DOE (Design of Experiments) protocol was not adhered to in this study, as the primary objective was to assess the feasibility of a cost-effective solution utilizing an Arduino microcontroller equipped with a built-in 10-bit analog-to-digital converter (ADC) and an acoustic sensor. The aim was to capture the acoustic signals generated in self-lubricating coatings with varying concentrations of solid lubricants. Furthermore, the study sought to establish a correlation between the recorded acoustic signals and the friction coefficient throughout the intricate processes of friction and wear.

This text was added to the original manuscript, above Table 4.

4

What is the diameter of the EN31 ball?

Diameter 12.7 mm has been mentioned now.

5

5.- COF is first defined in line 180. Do you mean Coefficient of Friction? First time cited should be described.

COF has been defined now earlier in line no 107.

6

My major concern comes from figure 6. Now you have a correlation between COF and average amplitude. What happens if we have a different geometry? For sure the noise changes due to geometrical issues which change natural frequencies.

In this study, we employed a ball-on-disc configuration, representing a Hertzian contact scenario. The initial contact is characterized by a point contact, which transforms into a line contact through linear movement and further evolves into a shaped surface contact as wear progresses. Throughout the interaction between mating surfaces, sound is generated, with intense friction leading to a more pronounced sound compared to low-friction scenarios. The sound generation during these interactions varies across dry, wet, or self-lubricating conditions, with higher intensity observed under increased friction.

The primary objective of this work was to establish a correlation between the sound produced during friction in a specific set of experiments. Additionally, during the analysis, we observed that irregularities caused by Schellamac waves were detected in conjunction with the peaks of the generated sound. This finding supports the well-established correlation between friction and acoustic signals in the present study.

We appreciate your perspective on exploring different geometries, which will be considered in our ongoing research. Thank you for your valuable suggestions.

7

I think you should repeat the tests at the same frequency for all coatings to avoid the effect shown in figure 6 where you have mor wear rate with less COF for C3.

The relationship between wear and coefficient of friction (COF) is not universally consistent. While it is commonly understood that a high COF correlates with increased wear, this correlation does not hold true in lubricated conditions or when high concentrations of solid lubricants are present. For instance, C3 exhibits a low COF despite experiencing higher wear, attributed to the specific type of solid lubricant it contains. Notably, C3 lacks graphite, a well-known solid lubricant.

The primary focus of this study was to establish a correlation between the frictional coefficient and acoustic signals. To achieve this objective, experiments were conducted using different types of coatings. Through these experiments, we successfully mapped the frictional coefficient and associated sound emission. This accomplishment validates the core correlation between sound generation and friction during the experiments, emphasizing the significance of the work.

8

Figure 7 should be repeated with the same axis for a to d

The primary objective of this study was to establish a correlation between the frictional coefficient and acoustic signals. To achieve this goal, a series of experiments were conducted using various types of coatings. The experiments were carried out once, and through the utilization of different coating materials, we successfully mapped both the frictional coefficient and the corresponding sound produced during friction. This accomplishment reinforces the achievement of our research objective, demonstrating a clear correlation between the generated sound and frictional behaviour.

9

Figure 8 shows that you wear around 30um. What was the original coating thickness? Are already getting friction against the base material?

The average thickness of the coatings was found to be 19.66±3 µm. In certain instances, certain coatings experienced complete abrasion, suggesting an interaction between the metallic pairs involved. The progression of the acoustic signals in cases C3 and C4 indicates direct metal-to-metal contact. The pronounced fluctuations and increased intensity observed in these cases serve as clear evidence of this interaction.

We incorporated this text into our manuscript. Appreciation for your insightful suggestion.

After reviewing the paper "Mapping frictional properties - stick-slip of self-lubricating epoxy-coated bearing steel with acoustic emissions during friction test", here are six points for improvement:

We are expressing heartfelt gratitude to the reviewer for the dedicated efforts in meticulously revising our paper. The valuable collaboration and insightful suggestions have significantly enriched the quality of our work. We appreciate the kindness and constructive contributions to the refinement of our research.

Experimental Design Clarification: The methods section could be more detailed, particularly in explaining the choice of materials and testing conditions. This would enhance reproducibility and comprehension of the experimental setup.

The required section has been added to the revised manuscript.

Data Analysis and Interpretation: The paper could benefit from a more in-depth analysis of the results. Exploring why certain trends or anomalies occur would add value to the findings.

The deviation or change in the pattern of the acoustic sensor to that of COF is attributed to the following:

·        Sensitivity of the sensor, in terms of gain. The higher gain will help the sensor pick up even the faintest of sounds. On the contrary, ambient noise will also be picked up, reducing the efficacity of the data acquisition process by reducing the signal-to-noise ratio.

·        The resistors and capacitors in the acoustic sensor suppress the 60 Hz electrical noise; the static and smooth jitters tend to make the response sluggish, producing an averaging effect. This makes it difficult for the sensor to capture every slick-slip response produced during the experiment.

The next text was added also:

To estimate the error induced by the ambient noise, the ambient noise was recorded for 60 s (Figure 4b), the results indicating no significant noise level from surroundings. Post-processing of the acoustic signal insured that the influence of the noise is completely removed from the acquired frictional noise.

The normalization was done to compare the levels of the COF and the acoustic sensor mapped, i.e., the maximum, the minimum, and the range of COF vs the maximum of the acoustic sensor. Based on this, a mapping between the COF and the acoustic sensor’s value can be generated to standardize the acoustic sensor’s value in terms of COF.

Normalized sensor value=  (sensor value )/(maximum sensor value)                      (2)

Expand the state of the art by adding studies such as: Impact of air pollution on global burden of disease in 2019 10.3390/pr9101719

The mentioned paper is not in line with the present work.

Nonetheless, we thank a lot for this suggestion, and we assure the esteemed reviewer that will use it in our future research.

Comparative Analysis: Including a comparison with other related studies or alternative materials could provide a broader context and significance of the findings.

We cited more studies. The objective of the present work is to find out the capability of acoustic to map the friction between a steel and self lubricating polymer. So, different compositions of lubricating coatings were used. The results were quite capable of mapping the acoustics and frictional properties.

Graphics and Visuals: Improving the quality and clarity of figures and tables would enhance the paper's visual appeal and make the data more accessible to readers.

The quality of the images has been improved.

6. Broader Impact and Applications: Expanding the discussion on the practical applications of this research in industrial or automotive contexts could make the paper more relevant to a wider audience. 

Discussions were added in introduction as highlighted:

“Acoustic emission has been widely applied in many tribo systems, for example, acoustic emission has been used to monitor wear in self-lubricating composite bearing liners used in aerospace systems [39].”

“Hase et al. [43] investigated the wear state of material under sliding friction conditions using a pin-on-disk-type sliding friction tester in the presence of electric current using the acoustic emission frequency spectrum. Maia et al. [44] investigated the AlCrN coated and uncoated cemented carbide cutting tool wear during turning hardened AISI 4340 steel using acoustic emission, in which tool wear and tool wear mechanism is correlated with the acoustic emission signal spectrum using the power spectrum density method and auto-covariance method.”