Characterization of the Solid Particle Erosion of the Sealing Surface Materials of a Ball Valve
Round 1
Reviewer 1 Report
In this paper, the authors characterized the erosion behavior of different materials used in ball valves via dry direct impact tests, comparing the experimental data with simulations results and the prediction of a not well specified theoretical model. Strenghts of this work are that the experimental campaign was very extensive in terms of testing conditions, and that the paper is relatively well structured and written.
At the same time, a fundamental issue left me perplexed. I think that, before going on with a more detailed, technical review, the authors should provide some definitive answer about the scope and the novelty of this manuscript, since, at the moment, I do not see any advancement over the state of the art.
Firstly, the title is deceptive, since it leaves the reader the impression that erosion data on ball valves will be presented, whereas only dry direct impact test are discussed. If this study was indended to have some technological-practical scope, it should include, at least, some numerical and experimental wear results on the actual ball valve geometry, providing evidence that the dry direct impact tests provide meaningful information for the device.
Conversely, if the study was aimed at characterizing the erosion behavior of materials used in ball valves only through dry direct impact tests, and, therefore, its scope was mostly “scientifical” rather then technological, the authors should definitely clarify the novelty of their study. There is plenty of literature concerning similar results on metals and other materials, obtained both numerically and experimentally. Honestly, the only added value I can see compared to other papers is that the experimental campaign was quite comprehensive. However, adding new data without a real advancement in terms of knowledge does not make a paper worthy of publication in a scientific journal.
Actually, I have many other technical comments. However, I think that going through these aspects would be a waste of time unless the authors prove that their paper provides significant scientific and/or technological advancement over the state-of-the-art. Up to that moment, this paper should not be considered further for publication.
Author Response
Point 1：The authors should provide some definitive answer about the scope and the novelty of this manuscript
Response 1: Your advice is very helpful to us. We think that our innovation lies in the variety of experimental conditions. We considered different erosion angles, different velocities and particle sizes. In the research on the erosion process of ball valve sealing surface, we have carried out comparative test and orthogonal test on high-speed impact of particles, different particle sizes and different erosion angles. In the next step, we will continue to improve the quantitative influence mechanism of gas-solid erosion, and verify it. At the same time, establish some life prediction methods of ball valve, and finally strive to form a complete set of ball valve damage evaluation system. I hope our work can be completed, we will continue to enrich our research content, and more innovative work, thank you.
Point 2: The title is deceptive, since it leaves the reader the impression that erosion data on ball valves will be presented, whereas only dry direct impact test are discussed.
Response 2: Thank you for pointing out the mistake. We have decided to change the title to “Erosion of the Ball Valve in Natural Gas Pipeline”. It's simple and clear. Through the experiment of simulating ball valve with plate specimen, and combined with the model to verify, we revealed some mechanism of ball valve erosion. We will continue to carry out the related work in the future, thank you.
Point 3: If this study was indended to have some technological-practical scope, it should include, at least, some numerical and experimental wear results on the actual ball valve geometry, providing evidence that the dry direct impact tests provide meaningful information for the device.
Response 3: Thank you for your suggestion. We have revised it according to your suggestion. At the same time, the fourth chapter of the manuscript focuses on the verification and comparison of experimental and simulation results. We are sorry that we did not give too much data, because the manuscript would be too long. For spherical ball valve, we think that every particle erosion to every point on the surface of the specimen is a tiny plane, so we assume that every specific contact point on the surface of the specimen is composed of several planes, a geometric diagram and some parts of the experimental data are given below. thank you. All the figs. and tables are given in the attachment.
Fig.1 Geometric diagram of ball valve erosion
Table 1 Q235 of experimental groups at different air flow rates
Table 2 Experimental terminal mass loss at different air velocity experiments
Thank you for all your suggestions. We have made major revision to the manuscript according to the requirements of editors and reviewers. At the same time, the authors once again discussed some defects and shortcomings of this work, we will make improvements in the next work, we believe that the follow-up research will do better. Thank you
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Please see the attached file for suggestions to improve the manuscript.
Comments for author File: Comments.pdf
Author Response
Point 1: The results obtained are not grid independent yet. You should run another mesh with 0.5 mm grid size.( On line 251 of the second reviewer's attached file)
Response 1: Thank you for your advice. We will continue to improve. We will run a simulation with a grid size of 0.5mm in the future.
Point 2 : Please add the definition of this dimensionless number. And add also the amount it reached. ( On line 267 of the second reviewer's attached file)
Response 2: Thank you for your advice. According to the meaning of Stokes number, Stokes number is the ratio of momentum response time and fluid characteristic time of particles, which represents the relative size of inertia force and drag force of particles, and is an important dimensionless number group to characterize the motion characteristics of particles. We believe that when Stokes Number St>1, the particles undergo denaturation and jump, and the erosion of the sealing surface of the ball valve presents discontinuous point erosion. The maximum erosion rate increases slightly with the increase of St, and the increase is small.
Point 3：You should add a sentence here to explain that you will compare the simulation results to the experimental ones of Vieira. ( On line 286 of the second reviewer's attached file)
Response 3: Thank you for your suggestion. The main difference between our work and Vieira's is the different research objects. They focus on the elbow of the pipeline, and we mainly consider the ball valve. Of course, the regularity of erosion results is consistent. Under the condition of different particle size and different velocity, the results are basically the same, but at different erosion angles, Vieira et al. Think that the maximum erosion angle is 45 ° and we think it is 30 °. In the next step, we will continue to carry out more in-depth research. Thank you.
We have made some changes to the format and sentences you proposed in the manuscript. Thank you.
Author Response File: Author Response.pdf
Reviewer 3 Report
Review of the manuscript entitled: Simulation and Experimental Analysis of the Erosion of the Ball Valve Sealing Surface in Gas-solid Two-phase Flow
The reviewed manuscript is quite interesting. It indirectly concerns the problem of erosion of the executive element regulating the flow in the valve. However, the manuscript contains errors of a methodological nature. It also requires explanations and expansion regarding the elementary issues contained in the work. Below are the main shortcomings and elements that require clarification and/or extension.
The cardinal point:
The proposed case is not a constructional element of the Ball valve. The ball valve has a ball-shaped executive element with a hollow cylindrical bore. As a result, the plane on which the impacted stream has a different shape (cylindrical - as a part of the cylinder wall), and thus the erosion trace (and flow trajectory, flow reflection stream) will be completely different. Up to this position of the half-open (or half-close) valve, they change the hydraulic diameter and the direction of the air stream. These facts make the topic and background of the simulation analysis inadequate. Rather, the simulated element resembles a Butterfly valve executive element, with its axis of rotation typically being vertical rather than horizontal as proposed by the authors. However, such valves are not used to regulate the flow of air, but liquids, e.g. in the food industry. This point alone may constitute a rejection of the manuscript. I suggest you search the catalogs of valve manufacturers and the Internet, where you can easily find videos with CFD analyzes of the flow through the Ball valve.
Main lacks and elements that need to be completed or clarified:
1. A mathematical model of the simulated phenomenon should be presented, taking into account the occurring of the dispersed phase and the interaction between the phases.
2. The boundary conditions of the mathematical model should be presented.
3. The assumptions and simplifications of the model should be clearly presented.
4. Provide data on the mesh, a number of elements, mesh quality (e.g. y +), the sensitivity of the results to the number of mesh elements.
5. Provide data on the type of simulation (steady-state or transient), convergence criteria and their testing, etc. Due to its nature, the case should be modeled as a transient. There is no information about this.
5. Simulation of the Erosion phenomenon is a typical FSI case. How are the "dead nodes" of mesh for a solid (valve executive element) declared in the simulation? Please provide an example image of the surface texture as a result of CFD.
6. What do the colors in Fig. 11 mean? Please provide the scale and unit.
7. Compare the obtained results with the experimental measurements. For example, when presenting the erosion on the surface as a simulation result, the results of surface texture (surface microtopography) obtained experimentally should be presented and compared.
Summary: The main text requires some detail. It concerns a phenomenon difficult to model. The very application of the Oka erosion model in the form of an implemented UDF requires clarification and extension in terms of the CFD model and the case of flow and interaction with the eroding surface. I recommend a major revision because I believe there is a chance to correct the manuscript.
Author Response
Point 1: A mathematical model of the simulated phenomenon should be presented, taking into account the occurring of the dispersed phase and the interaction between the phases.
Response 1:
Thank you for your advice. The Eulerian–Lagrangian method was used in this study to treat gas as a continuous phase, while the particles were treated as a discrete phase.
(All the specific theoretical numerical formulas are attached below)
Gas liquid phase model
Fluid follows the mass conservation law, momentum conservation law and energy conservation law during the process of flow. The continuous phase is solved with the Navier-Stokes equations, and the continuity and momentum equations can be expressed as:
Where, is the velocity vector in three different directions; ρg is the gas density; p is the pressure; ρggy is the gravitational force; SM is the added momentum due to the solid phase; is the stress tensor; μ is the viscosity; I is the unit tensor.
Displaying high accuracy and reliability, the Shear Stress Transport(SST) k-ω model, based on k-ε and k-ω turbulence mode, can better simulate the turbulent flow in near-wall boundary layer[20], the transport equations for the SST model that we used in this work can be expressed as：
where, Gk is the generation of turbulence kinetic energy due to the mean velocity; Gω is the generation of ω; Γk and Γω are the effective diffusivity of k and ω respectively; Yk and Yω are the dissipation of k and ω due to turbulence respectively; Dω represents the cross diffusion term; Sk and Sω are source terms.
Sand discrete phase model (DPM)
The motion of discrete phase particles in a fluid is determined by Newton's second law, the motion equation of the particles under the Lagrangian coordinates can be expressed as:
where, vp is particle velocity; vg is gas velocity；FD is the drag force per unit particle mass; ρp is the particle density; dp is the particle diameter; Fother donates other forces acting on the particles; Re is the particle Reynolds number; CD is the drag coefficient;α1、α2、α3are constants for the smooth spherical particles given by Morsi and Alexander[21].
Point 2: The boundary conditions of the mathematical model should be presented.
Response 2: Thank you for your advice, we have made corresponding adjustments.The boundary conditions of the mathematical model has been described in Section 3.1 and Figure 6. The grid independence test has been described in Section 3.2.
Point 3: The assumptions and simplifications of the model should be clearly presented.
Response 3: Thank you for point out our mistakes. In fact, our assumptions include the following aspects. The sand output from the nozzle is uniform, the air flow is stable, and the plane of the specimen is ideal. At the same time, we have also made a modification, pointing out that in order to not consider the influence of sand movement on fluid flow, we use the unidirectional coupling method, obtain the flow field movement, particle impact velocity and angle through CFD and particle tracking, and then use Oka erosion prediction model to get the erosion situation of the specimen.
Point 4: Provide data on the mesh, a number of elements, mesh quality (e.g. y +), the sensitivity of the results to the number of mesh elements.
Response 4: Thank you for pointing out our negligence. We have made amendments to explain numerical model and boundary condition settings and grid independence test in Sections 3.1 and 3.2. Also added to propose to avoid the influence of the number of particles on the final erosion result, we refines the nozzle entrance grid size to increase the number of particles incident from the nozzle, and sets the entrance grid size to 0.06mm, making the number of entrance grids reach 12520.
Point 5：Provide data on the type of simulation (steady-state or transient), convergence criteria and their testing, etc. Due to its nature, the case should be modeled as a transient. There is no information about this.
Response 5: Thank you for your advice. In the ideals and assumptions of the model, we assume that the specimen is in a stable state during the erosion process, for example, the airflow is a steady flow, so we use a single coupling method to carry out this work.
Point 6: Simulation of the Erosion phenomenon is a typical FSI case. How are the "dead nodes" of mesh for a solid (valve executive element) declared in the simulation? Please provide an example image of the surface texture as a result of CFD.
Response 6: Thank you for reminding. In the CFD simulation, because the eroded area is affected by high pressure, some material spalling and peeling may occur on the surface of the specimen.
Point 7: What do the colors in Fig. 11 mean? Please provide the scale and unit.
Response 7: Thank you for reminding. We have added the unit labels of Figure 10 and Figure 11 in the manuscript. Thank you.
Point 8: Compare the obtained results with the experimental measurements. For example, when presenting the erosion on the surface as a simulation result, the results of surface texture (surface microtopography) obtained experimentally should be presented and compared.
Response 8: Thank you for your advice. In the experiment, the concentrated stress of the specimen is not intuitive, we can not directly see the change of the surface of the specimen through the macro observation, and we can not see the spalling of the surface, so we use SEM to observe the microstructure of the specimen. In this way, a certain contrast is formed between the experimental microscopic observation and the CFD simulation surface.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
My comments are provided in the attached report.
Comments for author File: Comments.pdf
Author Response
Thanks to the reviewer, our reply is in the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Re-review of the manuscript entitled: Erosion of the Ball Valve in Natural Gas Pipeline, formerly Simulation and Experimental Analysis of the Erosion of the Ball Valve Sealing Surface in Gas-solid Two-phase Flow
As well I can see the manuscript is corrected. You prepared adequate corrections and improve the manuscript. It should be more readable and clear for potential readers. The paper concern with the current FSI simulation aspects so should be citable also.
In the view of responses and prepared corrections, I recommend accepting the paper in the present form. However the English corrections still necessary.
Congratulations.
Author Response
Thank you for your suggestion. We have made some sentence changes in the manuscript, such as the chapter of grid independence test, etc.
Round 3
Reviewer 1 Report
I acknowledge the effort put in by the authors in responding to my comments. I am satisfied with the revisions made and I am pleased to recommend acceptance of the revised manuscript for publications in the "Metals" journal.
Just a suggestion. Personally, I would explain in the manuscript that the “theoretical model” is the Oka erosion model at the bulk scale, that is, based on the jet bulk-mean velocity and the nozzle-to-specimen angle. In fact, some misunderstanding might happen due to the fact that, in the CFD simulations, the same Oka erosion model was applied at the local scale of particle-wall impingements, after coupling with the DPM tracking. I would also stress that the good agreement between the “theoretical” and “CFD” results further confirms that the motion of the particles in the gas-solid jet is dominated by inertia, which is not surprising given the high velocity of the jet and the high density ratio.
Author Response
Thank you for your advice and pointing out our shortcomings. In the manuscript, our general idea is to verify the applicability of Oka erosion model. We have carried out multi factor experiments and simulation verification. After the completion of this manuscript, we have also done comparative verification of other influencing factors. We think that this can enrich the connotation of Oka erosion model, and lay a solid foundation for proposing other modified models in the future. To better understand this manuscript, we decided to change the theoretical model in Section 4 to erosion model. Thank you for your guidance.
