Design of a Tree Micro Drill Instrument to Improve the Accuracy of Wood Density Estimation

Round 1

Reviewer 1 Report (Previous Reviewer 2)

A first important point is that the changes made to the original manuscript are not traced and this makes it difficult to make some assessments.
However, the current text has no substantial improvement over the first one I have assessed. The only real improvement is in the details of the results for the different timbers, which made me more understandable some evaluations.

My impression that the authors have little experience with wood in general, and in particular with the use of resistographic drills and the interpretation of the results, is confirmed. Most of the findings I made in the review were not included. One example is my notation that all the different applications of the resistographic drill derive from what the equipment can estimate: density differences, through a mechanical action on wood. It is the relationship between these two aspects that must be understood to interpret the results.

This would help to understand the differences between the characteristics of wood in the living tree and those on dry wood, and how the equipment is able to estimate its characteristics. It would allow the authors to understand why the tests done by them are not on trees (as they often say) but on dry wood. We cannot speak of tests on trees (lines 219-221 and captions of figures 6-7-8 and line 397-400), because wood is not in green conditions, but oven-dried.

Finally, it is quite clear to why the estimation doesn not work on oak wood, but it is not up to me to give the explanation.

Author Response

1. I'm sorry that I was on a business trip when I received the first paper revision comments and couldn't make the necessary revisions.
2. According to your suggestions, we have made the necessary modifications.
3. Your suggestion is very good, we will perform living tree test in the future.

Author Response File: Author Response.docx

Reviewer 2 Report (New Reviewer)

Summary:

The manuscript aims to design a microdrill for accurate wood density estimation and compares self‑made microdrill with commercially available Rinntech Resistograph 650-SC. The authors conclude that their self-made microdrill is more accurate and that it is beneficial to consider both rotation and feed resistance.

Comments:

There are several machines based on drilling resistance available. I understand that full details of their construction are part of companies’ know-how and thus are not publicly available. It is then however hard to say how much your tool differs from the others. Could you please try to describe the main differences? Are you planning to supplement full a documentation for your microdrill so that the measurements are reproducible by another research group? The blurry Figure 1 does not seem to be sufficient.

You emphasize that your microdrill is unique because, unlike Rinntech Resistograph, it also records feed resistance. However, it appears that IML power drill can also record feed resistance.

Lines 55–58: Trademark “Resistograph” belongs to Rinntech company as they patented the tool first. It does not necessary mean that it is the best.

Line 73: Isik et al. used IML-RESI B400.

What is the resolution of your microdrill?

How did you calculate F1, F2 and FR? Did you include only the middle part of the resistance curve? Could you please highlight the included part in Figure 4?

Did you perform any adjustments of the resistance curve? Did you notice any curve elevation caused by needle friction?

Is Figure 4 based on a measurement of your microdrill or Rintech resistograph? Could you show two graphs for the same sample and the two tools?

Line 259: “based on”

Tables 1–10: I suggest that comments to the tables’ content are below the tables and not as a part of their captions.

Figure 5c is missing.

Please include figures for LM3 with R2 so that it is possible to compare all four models.

Please include model name in the figures’ caption or as a figures’ title.

Line 299 and 300: “our newly designed”

Table 9: F2 is not significant. Actually, none of the parameters is significant.

Could you please discuss in a more detail why you were not successful in estimating wood density in oak and how that could be improved?

Overall, the discussion is missing any connection with other studies.

Did you test you microdrill outside the laboratory conditions as well? It would be interesting to see how it performs in “real life”.

Author Response

1. There are several machines based on drilling resistance available. I understand that full details of their construction are part of companies’ know-how and thus are not publicly available. It is then however hard to say how much your tool differs from the others. Could you please try to describe the main differences? Are you planning to supplement full a documentation for your microdrill so that the measurements are reproducible by another research group? The blurry Figure 1 does not seem to be sufficient.

       Due to technical confidentiality, we are not aware of the detailed technology       of Resistograph and IML resi, and we cannot conduct comparative analysis on technology. We have opened some design proposals and hope that more researchers will conduct research on the micro drill resistance instrument.

2. You emphasize that your microdrill is unique because, unlike Rinntech Resistograph, it also records feed resistance. However, it appears that IML power drill can also record feed resistance.

       IML power drill can also record feed resistance, but the users are not aware of the detailed technology.

3. Lines 55–58: Trademark “Resistograph” belongs to Rinntech company as they patented the tool first. It does not necessary mean that it is the best.

Revised.

4. Line 73: Isik et al. used IML-RESI B400.

Revised.

5. What is the resolution of your microdrill?

Added.

6. How did you calculate F1, F2 and FR? Did you include only the middle part of the resistance curve? Could you please highlight the included part in Figure 4?

Revised.

7. Did you perform any adjustments of the resistance curve? Did you notice any curve elevation caused by needle friction?

I didn’t perform adjustments of the drill resistance. I noticed the curve elevation caused by needle friction, but the wood length is 3 cm, the needle friction is little compare to the tip friction.

8. Is Figure 4 based on a measurement of your microdrill or Rintech resistograph? Could you show two graphs for the same sample and the two tools?

Revised.

9. Line 259: “based on”

Revised.

10. Tables 1–10: I suggest that comments to the tables’ content are below the tables and not as a part of their captions.

Revised.

11. Figure 5c is missing.

Revised.

12. Please include figures for LM3 with R2 so that it is possible to compare all four models.

We can't find a suitable graphical form that can present the R2 of LM3.

13. Please include model name in the figures’ caption or as a figures’ title.

Revised.

14. Line 299 and 300: “our newly designed”

Revised.

15. Table 9: F2 is not significant. Actually, none of the parameters is significant.

Yes, this may be due to insufficient experimental data.

16. Could you please discuss in a more detail why you were not successful in estimating wood density in oak and how that could be improved?
17.  
18. Overall, the discussion is missing any connection with other studies.

Most of the literature studies the wood density of standing wood, but we measured due the absolute dry wood density, there is no comparability.

19. Did you test you microdrill outside the laboratory conditions as well? It would be interesting to see how it performs in “real life”.

At present, our prototype cannot be tested in the field. We will gradually improve our prototype and conduct field experiments as soon as possible.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report (Previous Reviewer 2)

The paper certainly improved, but only in the discussion. Many of the elements included in the new discussion actually need to be moved to the introductory part.

I would advise authors to deeply rethink the layout of the work and write a well-balanced one in the different parts.

For example, the very detailed part about the design and implementation of the new tool is certainly excessive. On the other hand, there is no indication of the feed and rotation speeds during the tests, while all available equipment allow speed adjustment based on the density of the wood to be drilled. If I remember correctly the real Rinntech Resistograph adjusts automatically the speeds on the basis of the resistance opposed by wood.

I would like to emphasize a couple of basic notations:

-The advancement of a thin drill tip inside wood is an indirect method of estimating (not measuring) the density of wood. All the different uses of this resistographic methodology listed in the paper derive from this. The estimation of the thickness of the annual increments also derives from the possibility of the tip to estimate the differences in density between early and late wood.

- The sensitivity of the instrument finds a limit in the ratio between the size of the tip and the thickness of the annual rings. This is a geometrical problem: if the thickness of the rings is less than or much less than a millimetre, then the 3 mm wide tip finds itself entering several rings at the same time, fatally mixing the measurements. The problem, of course, cannot be solved by varying the engines or speeds.

- Secondary is the question of the measurement made forward and backward. Surely the IML I used measures only forward. But it still seems very strange to me thinking of doing backward resistance measures.

Author Response

1. Resistograph can not adjust feed speed on the basis of the drill resistance.
2. To avoid misunderstandings, we have added the drill geometry in the paper.
3. To avoid misunderstanding, we have made modifications.

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

The paper deal with the design of a resistance drilling instrument and the comparison of its measurements with another commercial instrument to estimate the density of wood.

The work has several flaws that needs to be addressed and should be rejected in its current form.

Generally speaking,

1-    English should be improved

2-    the description of the instrument is far too detailed, while the experimental part is not sufficiently described and discussed

3-    some methodological errors should be addressed

4-    the statements are not always supported by the results.

Specific comments:

Abstract

Line 17: softwood and conifer are usually used as synonymous.

Lines 19-20: the verb is missing.

Lines 23-25: this sentence is not properly correct, see the comments further on.

Line 26: “increasing the feed”, what does it mean?

Introduction: most of the literature cited is about the estimation of density in standing trees, but the drilling resistance instrument are used much more in timber structures and there are several studies on beam density prediction in existing buildings.

Lines 71-72 and 84-85 are repeated.

Lines 97-99: is this statement from the literature? Please, cite the reference source.

Lines 103-113: too repetitive.

Lines 123: the paper?

Lines 327-328: are these the settings used during the experiment?

Lines 341-347: English is not correct.

Lines 348-350: the authors decided to measure the drilling resistance of the wood in dry state, but this is a condition never found in the common situations (the standing trees has higher moisture content, and the timber structures usually are at the equilibrium moisture content, 12-15%). To avoid the effect of the moisture content on the drilling it was sufficient to condition all the specimens under the same environmental conditions.

Lines 376-380: not clear!

Lines 383-385: improve English.

Lines 389: the third model is not showed in Figure 7.

Figure 7: the figure shows that the different density of the several species is affecting the results. When there are two separate groups like here (very high density and lower density) the linear regression approaches to a line through two points. This is the reason you get so high coefficient of determinations and not because the prediction is so good. Please, keep wood species separated.

Table 1: In the regression, F2 is not significant (p>0.05). This means that including F2 in the regression does not add any improvement to the prediction; exactly the contrary of what the authors claim.

Table 2: LM3 in the Figure 7 is another model. Furthermore, LM1 and LM3 has a very similar R2, a consequence of the fact that F2 is not a significant variable (see table1). If you compare the models with a statistical analysis LM1 and LM3 will result not different at all.

Lines 432-434: where does this statement come from?

Line 440: “measurement accuracy”, you are not measuring density, you are estimating it!

Line 455: researchers  research

Line 464-467: incorrect, see comment above.

English is often incorrect and not clear.

Reviewer 2 Report

General remarks

Non-destructive or low-invasive equipment for estimating wood density are in great demand on the market because estimating wood density is a first parameter of great importance for the physical-mechanical qualification of wood.

Being able to estimate the density of the wood in plantations without having to cut down the trees ahead of time can allow better planning of the cultivation treatments in view of a certain objective, while in the diagnosis of historical wooden structures, the estimation of the density is fundamental both for evaluate the presence of degradation and to better set the structural parameters of the individual elements.

Further applications may still be numerous, but always derived from applications based on the punctual estimation of density, such as for the variations between early and late wood within the growth ring.

This paper fits into this groove, trying to develop an equipment that allows very high precision measurements in the punctual estimation of the density of a wood sample, using as a benchmark one of the most precise micro drilling tools on the market.

From a general point of view, my first remark concerns the title of the paper: reference is made to "tree micro drilling", while the tests on wood are made on completely anhydrous specimens (at least that's how it seems). So if water content is an important parameter (and it is), and the tests are done on dry wood, then this equipment is not parameterized for standing trees.

In my opinion, another aspect to correct is that this paper is very unbalanced on the technical description of the instrument (6 pages out of a total of 12 written), while the more scientific part which concerns the execution of the tests, data processing and discussion they are not very detailed and leave many doubts about the repeatability of the test.

Furthermore, the extreme detail of the design and construction of the new equipment does not correspond to any description of the benchmark equipment, so it is not possible to understand how the new instrument improves on the one found on the market.

Finally, parts of the paper written in easily understandable English alternate with other paragraphs with verbs only in the infinitive (or imperative? but it would be incoherent) and difficult to understand the text. A revision is definitely needed, in my opinion, even though I'm not a native speaker.

Detailed remarks

Lines 58-60: The starting point is always the density estimate, the other parameters are all derived.

Lines 65-68: I guess from my personal experience that the problem is related to the geometry of the tip: being 3 mm wide, when the rings are less than 0.5 mm thick it is clear that the tip digs more than one ring at the same time, mixing the density of early-late wood of several rings. If this underestimation error is written in this way, without taking into account the geometric parameter, it almost seems that the same error can always be considered for a given wood. Written like this it is misleading.

Lines 81-84: The comparison between micro drilling and core analysis is also misleading here: it is clear, as written by the authors in the discussion, that microdrilling is much less harmful to the wood of the standing plant and to the plant itself. It is strange that in the previous paragraph we write about how accurate the PD400 IML is in estimating density and in the following one we claim the opposite. Probably the density estimate parameter is effectively estimated and the ring reading parameter is not (at the end of discussion there is a similar consideration). But it's not written well. It is unacceptable that between two paragraphs one thing is said and then the opposite.

Lines 339-341: different timbers are declared only here, then they disappear (eg. Which wood is it in figure 6?). What is the expected density of each wood species, according to the literature and according to your experimental measurements?

Line 344 and following: are the wood block oriented correctly to wood directions? Following paragraph: were dills operated on completely dry (anhydrous) wood blocks? Can authors better explain how blocks were clamped? The correct procedure is not completely explained.

Figure 6: what wood species is this one?

Lines 376-392: very difficult to follow and understand. Please, check the English. So the average resistance (rotation/feed) value was compared to the mean density of the block? I mean: the comparison was between a single value of the resistance and a single density of the block?

Lines 424-427: IML PD400 gives both resistance measurements, rotation and feed. Why your benchmark is only Resistograph?

Lines 453-456: why nothing is said about a possible improvement of the needles, like tip typology, etc.? From this paper, but also from the personal experiences, a tool giving an estimation of wood density around 90% is a very good achievement. The problem is probably not the density estimation but growth ring measurements.

Lines 468-470: What does it mean “increasing the feed resistance of the drilling needle”? Increase the mechanical performances of the needle?