General comment:

The paper makes a significant contribution to scientific progress in analysis and data with regard to WES scope. The main objective of the paper is to show the effect of a specific site condition on the wear in the pitch bearing at the below-rated wind speed. The study is new and has not been studied before.

The paper did a study on the pitch bearing wear. The results are qualitative, not quantitative, but have interesting findings. The test cases are different bearings, and a real multi-megawatt pitch bearing of a wind turbine is used in the cases, which is beneficial to apply to large wind turbines. The experiment procedure are described thoroughly. Moreover, the results are presented quite in depth.

There are general notes as follows:

• As there are three different heights of measurement, it is not clear from which height wind shear and turbulence intensity are considered for simulation. Please add it to the related section.
• Please explain why six seeds are used for each wind speed or reference it. According to IEC 61400-1, in the ultimate analysis, six seeds were used at a (below-rated wind speed – 2 m/s).

Specific comment:

1. In Section 2.1, please add the goodness-of-fit of the Weibull distribution.
2. The phrase "the wind speeds do not correlate well with any of the IEC wind classes" in Section 2.1 needs to be quantified.
3. It is not clear what the height of measurement is in figures 1 and 2. Please add the height in both figures.
4. In figure 2, the last bin does not have a mean plus (red asterisk) value.
5. In Section 2.1 and related to Figure 3, it seems that heights of 119 and 139 are used for the calculation of the shear exponent, and the projected wind speed results on 159 meters are compared with measurements. If that is the case, please include the results of the comparison. Otherwise, explain more about the process and comparison with measurement.
6. In the phrase "While the shear induces more pitch activity, the low TI leads to less variation in the oscillations of the bearings. It is therefore reasonable to assume a more wear-critical operation under the measured wind speeds in comparison to IEC classes." It is not clear how this judgment was reached. Please explain more.
7. Wrong referred In line 238, Eq. (1) should be changed to Eq. (2).
8. It seems Equation 5 has the wrong term. It has to be rewritten.

The authors have undertaken an examination of an oft-talked about but seldom looked-at relationship between wind turbine controls and reliability: in this case individual pitch control and risk to the blade bearings because of wear. Having said that, I will admit I had a bit “higher hopes” for the article based on reading the Abstract and Introduction as compared to the Test Results, Discussion and Conclusions. Upon first reading the former, I had the thinking that the work would examine, assess and/or quantify how “lubrication runs” as enabled by individual pitch control (IPC) could reduce the risk of pitch bearing wear. However, the latter was much more limited and described a process by which the characteristics of a particular wind site, turbine, pitch bearing, and controller with IPC were distilled and shown to result in mild wear in a relatively short period of time on both scaled bearings and the full-scale bearing on different test benches. I don’t necessarily want to downplay the importance of this conclusion that mild wear can occur quite rapidly in certain conditions, as it is indeed important. It’s just that I thought that the paper might offer a means to first assess or even quantify the risk of wear and then validate that it can be avoided through validation testing by comparing the test results with and without certain individual pitch control actions. This primarily relates to how key sentences in the article are worded as noted in specific comments that follow. After giving some thought about trying to sum up the article in a single sentence, I personally would conclude that the work shows that “Pitch bearing wear can quickly result from individual pitch control”. Can that be as broadly implied as written? Or does it entirely depend on the nature of the site, turbine, and such? Can we quantify in anyway what we thought the risk of the conditions might have been, compared to what they were demonstrated to be? Then again, maybe a more important point is that the actual site-specific characteristics might give different results than the standard IEC classes – maybe the latter would yield little to no wear? I guess that’s my point, I am not sure I understand what the authors think the main conclusion(s) from the article to be, and then focus on it/those throughout.

Aside from this more general overall comment, I also provide the following more specific comments that are mostly tailored to allowing a reader to a bit more easily understand the presented material.

Title

I have to wonder, is the aspect of “site-specific” worthy of being in the title? It is certainly true that this examination was accomplished using the particular wind characteristics of an actual site and for a reference (or representative of the state-of-the-art) large, modern wind turbine. But, I believe those to be secondary. Would the authors agree? It was also not shown/quantified (or at least I didn’t pick up on it) that the site-specific conditions were any more problematic in terms of wear than standard IEC wind classes and NTM. I think a simpler title like “The effect of individual pitch control on blade bearing wear” is more accurate and better in line with the Test Results, Discussion and Conclusions.

Abstract

I think it is common that the Abstract should be a single paragraph. I also think what is now the second paragraph can be improved to be more specific as in “This work demonstrates that a site-specific wind and pitch controller conditions can quickly result in pitch bearing wear.” That is, the article doesn't generally analyzes (plural) “wind and operating conditions”, just one particular combination. OK, across wind speeds yes, but really seems to be specific to this particular site. I also recommend adding the adjective “mild” to wear in the Abstract to be in line with the last sentence in Section 6.

1 Introduction

There are some minor grammatical changes needed here. In line 29, I recommend “…for evaluating the wear risk to pitch bearings…” This paragraph is also only 2 sentences long. I believe it can be combined with the next paragraph beginning on line 31.

The sentence describing theta_crit in lines 32-33 can be improved, specifically “gets in touch”. I believe per DG03 and per Rumbarger and Jones that theta_crit is “the relative angle of rotation of the raceways for which the portion of the raceway stressed by one rolling element touches—but does not overlap—the raceway that is stressed by adjacent rolling elements.”

The sentence on lines 36-37 also says “Tests on small scale bearings performed by showed…”, which reads as though a name and/or reference is missing, or “performed by” can be deleted. I suggest the former though.

I will admit I got a bit confused reading lines 40 – 47 and “the work related to this paper focuses”, lines 53-54 and “aim of this test program is to”, and lines 57-58 and “The study in this paper is not”. I kept thinking that they were describing this manuscript, when I think they are describing nearby citations. I recommend these be past tense and/or “this” changed to be more specific. The statement about this manuscript does not come until lines 76-79. Generally speaking, I found lines 45-75 a bit hard to follow.  It might also help to begin a new paragraph with “Stammler (2020)…” on line 47 and maybe trim some of this material in lines 47-75. At a minimum, I would suggest defining the acronyms for HBDV, HAPT, and iBAC, and if possible adding a citation. I’m not sure simply mentioning them is worthwhile if no other information is given.

2 Wind data analysis

2.1 Wind data characterization

The colors in Figure 1 (bottom) were a bit hard to sort out. I might suggest that because the Weibull distribution fit is not mentioned in the text, it can be removed from the figure – or do the listed a and b values quantify that the measured wind speeds “do not correlate well with any of the IEC wind classes”?

2.2 Wind data and turbine operation conditions

I had difficulty in following the descriptions in this section. At first, I thought this section was intended to describe the process by which wear-critical operations are determined, but then the first sentence in Section 3 says that is the point of Section 3. I think Section 2.2 maybe just gives an example of a particular interesting/likely to induce wear condition, but it’s probably longer and more convoluted than needed at several pages and 3 figures.

4 Simulation results and test program

4.1 General description of the test program and simulation results

I think some of my confusion about Section 2.2 is not surprising, as the first sentence in Section 4.1 says “To evaluate if the critical sequence described in Chapter 2 is in fact critical in terms of wear...” That is, what is the difference between “critical sequence” and “critical (sequence) in terms of wear”? I found the process description and Figures 10 and 11 helpful here – it just felt like it took a long time to get to this point.

6 Discussion and 7 Conculsions

The statement in 6 “While the test results of type 7220 bearings and real size blade bearings of type 51818 show almost similar results, the results from the scaled blade bearings of type 13229 and the real size blade bearings of type 12480 are different” seems to conflict with the statement in 7 that “Despite the differences in size and bearing design, the test program leads to comparable wear damages on the raceways.”

In 7 Conclusions and to my earlier questions about the importance (or not) of “site-specific” conditions to pitch bearing wear, where it is stated in the first sentence on line 515 that “…wear…can occur quickly under certain operating conditions…” I do agree that this has been shown but more than anything what I do not have a clear picture of is exactly what “certain” means here. What exactly are the characteristics that lead to wear? Can we say, or is every site a snowflake? I take it the IPC characteristics and maybe the turbulence and/or shear are the most important things? Or maybe the main point is that the article proposes a process by which a particular set of characteristics can be assessed for a given turbine/controller/bearing, but then again leads me back to the thinking that every site is a snowflake. That is, the characteristics of any given site must be known and then tested to see the extent to which wear may result and then the IPC potentially changed, which is implied where it says “It can be adapted to other turbine and controller designs as well as to other site-specific wind conditions.”