Next Article in Journal
Relationship between Peak Stage, Storm Duration, and Bank Storage along a Meandering Stream
Next Article in Special Issue
Effects of Spur Dikes on Water Flow Diversity and Fish Aggregation
Previous Article in Journal
Red Lentil (Lens culinaris) Extract as a Novel Natural Coagulant for Turbidity Reduction: An Evaluation, Characterization and Performance Optimization Study
Previous Article in Special Issue
Quantitative Assessment of the Influences of Three Gorges Dam on the Water Level of Poyang Lake, China
Open AccessArticle
Peer-Review Record

Physical Modeling of Ski-Jump Spillway to Evaluate Dynamic Pressure

Water 2019, 11(8), 1687; https://doi.org/10.3390/w11081687
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Water 2019, 11(8), 1687; https://doi.org/10.3390/w11081687
Received: 18 June 2019 / Revised: 11 July 2019 / Accepted: 11 July 2019 / Published: 15 August 2019
(This article belongs to the Special Issue Environmental Hydraulics Research)

Round 1

Reviewer 1 Report

(a) The authors should explicitly state in the introduction what is novel in this article.

(b) The authors should be aware of the fact that the font size (of texts and equations) should be adjusted.

(c) Please improve Figures-4, 5, and 6.

Author Response

Response to Reviewer 1 Comments

We thank you for providing us the useful and valued comments. In the revised manuscript, we have incorporated all your suggestions and other Reviewers’ comments to improve the quality of the work. Our response to your comments is given below.

(a) The authors should explicitly state in the introduction what is novel in this article.

We thank you the referee for concerning the novelty of the paper. In this research, the entire dominant parameters affecting the pressure at a ski jet were investigated. In addition, apart from most literary works, we used various angles of impact plate. The objectives of the works have been modified as requested by the referee. Kindly see lines 71-73.

 (b) The authors should be aware of the fact that the font size (of texts and equations) should be adjusted.

This has been modified in the entire text. Thanks.

 (c) Please improve Figures-4, 5, and 6.

Thanks for your useful comments. This has been modified and the Figures (4, 5, and 6) were improved.


Reviewer 2 Report

Measurements of pressures caused by a ski-jump spillway are measured in laboratory experiments.  The experiments are carefully conducted, the measurements are detailed and clearly presented, and there is some useful discussion of the results.  The paper is well-structured, and the figures are clear and appropriate.  Overall this paper would be suitable for publication, but there are a number of minor matters to address first, detailed by line number below.

 

18 First, a general point: I don’t think the term “sidewall” is appropriate for a surface that is often close to horizontal and I found the term confusing.  I suggest it would be better to use something like “impact plate” or “impingement plate” throughout.

 

19 by ski-jump buckets (delete “the”)

20 depths were used.  Discharges of… were chosen.

22 by a transducer.

27 depth varied from

32 pressures resulting

33 scouring [1,2].

35 by a ski-jump

39 on the deflection

44 results show that

45 bubbles disappeared as

46 was reduced… caused by submerged..

49 based on the major effects of

54/5 pressure into dynamic and hydrostatic components

56 time step procedure

58 in a physical

61 results show that

62 fluctuations grow.

79 Adapted from

81 Region A is composed of three sub-regions: A1, A2..

90 turbulence

91 with less energy (delete on downstream)

99 is the acceleration due to gravity (or the gravitational acceleration)

119 It is 75.5 m in height

123 Four discharges of 67,..

124 real discharges of

126 pool were 0, 15,..

151 Piezometers were used to observe the fluctuations of the dynamic pressures on the..

189 part of the jet is

191 or has no significant reduction, while, from a certain depth, this..

199 impingement of the core.. is a result of the plunge pool

202 consequently so does.. As seen in Figure 4,..

208 reduction in the

228 along with increases in discharge.

248 it decreases away from the

251 At 30 cm depth (delete Reaching)

254 decline in Cp

258 the highest pressure

260 This may be related to the

280 relative to the horizontal increases

281 decreases in all depth situations

283 behind this is that

289 and vortices are

294 a lower dynamic

306 The greater the plunging depth

308 A low plunging depth did not give a reduction of

Author Response

Response to Reviewer 2 Comments

Measurements of pressures caused by a ski-jump spillway are measured in laboratory experiments.  The experiments are carefully conducted, the measurements are detailed and clearly presented, and there is some useful discussion of the results.  The paper is well-structured, and the figures are clear and appropriate.  Overall this paper would be suitable for publication, but there are a number of minor matters to address first, detailed by line number below.

 We thank you for the encouraging comments. In the revised manuscript, we have attended and incorporated all your suggestions to improve the quality of our work. Following is our response to your comments:

 

18 First, a general point: I don’t think the term “sidewall” is appropriate for a surface that is often close to horizontal and I found the term confusing.  I suggest it would be better to use something like “impact plate” or “impingement plate” throughout.

 We change the “sidewall” to “impact plate” entire the manuscript. Thanks for your careful comment.

 

19 by ski-jump buckets (delete “the”)

Done. Thanks for your accurate comments on the paper presentation. We have modified the sentences as requested.

20 depths were used.  Discharges of… were chosen.

Done

22 by a transducer.

Done

27 depth varied from

Done

32 pressures resulting

Done

33 scouring [1,2].

Done

35 by a ski-jump

Done

39 on the deflection

Done

44 results show that

Done

45 bubbles disappeared as

Done

46 was reduced… caused by submerged..

Done

49 based on the major effects of

Done

54/5 pressure into dynamic and hydrostatic components

Done

56 time step procedure

Done

58 in a physical

Done

61 results show that

Done

62 fluctuations grow.

Done

79 Adapted from

Done

81 Region A is composed of three sub-regions: A1, A2..

Done

90 turbulence

Done

91 with less energy (delete on downstream)

Done

99 is the acceleration due to gravity (or the gravitational acceleration)

Done

119 It is 75.5 m in height

Done

123 Four discharges of 67,..

Done

124 real discharges of

Done

126 pool were 0, 15,..

Done

151 Piezometers were used to observe the fluctuations of the dynamic pressures on the..

Done

189 part of the jet is

Done

191 or has no significant reduction, while, from a certain depth, this..

Done

199 impingement of the core.. is a result of the plunge pool

Done

202 consequently so does.. As seen in Figure 4,..

Done

208 reduction in the

Done

228 along with increases in discharge.

Done

248 it decreases away from the

Done

251 At 30 cm depth (delete Reaching)

Done

254 decline in Cp

Done

258 the highest pressure

Done

260 This may be related to the

Done

280 relative to the horizontal increases

Done

281 decreases in all depth situations

Done

283 behind this is that

Done

289 and vortices are

Done

294 a lower dynamic

Done

306 The greater the plunging depth

Done

308 A low plunging depth did not give a reduction of

Done


Reviewer 3 Report

English language should be improved, especially the technical terms should be used in usual sense.

Dynamic pressure is normally given in Pascals, here is in meters. Obviously this is in the „head“ form, then the corresponding physical quantity is the “velocity head” and not “dynamic pressure”.

The physical dimensions are in form “184 ls-1” (line 21), should be “184 l/s” or “184 l.s-1”. Similar in lines 123, 124,…

Line 110, eq. (4) is not the “break-up length”, but the pressure coefficient, the same as (3)!

Line 142, in figure 2 the quantities should be shown clearly, e.g. Bj and points of pressure measurements. The experiment is described not enough in my opinion, the volumetric rate is not the right input parameter for the jet, as velocity is the governing parameter for dynamic pressure.

Generally, all used geometrical quantities should be defined and shown in figures. List (index) of quantities could help.

Line 140: the “side” wall is in reality the “bottom” wall?

Line 150-163 “measurement of dynamic pressure”, the fluctuations of dynamic pressure are not measured properly in my opinion, transducers, tubing, frequency response, etc. are not specified.

In Figure 5 the comparison is not clear, as the presented experiments are in very different regimes than those from literature.

In Figure 7, in graphs the axes are not described.

The paper should be reworked completely, better description, more details should be given and clear figures.


Author Response

Response to Reviewer 3 Comments

Thanks for your observation and comments. We did our best to incorporate all your valued comments. In the review process and based on your comment, we believe the paper has been significantly improved.

 

English language should be improved; especially the technical terms should be used in usual sense.

Thanks for your comment. We agree with you. In the revised manuscript, we have improved the language and paper presentation, as suggested. Hope our attempt satisfy respected referee. 


Dynamic pressure is normally given in Pascal, here is in meters. Obviously this is in the „head“ form, then the corresponding physical quantity is the “velocity head” and not “dynamic pressure”.

We agree with the referee which the dynamic pressure (P) is in Pascal (Pa). However, in the Bernoulli equation, the corresponding terms express as P/γ, which is in meters as is for “velocity head” which mentioned by the referee. In this research, the dynamical pressure coefficient (CP) was used to represent the dynamic pressure which, according to equation 3, is a dimensionless parameter. Thanks for your accurate comments.

The physical dimensions are in form “184 ls-1” (line 21), should be “184 l/s” or “184 l.s-1”. Similar in lines 123, 124,…
We are sorry for this typo error. The dimension has been modified as l/s in the entire text.


Line 110, eq. (4) is not the “break-up length”, but the pressure coefficient, the same as (3)!

Thank you. We have modified this point in the revised manuscript as presented in lines 109 and 110.


Line 142, in figure 2 the quantities should be shown clearly, e.g. Bj and points of pressure measurements.

We modified Fig. 2 in response to your valued comment. The parameter Bj and the location of pressure measurements by piezometers were schematically added in Fig. 2. Thanks for your comment.

 

The experiment is described not enough in my opinion, the volumetric rate is not the right input parameter for the jet, as velocity is the governing parameter for dynamic pressure.

We agree that the velocity of the jet is an effective and dominant parameter in dynamic pressure, as presented in Equation 3. Moreover, it is well-known that the velocity is descriptive of discharge while the area is a constant value as mentioned in the continuity equation as Q = AUj, where A is flow area (Jet area herein). However, discharge is also an important parameter in the amount of dynamic pressure, as reported in the literature. However, as a finding of this research, according to Fig. 4, at 0°, the maximum pressure coefficient for a discharge of 67 l/s was about 0.35, and for a discharge of 186 l/s, it was about 0.85 which are significantly different.


Generally, all used geometrical quantities should be defined and shown in figures. List (index) of quantities could help.

We thank you for this comment. We have included the variables in Fig. 2a. Regarding the list of variables, we agree that the abbreviation list could help the readers. Since it is not recommended by the journal, it is not incorporated in the manuscript.


Line 140: the “side” wall is in reality the “bottom” wall?

Yes, we agree with you. In the revised manuscript, to overcome this and additional issues raised by other referees, the “side wall” was replaced with “impact plate” throughout the text. .

Line 150-163 “measurement of dynamic pressure”, the fluctuations of dynamic pressure are not measured properly in my opinion, transducers, tubing, frequency response, etc. are not specified.

We agree with you. In the revised manuscript, we have improved the related statements. Kindly see lines152, 157-158.

  

In Figure 5 the comparison is not clear, as the presented experiments are in very different regimes than those from literature.

We have considered and incorporated this point in the revised manuscript. The graphs are presented in case of H/Lb.  Kindly see Fig. 5 which is now consistent with those reported from previous works. Thanks for your valued comments.


In Figure 7, in graphs the axes are not described.

We understand your concern. Therefore, we highlighted the plate dimension in the text as 0.5×0.5 m. The original file is with more resolution which provided enclosed this manuscript.

The paper should be reworked completely, better description, more details should be given and clear figures.

We have incorporated all your comments and other reviewers’ to alter the quality of our works. We believe that your comments improved our manuscript and it is now suitable for publication in the journal. Thanks!


Round 2

Reviewer 1 Report

The authors addressed all my questions and concerns. Thanks a lot! With some of them, I am, however, not yet satisfied requiring another review round. Please improve Figures 4-8.

(a) For figure 4, 5, 6 and 8, please remove the greenish background colour.

(b) For figure 5, please use point & line together to show comparison between present research and previous works. And please explicitly state what is novel here.

(c) Please use rainbow colour scale for figure-7.

(d) Please use point & line together for figure-8.

Author Response

Response to Reviewers

The authors addressed all my questions and concerns. Thanks a lot! With some of them, I am, however, not yet satisfied requiring another review round. Please improve Figures 4-8.

Thanks for your valued comments which improved the contents and presentation of the manuscript.

(a) For figure 4, 5, 6 and 8, please remove the greenish background colour.

We checked the original files and removed any fills and colors as a background for the figures. Kindly see figures 4, 5, 6, and 8. Thank you.

(b) For figure 5, please use point & line together to show comparison between present research and previous works. And please explicitly state what is novel here.

Thanks for your accurate comment. As requested, we used lines and markers for graphs in Figure 5 to be consistent with the charts from other researches. Kindly see the revised version of Figure 5.

As stated in lines 205 -206, the diminishing of the pressure coefficient in this study starts from the range of 2 <Y/Bj <4. In lines 219- 224, the results obtained from Fig. 5 are described and justified. 

(c) Please use rainbow colour scale for figure-7.

Done. Kindly see figure 7.

(d) Please use point & line together for figure-8.

Done. Kindly see figure 8.


Reviewer 3 Report

The paper is improved, however still important information is missing.

There is no info about precision of measurements (measuring error) – this is an obligation for any experimental study.

The results in Figure 7 are not described properly. Still, the axes are not assigned (which are streamwise and spanwise directions?). The results are strange, as the distribution should be symmetrical with respect to the flow axis and it is obviously not. What are the small peaks in positions 30,10 and 30,40? Please make a comment.


Author Response

Comments and Suggestions for Authors

The paper is improved; however still important information is missing.

The authors thank the referee for his/her valued comments.

There is no info about precision of measurements (measuring error) – this is an obligation for any experimental study.

The accuracy of the measurements is specified in lines 159-160. Thanks for your accurate comment.

 The results in Figure 7 are not described properly. Still, the axes are not assigned (which are streamwise and spanwise directions?). The results are strange, as the distribution should be symmetrical with respect to the flow axis and it is obviously not. What are the small peaks in positions 30,10 and 30,40? Please make a comment.

As the referee correctly stated, the direction of flow was missed in the figure 7. We added the flow direction to recognize the streamwise and spanwise directions of the flow. Moreover, in response to reviewer 1,   figure 7 is given in color. Kindly see figure 7.

In case of asymmetry in pressure on the impact plate, depending on the amount of flow and flow rate, the jet position on the impact plate is varying, and the jet may not hit the center of the screen. As stated in manuscript (Lines 196- 199; and lines 260-266) developed jets create both high and low frequency vortices. The formation of such turbulent currents may not cause the maximum pressure to occur in the enclosure. Therefore, the distribution of pressure on the impact plate may not be symmetrical. Also, the formation of these rotational currents may result in the formation of small peaks in different positions on the screen. Few references to support such findings have been cited in the manuscript. Thanks a lot for your efforts to clarify the presentation of the paper.


Back to TopTop