Numerical Simulation of Flame Propagation in a 1 kN GCH4/GO2 Pintle Injector Rocket Engine
Round 1
Reviewer 1 Report (New Reviewer)
Comments and Suggestions for AuthorsThis paper focuses on the ignition and flame propagation of a methane-oxygen rocket engine, using a 1 kN thrust pintle-type injector as the research subject. Advanced numerical simulation methods, including the Eddy Dissipation Concept (EDC), the Partially Stirred Reactor (PaSR) model, and Shielded Detached Eddy Simulation (SDES), were employed to analyze the transient ignition process. The study examined flame propagation and stability, uneven mixing regions, pollutant formation, and thermal load distribution on the injector. The results provide critical data for optimizing the geometry of the injector, reducing the reliance on iterative prototyping in the early design stages.However, there are several areas in the paper that need improvement.
1. The description of experimental parameters and photos should not be placed in the introduction but should be moved to a section like "Materials and Methods." The literature review section should be placed in the introduction.
2. In Line 74, the phrase "The paper written by V.M. Zubanov et al [5], presents a method for simulating transient combustion in a rocket engine that uses gaseous hydrogen and oxygen" is suggested to be rephrased as "V.M. Zubanov presents a method...". Similarly, the citation in Line 119, "the paper by A.V. Brito et al", should also be refined for clarity.
3. When EDC is mentioned in the main text, its full name should be provided. All abbreviations must have their full forms given when they first appear in the abstract and throughout the paper.
4. The main contributions of the paper should be clearly outlined and detailed in the introduction.
5. The reason for choosing GCH4/GO2 as the fuel is not fully explained. The advantages of these fuels should be thoroughly described in the paper.
6. The description of the key sub-models used in the simulation is too detailed. It is suggested to simplify this part.
7. There is a lack of comparison between the numerical simulation and experimental results. How can the accuracy of the simulation results be ensured?
8. In Figure 5, it can be observed that near the Y-coordinate = 0.025, Adapt1 and Adapt2 are closer to each other, while the 400,000 rectangular grid shows a discrepancy compared to the two cases. Therefore, I am curious whether further increasing the grid count would lead to significant changes in the results.
9. In the paper, it is mentioned that the backflow species consist of 76.6% N2 and 23.4% O2. Are these mole fractions or mass fractions?
10. The ignition location should be specifically marked in the figure. The scale for temperature and velocity should be unified for consistency.
11. The paper lacks a conclusion section.
Author Response
|
Comment 1: The description of experimental parameters and photos should not be placed in the introduction but should be moved to a section like "Materials and Methods." The literature review section should be placed in the introduction.
|
|
Response 1: The Literature review has been moved to the introduction section, while the description of the parameters has been moved to a new chapter.
|
|
Comment 2: In Line 74, the phrase "The paper written by V.M. Zubanov et al [5], presents a method for simulating transient combustion in a rocket engine that uses gaseous hydrogen and oxygen" is suggested to be rephrased as "V.M. Zubanov presents a method...". Similarly, the citation in Line 119, "the paper by A.V. Brito et al", should also be refined for clarity.
|
|
Response 2: Thank you for the suggestion. The citations have been refined in order to provide more clarity. Page 2, paragraph 1, line 56.
Comment 3: When EDC is mentioned in the main text, its full name should be provided. All abbreviations must have their full forms given when they first appear in the abstract and throughout the paper.
Response 3: The first usage of the EDC model in the main text has been rectified, using the full name. Page 2, paragraph 4, line 84.
Comment 4: The main contributions of the paper should be clearly outlined and detailed in the introduction.
Response 4: The main contributions have been added in the introduction section, on page 3, paragraph 4, lines 121-132.
Comment 5: The reason for choosing GCH4/GO2 as the fuel is not fully explained. The advantages of these fuels should be thoroughly described in the paper.
Response 5: The explanation has been introduced on page 3, line 59-70
Comment 6: The description of the key sub-models used in the simulation is too detailed. It is suggested to simplify this part.
Response 6: Thank you for noticing that. It has been a request from previous reviewers to provide more detail.
Comment 7: There is a lack of comparison between the numerical simulation and experimental results. How can the accuracy of the simulation results be ensured?
Response 7: While the experimental campaign has not yet been performed, the simulation uses methods already validated by other papers, such as:
Son, M., Acta Astronautica (2017), http://dx.doi.org/10.1016/j.actaastro.2017.02.005
Frassoldati, Alessio et al. “Simplified kinetic schemes for oxy-fuel combustion.” (2009).
De Giorgi, M.G.; Sciolti, A.; Ficarella, A. Application and Comparison of Different Combustion Models of High Pressure LOX/CH4 Jet Flames. Energies 2014, 7, 477-497. https://doi.org/10.3390/en7010477
Comment 8: In Figure 5, it can be observed that near the Y-coordinate = 0.025, Adapt1 and Adapt2 are closer to each other, while the 400,000 rectangular grid shows a discrepancy compared to the two cases. Therefore, I am curious whether further increasing the grid count would lead to significant changes in the results.
Response 8: Thank you for the comment. The increase in grid count would probably not lead to major changes in the simulation, as the error between the 4000000 mesh and the other adapts is of about +/- 0.05%.
Comment 9: In the paper, it is mentioned that the backflow species consist of 76.6% N2 and 23.4% O2. Are these mole fractions or mass fractions?
Response 9: Thank you for covering this. The backflows species and domain species are represented in mass fractions. The corrected phrases can be found on page 9, paragraph 6, line 341.
|
Comment 10: The ignition location should be specifically marked in the figure. The scale for temperature and velocity should be unified for consistency.
Response 10: Thank you for pointing this out. The location of the spark generator is specified in Figure 2 b). We have also included it in the caption, based on your recommendation. Page 4 Figure 2 b).
Comment 11: The paper lacks a conclusion section.
Response 11: Thank you for pointing this out. The reasoning behind not including a conclusion section was based on the manuscript template which indicates the following: “5. Conclusions
This section is not mandatory but can be added to the manuscript if the discussion is unusually long or complex.”
A conclusion section has been introduced on page 18.
Reviewer 2 Report (New Reviewer)
Comments and Suggestions for AuthorsThe work presented in the article focuses on igntion and flame propogation studies of methane fuelled pintle injector for a 1KN model rocket engine. The following major corrections are required to consider the article for publication.
1. The title of the article does not clearly depict the rocket engine type. Suggesting to change the title from "pintle rocket engine" to "pintle injector rocket engine" for more clarity.
2. Some of the literatures presented in the article are irrevalent to the study. Add literatures relevant to the ignition and flame propogation studies.
3. How did you choose the rocket engine design? Reference about the design is missing.
4. The wedge axisymmetric configuration of the rocket engine is taken for the simulation is mentioned in section 4. But, the images corresponding to the wedge configuration are not shown in the article. It looks like 2D axisymmetric configuration. Clarify?
5. The section 4 is not properly presented. Restructure the section 4 in a proper way for better understanding.
6. The details of the adapted mesh strategy used for the computational study is not given.
7. What are the important parameters compared among different meshes? These are not mentioned in grid independence study
8. The design chamber pressure is mentioned as 2 MPa in table 1. But the simulation is carried out only for atmospheric pressure (gauge pressure = 0 Pa). Why?
9. The picture resolution of figure 18 is poor. Mention the ignition events in the figure 18.
10. In the abstract of the article, the studies on flame stability of the rocket engine is mentioned. Does the velocity vector and recirculation zones are enough for the flame stability study.
Author Response
|
Comment 1: The title of the article does not clearly depict the rocket engine type. Suggesting to change the title from "pintle rocket engine" to "pintle injector rocket engine" for more clarity.
|
|
Response 1: Thank you for pointing this out. We have modified the title: “Numerical simulation of flame propagation in a 1kN GCH4/GO2 pintle injector rocket engine”
|
|
Comment 2: Some of the literatures presented in the article are irrevalent to the study. Add literatures relevant to the ignition and flame propogation studies.
|
|
Response 2: Thank you, more relevant literature has been introduced. Some include 1. : J. Liberatori, R. M. Galassi, D. Liuzzi, et al, Uncertainty quantification in RANS prediction of LOX crossflow injection in methane, AIAA Propulsion and Energy Forum, Virtual Event, August 9–11, 2021, 1–18. 2. S. Blanchard, Q Cazères, B. Cuenot, Chemical modelling for methane oxy-combustion in Liquid Rocket Engines, Acta Astronaut. 190, 2022, 98–111. 3. Peters, N. Laminar diffusion flamelet models in non-premixed turbulent combustion. United Kingdom: N. p., 1984. Web. doi:10.1016/0360-1285(84)90114-X. 4. Peters, Norbert. “Laminar flamelet concepts in turbulent combustion.” (1988).
Comment 3: How did you choose the rocket engine design? Reference about the design is missing.
Response 3: The method used for the engine design has been introduced in text on page 5, lines 171-176.
Comment 4: The wedge axisymmetric configuration of the rocket engine is taken for the simulation is mentioned in section 4. But, the images corresponding to the wedge configuration are not shown in the article. It looks like 2D axisymmetric configuration. Clarify?
Response 4: Thank you for your comment. The images presented are a cross-section plane for a better visualization. In Figure 2, page 4, the studied wedge section can be seen.
Comment 5: The section 4 is not properly presented. Restructure the section 4 in a proper way for better understanding.
Response 5: Thank you for the feedback, the entire section has been reorganized.
Comment 6: The details of the adapted mesh strategy used for the computational study is not given.
Response 6:.Thank you for pointing this out, the mesh strategy has been written on page 10, paragraph 2 line 359-369
Comment 7: What are the important parameters compared among different meshes? These are not mentioned in grid independence study
Response 7: The parameters are discussed in the revised mesh strategy paragraph, page 10, lines 359-369.
Comment 8: The design chamber pressure is mentioned as 2 MPa in table 1. But the simulation is carried out only for atmospheric pressure (gauge pressure = 0 Pa). Why?
Response 8: Thank you for pointing this out! The operating pressure of the domain has been set at atmospheric conditions, making the gauge pressures of the outlet equal to 0 Pa, with respect to the atmospheric pressure. The design chamber pressure refers to the pressure inside the rocket engine combustion chamber at which it operates during steady state conditions. The text has been revised and can be found on page 9, paragraph 5, lines 338-341 .
Comment 9: The picture resolution of figure 18 is poor. Mention the ignition events in the figure 18.
Response 9: The picture has been improved and the ignition time has been introduced
Comment 10: In the abstract of the article, the studies on flame stability of the rocket engine is mentioned. Does the velocity vector and recirculation zones are enough for the flame stability study.
Response 10: The recirculation zones and velocity vector fields prove the central element regarding combustion stability, as they help visualize the flame anchoring regions. Other important elements regarding the flame stability can also be found in the result section, (5. Results page 11, second paragraph, lines 383-514), and conclusion on page 18:
|
Reviewer 3 Report (New Reviewer)
Comments and Suggestions for AuthorsThis work investigates the ignition and flame propagation processes of a 1kN gaseous methane-oxygen rocket engine using a swirl injector nozzle. The study employed advanced numerical simulation methods, combining the Eddy Dissipation Concept (EDC) and Partially Stirred Reactor (PaSR) models with the Shielded Detached Eddy Simulation (SDES) model, to simulate the complex transient ignition process. The results indicate that this method can provide important information about flame propagation and stability, pollutant formation, and temperature distribution during engine startup, highlighting the regions of mixing non-uniformity and the thermal load on the nozzle. This information can be further utilized for geometric optimization of the swirl injector nozzle, addressing key design challenges without the need for iterative prototype designs in the early stages of development.
1. Figures 5, 18, 19, and 20: To ensure the images are clean and easier to read, please remove the background grid, increase the font size of the axis titles, and enlarge the font size of the figure captions.
2. Was the reliability of the computational method validated?
3. This work has yielded many valuable results. To facilitate reader comprehension, please present the research findings in a table in the end of Section 5. Results.
4. Why is there no Conclusion section?
Author Response
|
Comment 1: Figures 5, 18, 19, and 20: To ensure the images are clean and easier to read, please remove the background grid, increase the font size of the axis titles, and enlarge the font size of the figure captions.
|
|
Response 1: Thank you for the feedback, the images have been improved.
|
|
Comment 2: Was the reliability of the computational method validated?
|
|
Response 2: While the experimental campaign has not yet been performed, the simulation uses methods already validated by other papers, such as:
Son, M., Acta Astronautica (2017), http://dx.doi.org/10.1016/j.actaastro.2017.02.005
Frassoldati, Alessio et al. “Simplified kinetic schemes for oxy-fuel combustion.” (2009).
De Giorgi, M.G.; Sciolti, A.; Ficarella, A. Application and Comparison of Different Combustion Models of High Pressure LOX/CH4 Jet Flames. Energies 2014, 7, 477-497. https://doi.org/10.3390/en7010477
Comment 3: This work has yielded many valuable results. To facilitate reader comprehension, please present the research findings in a table in the end of Section 5. Results.
Response 3: An enumeration of the most significant findings has been introduced in a the conclusion section on page 18.
Comment 4: Why is there no Conclusion section?
Response 4: Response 11: Thank you for pointing this out. The reasoning behind not including a conclusion section was based on the manuscript template which indicates the following: “5. Conclusions This section is not mandatory but can be added to the manuscript if the discussion is unusually long or complex.” A conclusion section has been introduced on page 18. |
Round 2
Reviewer 1 Report (New Reviewer)
Comments and Suggestions for AuthorsThe revised version improves the quality of the manuscript and is recommended for acceptance.
Author Response
Nothing to reply. The reviewer agreed with present form of the ms.
Reviewer 2 Report (New Reviewer)
Comments and Suggestions for Authors1.Cite the operating conditions and engine parameters (table 1) in section 2
2.How did you select the RPA dimensions? A reference to the parameters of operation.
3.Mention the wedge axisymmetric cut section angle.
4.How is the transient adaptive time step technique calculated?
5. Figure 5 needs a quantitative explanation, and the legend name needs to be changed.
6. A rough estimate of flame propagation is provided at the end. It is also stated that uniform flame propagation is the means by which combustion stability is attained.
7.Clarity on conclusion is required.
Comments on the Quality of English LanguageSentence formation and grammatical errors needs to be corrected.
Author Response
|
Comment 1: Cite the operating conditions and engine parameters (table 1) in section 2
|
|
Response 1: Both the engine parameters and the operating conditions were chosen by us in the engine model development stage (lines 141 to 143) to achieve 1 kN of thrust at sea level. The table has been referenced on page 5, paragraph 1, line 174.
|
|
Comment 2: How did you select the RPA dimensions? A reference to the parameters of operation.
|
|
Response 2: The dimensions were based on the parameters from table 1 in section 2, as described in: “The engine geometry has been determined using the Rocket Propulsion Analysis (RPA) software [28] by introducing operating parameters including the chamber pressure, total mass flow rate, mixture ratio and operating atmospheric conditions. With these inputs, the software computed the chamber’s initial dimensions and the exhaust nozzle. The exhaust nozzle has been computed by using the method of characteristics.”, page 5, paragraph 1, lines 171-176.
Comment 3: Mention the wedge axisymmetric cut section angle.
Response 3: The wedge angle has been mentioned on page 9, paragraph 4, line 331 and on page 10, Figure 3 caption, as requested.
Comment 4: How is the transient adaptive time step technique calculated?
Response 4: The adaptive time step method has been introduced on page 11, paragraph 1, lines 380-388. In this case, the time step size is proportional to the characteristic time taken for the fluid to cross a control volume and the given Courant number. For convection dominated and/or wave propagation problems, it is recommended to set CFL close to 1 to obtain more accurate results. Usually, the characteristic time scale is taken to be the minimum over all control volumes time scales. The adaptive strategy starts with an initial time step which, during the computation, can be changed based on multiplication with some step size factors ranging between 0.75 and 2 to ensure the time step falls between a minimum and a maximum prescribed value and the CFL number remains close to the value set by the user. |
Comment 5: Figure 5 needs a quantitative explanation, and the legend name needs to be changed.
Response 5: A table has been added to present the quantitative assessment of the goodness for the three grids. Legends for figure 5 have been updated.
Comment 6: A rough estimate of flame propagation is provided at the end. It is also stated that uniform flame propagation is the means by which combustion stability is attained.
Response 6: Thank you for pointing this out, the phrase has been revised. Page 19, paragraph 4, lines 584-586.
Comment 7: Clarity on conclusion is required
Response 7: The conclusions have been updated.
Reviewer 3 Report (New Reviewer)
Comments and Suggestions for AuthorsAccept in present form
Author Response
Nothing to reply.
The reviewer agreed with the present form of the ms.
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
Comments and Suggestions for AuthorsThe present study is concerned with discovering insights into the ignition process and flame front propagation at engine start-up.
It is one of the research directions on the engine injector design.
In my opinion, the paper could be published with some revisions.
The revisions regard the following points:
1. Grammar should be double-checked.
2. The motivation mentioned in the first paragraph is not well connected with the research content of this manuscript. Modifying the relevant content is recommended.
3. The description in Figure 2 on page 2 does not clearly explain the location of the spark generator. It is recommended to add relevant content.
4. In the boundary condition setting of the simulation, the fuel inlet temperature of 300 K is different from the fuel inject temperature of 450K in Table 1 on the second page. It needs to explain why there is a difference.
5. The resolution of Figure 5 on page 9 is too low, it needs to be corrected
6. In the results section on page 10, the content needs to clarify which figure number the content refers to.
7. The last section of the manuscript needs to summarize the conclusion, not the discussion.
8. The best way to convey clout is to write the paper from a third-person point of view. Modifying the relevant content is recommended.
Reviewer 2 Report
Comments and Suggestions for AuthorsThe manuscript presents a numerical simulation of the ignition process in a scale model of a 1kN GCH4/GO2 rocket engine equipped with a pintle injector. The numerical setup relies on an axisymmetric configuration and leverages a Shielded Detached Eddy Simulation (SDES) approach. Turbulent combustion is closed by combining the Eddy Dissipation Concept (EDC) and the Partially Stirred Reactor (PaSR) models.
Due to the extremely high computational cost, the numerical simulation was carried out up to 6 ms, representing the main features of the ignition process and corresponding flame propagation. Nonetheless, as the authors explicitly state in Section 6, a steady state was not reached, and the numerical simulation should run longer, e.g., to provide reliable mean fields about the quantities of interest.
Moreover, the authors encountered numerical issues about the spark ignition process, with a significant overestimation of the initial temperature rise, which eventually led to the limitation of the maximum temperature available in the computational domain.
Nonetheless, this analysis's primary outcome is unclear, although the authors state that the present research study constitutes the background for a design optimization process. In contrast, peculiar attention is placed on pollutant formation, which should be deemed a secondary concern in model-scale rocket engine design optimization.
Moreover, while the authors repeatedly complain about the exceptionally high computational burden, they do not consider employing a skeletal yet accurate chemical kinetic mechanism. In their concluding remarks, they suggest adopting a large eddy simulation (LES) approach and introducing a Flamelet Generated Manifold (FGM) methodology, which, in principle, is less suitable than a finite-rate chemistry model to address complex ignition processes typical of pintle injectors.
Based on these considerations, the manuscript is not recommended for publication in Processes, even after a significant revision. The following highlights the most relevant remarks.
MAJOR REMARKS
1. First, the Introduction section needs to provide more context about the relevance of pintle injectors. It merely describes the experimental setup that inspired the numerical simulations carried out in the present work.
2. In Section 3.2 (NOTE: there’s a typo since two paper sections are named Section 3.2), on page 5, the authors discuss the partially premixed combustion modes established in rocket engines equipped with pintle injectors. Still, they do not precisely describe the flow field regions typical of a pintle injector configuration where non-premixed and premixed combustion modes are more prone to occur. Moreover, the characteristics of the resulting flame from pintle injectors are compared to those of “lean premixed combustors with diffusion pilot flames, etc.”, which is unclear to me.
3. In Section 3.2, on page 5, the authors introduce the mass and thermal diffusion coefficients, which are crucial in driving mixing and ignition in the test case being addressed. Still, more information about the turbulent Schmidt and Prandtl numbers needs to be provided. Are those values assigned according to the default standard in Ansys Fluent?
4. In Section 3.2, on page 6, relevant concerns arise about selecting the GRI-Mech 3.0 as the reference chemical kinetic mechanism. In the first place, as 53 chemical species are included, the associated computational burden is exceptionally high - as further highlighted on page 8 - and represents the fundamental reason why the authors could not carry out the numerical simulation until the attainment of a statistical steady state. Moreover, while the GRI-Mech 3.0 includes NO formation and reburn chemistry, this aspect is irrelevant to the design optimization of a model-scale rocket engine. In addition, the GRI-Mech 3.0 was developed to model natural gas and methane combustion in air [Smith1999] at moderate pressure, with significant discrepancies arising about ignition delay time estimation under intermediate-to-high pressure levels mainly due to the absence of the high-pressure alternative branching pathway involving CH3O2. In this regard, several alternatives were proposed during the last decade to address methane-oxygen combustion under high-pressure conditions typical of rocket engines, including several skeletal kinetic schemes able to trade off for predictive accuracy and computational cost.
5. In Section 3.2, on page 6, the authors discuss using the Peng-Robinson equation of state. Still, besides stating that pressure is expected to increase within the combustion chamber during the ignition process, the discrepancy between ideal and non-ideal equations of state in estimating relevant gas properties has yet to be thoroughly investigated. Indeed, by way of illustration, it should be interesting to provide insights into how much the Peng-Robinson model deviates from the ideal gas assumption at 20 bar.
6. In Section 3.3, on page 7, the sentence “The spark model works by burning a few cells near the defined location” does not contain any technicalities and should be considered for re-writing.
7. In Section 4, on page 8, the authors describe the computational domain as a 2D cross-section. This is still a 3D numerical simulation and should be correctly regarded as a “wedge axisymmetric configuration.” Moreover, Figure 3 does not feature any caption.
8. In Section 5, on page 12, the authors state that the highest rate of carbon monoxide generation mainly occurs near the mixture spraying cone, which is associated with imperfect mixing. Yet, this might be related to chemical kinetics, with CO-to-CO2 conversion pathways becoming relevant downstream.
9. On page 13 of Section 5, peculiar attention is paid to NOx formation. Is this aspect relevant in a design optimization context? Instead, greater attention should be paid to ignition and flame propagation characteristics, which are crucial in preventing a pintle tip exposure temperature that is too high.
10. In Section 5, Figures 15 and 16, which discuss the turbulent Reynolds number and the Schlieren density, do not significantly contribute to examining the numerical results.
11. In Section 5, Figure 17 illustrates a chemical time scale. Is this the chemical time scale from EDC-PaSR?
12. Although a statistical steady state is not attained according to the authors, Figures 21, 22, and 23 illustrate mean fields of the quantities of interest, which is questionable.
13. Overall, the image quality should be substantially improved, particularly for Figure 5.
14. Lastly, although the authors state that the present research study fits into a design optimization concept (see the Abstract and Section 6), it is not clear how the numerical simulation contributes to providing insights about ignition and flame propagation characteristics that could help drive the optimization of geometrical design and operational parameters.
References
[Smith1999] Smith G. et al., “GRI-Mech 3.0”, 1999, http://combustion.berkeley.edu/gri- mech/.
Comments for author File: 
Comments.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsA manuscript on “Numerical Simulation of Flame propagation in 1kN GCH4/GO2 pintle engine” is an unorganised and very poor attempt to convert numerical analysis in to research article. There is no justification and need for the current study. There are no specific conclusions/observation which will help scientist/engineers involved in rocket design.
Major Observations/Suggestion:
1 Abstract is chaotic and unorganised. First line talks about H2 usage. It is miss fit in abstract. Abstract is poorly written. Abstract should consist of 5 elements i.e Introduction/problem statement, objective/aim, method, results (qualitative and quantitative) and conclusion. Many of these elements are missing.
2. Reviewed literature is mostly about modelling of transient combustion flow phenomena. Literature review is not able to justify need of current study.
3. Reference for SDES model is missing.
4. Present grid independence study for transient case instead of for steady stateanalysis.
5. Enhance quality of figure 5. Text of axis label and curve labels are not readable.
6. In figure 18, label for temperature curve is not available.
7. Data, labels and captions are not readable in figure 18 and 19.
8. Discussion/Conclusions must be pointwise and specific which are the take away message for the readers.
