Detecting and Monitoring Cracks in Aerospace Materials Using Post-Processing of TSA and AE Data

Round  1

Reviewer 1 Report

This generally well-written manuscript presents an experimental study comparing the crack tip localization performance of thermoelastic stress analysis and acoustic emission via wave time-of-arrival analysis.  The manuscript has been written in the style of a short communication and has been reviewed on the assumption that it has been submitted for publication as such. 

The novelty of the work appears to be in the combined use of full-field stress imaging and AE methods for crack tip localization.  There is some sense in combining these methods.  They operate on fundamentally different physical mechanisms so their combination has the potential to yield insights into crack growth behaviour that are not available from either method alone.  This is reasonably clear from the conclusions drawn by the authors.

The main shortfall of the manuscript is its lack of detail, however this may be justified if the article is a short communication.  The deficit of detail is most evident in the AE component of the work, where many questions can be raised, e.g.

(1) what was the basis for selecting 4 sensors - why not 3, 5 etc.

(2) what was the basis for selecting the sensor positions - was it engineering judgment, was an analysis undertaken ...

(3) why use a training grid that is so coarse.  Presumably a finer training grid along the line of expected crack growth would yield much better performance.

(4) given the localization approach involves empirical training using pencil lead break sources, why were pencil lead breaks not used to evaluate the performance of the trained algorithm along the expected crack growth line prior to the fatigue test. 

(5) no representative AE time histories are given so the reader has no sense of the significance of the AE signatures relative to background noise, nor of the modal composition or frequency content of the emissions.

(6) the AE sensors are said to be nominally 50 mm from the hole.  Fig 1b suggests the lower right sensor is closer to 40 mm from the hole which stretches the meaning of nominal.  A schematic showing relevant dimensions would be useful, space permitting.

(7) the presentation of Fig 3 invites a question that in the opinion of this reviewer does need to be addressed.   The two figures present a qualitative comparison when it seems a more useful quantitative comparison is possible.  This could be easily achieved by overlaying the two sets of results in the one plot.

Author Response

Please see attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper present an interesting application of thermoelastic stress analysis (TSA). The technique is used for measuring the distribution of stress in the surface of a component subject to cyclic loading by using a sensitive infrared camera. The stress concentrations indicative of a crack can be located and tracked using an optical flow method, allowing the position of the crack-tip to be identified at a given time. Acoustic Emission (AE) has been used to validate the TSA algorithm.

The article could be interesting for publication after some minor revision.

As a first hint, the manuscript is written in a good English grammar, only some typos have been found within the text.

In the following a list of suggestions and requests of clarification is given.

Major comments:

(1) In the Introduction at lines 23-26 the Authors provide a list of different non-destructive testing (NDT) methods, without quote any reference.

The Referee suggests to improve the Introduction by providing a widen background (i.e. considering also as a further NDT the electric resistance variation) together with the relevant references.

Cite for example, among the others:

- Lemaitre J., Dufailly J. (1987) Damage measurements. Engineering Fracture Mechanics, 28: 643-661.

- Todoroki A., Ueda M., Hirano Y. (2007) Strain and damage monitoring of CFRP laminates by means of electrical resistance measurement. Journal of Solid Mechanics and Materials Engineering,1: 947-974.

- Niccolini G., Borla O., Accornero F., Lacidogna G., Carpinteri A. (2015) Scaling in damage by electrical resistance measurements: An application to the terracotta statues of the Sacred Mountain of Varallo Renaissance Complex (Italy). Rendiconti Lincei, 26: 203-209.

- Chen B, Liu J (2008) Damage in carbon fiber-reinforced concrete, monitored by both electrical resistance measurement and acoustic emission analysis". Construction and Building Materials, 22: 2196-2201.

(2) From the experimental point of view, the Authors declare that a threshold of 55dB has been used due to the presence of electrical noise.

However, it is known that some of the problems in the use of piezoelectric sensors  is due to rapid changes of capacitance between conductors. In addition, flexing, twisting or transient impacts on coaxial cables could cause electromagnetic noises in the signal (with a spectrum from few Hz to tens THz). This phenomenon is generally called triboelectric effect and can induce false signals generation.

Many Authors tend to ignore this inconvenience, or they declare that no triboelectric effect is observed during their experimentation,  without however providing detailed information on how the latter is eliminated. Does the 55 dB fixed threshold high enough to overcome the problem of the triboelectric effect? It would be very helpful to improve the quality of the manuscript that the Authors spent a few words on this subject.

(3) At lines 98-99 the Authors assess that the acoustic events shown in figure 2a refer to hits detected at three or more sensors. Can the Authors explain why they consider only the AE detected by at least three sensors? Why not considering all the signals, even those monitored by only one sensor? If the Authors, on the other hand, have conducted the experimentation according to a standard protocol, then please indicate which normative has been taken into account as reference. This part seems to be not so clear. Is a misunderstanding by the Referee?

Minor comments:

(1) At lines 69-70 the Authors declare that a silicone sealent was used as couplant between the sensors and the specimen. Why a silicone glue, and not a bi-component glue (i.e. X-60) that is more rigid and tend to better transmit the AE?

The final comment of the Referee is that the manuscript is suitable for publication after the above suggestions have been implemented.

Author Response

Please see attached file.

Author Response File: Author Response.pdf

Round  2

Reviewer 1 Report

The revisions have adequately addressed the points raised.

Author Response

We thank the reviewer for re-reading the manuscript, and for their original comments, which have improved this manuscript.

Reviewer 2 Report

The Referee recognizes the effort done by the Authors to fully respond to the requests.

In general, the manuscript appears to be greatly improved if compared to the first version, and the requests made by the Reviewer seem to be well developed and integrated into the text.

The Referee only reports some typos within the text.

The final comment of the Referee is that the manuscript is suitable for publication after the above suggestions have been implemented.

Author Response

We thank the Reviewer for re-reading this manuscript, and thank them for the comments which have improved this work.

Without line numbers for the typos reported by the Reviewer, we are not able to locate the particular problems they refer to. We are only able to identify words which have been spelt using British English rather than American English. According to the MDPI Instructions for Authors: “American English or UK English are fine so long as there is consistency.”

We have, however, changed references to “Delta-T” to “Delta-T Mapping” for consistency.

Also, for clarity – although the template in Microsoft Word identifies “couplant” as incorrectly spelled, this is in fact a technical term used in Non-Destructive testing and Acoustic Emission work.