Dual Numerical Solution for 3D Supersonic Laminar Flow Past a Blunt-Fin Junction: Change in Temperature Ratio as a Method of Flow Control
Round 1
Reviewer 1 Report
The paper numerically investigated the dual solution of hypersonic laminar flow over a blunt-fin on a plate. The authors found that the temperature factor impacts the solution type. The solutions of two types could be switched using a short time change in plate temperature factor. The problem is interesting and important for heating prediction. There are following issues, which authors need to address before the acceptance of this paper for the publication.
1. P1 1st paragraph (line 24 to 31): The research background introduction is much similar to that of Tutty et al.(2013). This identical expression should be avoided. The authors should describe it based on their own understanding.
2. The computational domain and grid need to be described in detail, especially the outflow scale and the grid near the wall, because the grid resolution in the boundary layer has a great impact on the heat flux simulation. Also, an illustration of the overall grid distribution needs to be given.
3. Section 3.1: It is understandable that different configurations or flow conditions (i.e. Reynolds number, Mach number and so on) would lead to different flow patterns. How do the authors obtain the dual solutions at a given configuration and fixed Re=1.25×104?
4. Section 3.2: It seems that the authors employed a quasi-steady method to evaluate the influence of the temperature factor. Is there any unsteady characteristics between the dual solutions due to the existence of separation and vortices in the flow? In addition, the simulation procedure was started from a dual solution temperature factor T_w/T_inf=4.76. Starting from a value of 5.95, only a single solution II exists. If decrease temperature factor further (for example, T_w =50K), can a single solution I be obtained? Further, when the simulation procedure was started from a higher T_w =450K or a lower T_w =50K, what is the difference to the present results?
5. Section 3.2: The mechanism of influence of temperature factor needs to be explained further.
6 Section 3.3: The authors used a short time change in plate temperature to control the switching between dual solutions. How is the time determined?
7. The boundary layer shown in Figure 11b is much thinner than that of Figure 3a and Figure 7a. Is there any explanation for this?
8. In discussion (line 298 to 299), the authors mentioned that solution II is more stable, why Tutty et al.(2013)’s experimental and numerical results obtain solution I?
Author Response
The authors are deeply grateful to the reviewer for valuable remarks and comments! In attachment there are our responses and a description of what has been changed in our manuscript.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Please address the following questions:
How can a step change of the wall temperature be implemented in experiments? How is the flow control affected if the wall temperature changes more slowly?
Why is the horseshoe system not unsteady? Did the authors vary the timestep size to check if this is a temporal resolution issue? Related question: The time-integration is implicit. How many iterations per timestep were used? The new timestep was converged to what accuracy?
What was the accuracy of the viscous terms?
Does the flow downstream of the junction remain laminar? It may be argued that the flow state downstream of the junction is of little importance for the junction. If that is the assumption, please say so.
How did you initialize the two different solutions? One of the two solutions is more stable than the other. If you compute for a long time, does the less stable solution persist? Related question, page 5, what is meant by metastable? How do you explain the temperature dependence of the solution?
Fig. 5: The agreement with Tutty is not good. In the experiment, the heat flux fluctuations are less pronounced, which suggests that the horseshoe vortex is weaker or unsteady. Please comment.
Tab. 1: x_v is distance to leading edge?
Page 7: You are writing that "the unique solution has the vortex structure". This is confusing: Both solutions have a vortex structure. For solution 2, the vortex is larger (or stronger) and closer to the leading edge.
Temperature factor is confusing. Consider calling it wall temperature ratio.
Author Response
The authors are deeply grateful to the reviewer for valuable remarks and comments! In attachment there are our responses and a description of what has been changed in our manuscript.
Author Response File: Author Response.pdf
Reviewer 3 Report
My congratulations to the Authors for a very interesting paper, and simple explanation of such complicated phenomena, along with high skills in presentation of the results. I've found only one, minor mistake: in line 229 should be "plate temperature" instead of "temperature plate".
Author Response
The authors are deeply grateful to the reviewer for his encouraging comments! The minor mistake noted by the reviewer has been corrected.
Round 2
Reviewer 1 Report
The authors responded appropriately to all concerns. I recommend publishing in its present form.
