Using numerical simulation to investigate the effect of injection configurations over the scramjet performance

https://doi.org/10.1016/j.matpr.2020.09.703Get rights and content

Abstract

Achieving combustion in the available residence time by efficient mixing of the injected fuel has been a challenge for scramjet designers since years. Design of the injection configuration do affect the patterns of mixing and in turn the rate of heat release for a given combustor geometry. The presented work attempts to investigate the effect of inclined injection configurations on the performance of a scramjet for a typical combustor geometry having intrusive fuel injection using strut. Over a decades of literature points towards the fact that numerical simulations are widely being accepted as promising method of investigation of flow inside a scramjet. Reacting simulations of hydrogen injection in an inlet air stream of Mach 2.5 are done with various fuel injection configurations. After validation of simulation results with the available experimental Sadatake Tomioka et al. (2001) results, different injection configurations were simulated for a constant equivalence ratio. The maximum combustion, mixing and heating was observed in near the strut surface, hence flow fields of these configurations are studied near the strut to understand the mechanism of mixing. Observations on performance parameters like total pressure loss, maximum top wall static pressure, flow enthalpy etc. are made to compare these configurations.

Introduction

Air breathing Scramjet engines are being investigated as an alternative to the existing bulky solid propellant fuel and oxidizer systems. These existing systems are much heavier than the payloads they deliver in space. Conventional ramjets using the oxygen in the air cannot be operated above Mach numbers of 3–4 due to the excessive temperatures reached while diffusing the incoming air in the combustor. These temperatures are above the practical limits of metallurgy and combustion. To deal with this problem, combustion in a supersonic stream without decelerating the air is being attempted in the engines called Scramjets. A limited time for mixing and combustion is available inside the scramjet due to very high velocities of the flow. Hence the flow fields in scramjet are critical to manage as we need to execute the combustion in a limited length of combustor. Different flight conditions varies the inlet conditions and makes the situation further complex. Combustor geometry and arrangement of injectors inside the combustor plays a key role in mixing of fuel and in turn the overall performance of the combustor [2]. Several studies have been reported to investigate the mixing and combustion phenomena particularly with the strut injection. K. M. Pandey et al. in his review paper exclusively on scramjet fuel injection, explained importance of different fuel injection schemes in mixing, combustion and thrust formation [3]. L. Abu-Farah et al. has carried out numerical simulation of single and multi-staged injection of hydrogen in a scramjet combustor. From the simulations it is clear that multi stage injection of H2 increases the H2 /air mixing by forming vortices and additional shock waves [4]. Juntao Chang et al. has done review on progress in strut based supersonic combustors and studied various phenomenon occurring during combustion, mixing, equivalence ratio and their effects on the flow [5]. Bhargav. A et al. has done computational study on effects of shockwave on combustion by predicting shock waves through CFD which in turn affects the combustion efficiency. Finally, through numerical simulation it is observed that shock wave impingement increases the combustion as well as mixing efficiency [6]. You et al. has worked on injection and mixing inside a scramjet combustor using DES and RANS method. Using these methods in the simulation it was found that flow features as well as the shock waves were captured more precisely. The formation of boundary layers and mixing layers can also be seen clearly through these studies [7]. Y. S. Chen et al. has done numerical study on the effect of mixing on the scramjet combustion and its effectiveness. It has been found that by the use of flame holding hydrogen, mixing of air and fuel is enhanced. This resulted in good combustion efficiency and improved thrust performance [8].

When a jet of gaseous fuel is injected in a supersonic stream a shear layer is created due to the velocity difference between the two streams. Turbulent micro mixing in these shear layers is known to be the prominent mechanism for mixing inside the scramjets [9]. Oblique shock wave impingement onto the fuel jet is also one of the known methods to enhance the molecular mixing between supersonic air and gaseous fuel [10]. The vorticity generated when a shock wave interacts with a shear layer results in enhanced combustion efficiency due to its immediate significance to the mixing enhancement in supersonic flows [10]. Along with the mixing enhancement, flame stabilization is also an important factor to be achieved to make the combustion effective. One of the methodologies for active flame stabilization is to create a local low velocity and possibly high chemically reactive regions so that the flame segment in this region anchors the bulk flame [11]. Three-dimensional bow shocks those are formed upstream due to the transverse injection causes the upstream wall boundary layer to separate. This separation region also acts as a flame -holder in combusting situations [12] where the boundary layer and jet fluids mix sub-sonically upstream of the jet exit. Many experimental and numerical analyses have been reported during the last few decades regarding the characteristics of the complex flow field resulting due to fuel–air mixing and combustion [13]. The flow fields especially downstream to the injection jet in a supersonic stream are complex and the modern simulation technique has emerged as a promising tool to investigate the same. Computer aided post–processing of the data obtained from computation can provide realistic picture of process occurring in combustors [13]. Injector shape and its orientation for optimum performance are the main interest of the researchers working in this field presently [13]. At supersonic speeds, the normal injection configuration is more useful in facilitating mixing [13] however, performance and mixing study for normal injection at some transverse angle across the direction of flow has not been studied in detail. The presented work attempts to understand the mechanism and nature of mixing behind the injected fuel jets, especially near the strut body by simulation. The study is essential since the region near strut surface act as a flame holding region affecting the whole combustion scenario inside the combustor. Results for various configurations and its impact on the performance of the combustor is studied.

Section snippets

Experimental work selected for the study

Based on the available experimental works a candidate geometry representing a typical combustor design is selected. Sadatake Tomioka et al. [1] experimented with a combustor having typical geometric features such as isolator, combustor and divergent section with an intrusive fuel injection using strut. The geometry details are as given in Fig. 1(a) for ready reference. It uses shock positioning as shown in Fig. 1(b) for mixing enhancement. The inlet conditions in the experiment corresponds to a

Configurations

To validate the code, initial simulation was carried out with 0.34 equivalence ratio with same injector configuration as used in the experimental case [1]. Based on identified gap in literature survey and the initial simulation different fuel injection configurations are proposed and performance in terms of mixing, pressure rise and total pressure loss was studied. The basis and thought process for designing these configurations could be noted down as below:

  • 1.

    Each injection hole of 2.5 mm is

Computational Setup: [15]

Omitting the facility nozzle, combustor geometry was modelled using CATIA modeller as given in Fig. 3 (a) with a symmetry plane dividing the 94.3 mm width into half to save the computational efforts. Total length of the geometry being 895 mm. The HPC facility provided by C-DAC Pune, on PARAM- YUVA- II was used for computations. Submission of the jobs remotely and managing the simulation runs was done by using PBS job script. A commercial solver code STARCCM + with the necessary licensing (POD)

Combustion chemistry and boundary conditions

For non-premixed combustion in a supersonic combustor the reaction is mainly governed by turbulent micro mixing [1] hence combustion is assumed to be mixing limited rather than kinetically limited. This assumption allows simplification in the modelling and is justified when model is validated with the experimental results [16]. Hence Standard Eddy Breakup model with K-ε turbulence was used to obtain the reaction rates.

Single-step finite rate chemistry performs reasonably well in predicting the

Methodology

Simulations for all configurations are carried out with combustion cases. The simulation results of A0 configuration are mainly used to validate the code. Further the results of all configurations including A0 are compared with each other. An attempt is made to understand the flow fields created by an oblique injection in a supersonic stream. Since the initial simulations have shown that most of the combustion, mixing and heating is occurring near the strut surface, a section plane 1 mm above

Results and discussion

The top wall static pressure graphs of experimental and simulated case for A0 configuration with 0.34 equivalence ratio are compared for validation of the CFD code. The results are as shown in Fig. 4.

The results show fairly good agreement with each other. The simulation overestimates the maximum wall static pressure by showing a spike at specific location. Reasons like the computational nature of solution and limited number of sensors in experimental case compared to high resolution in

Conclusions

Flow fields around the injected jet inclined in a plane perpendicular to the direction of supersonic flow are investigated. Representative sketches are prepared that can be used for reasoning of some behavioural aspects of reacting mixture behind injectors in scramjets. An exact three-dimensional sketch for the same can be created by obtaining the same kind of streamline like sketches over multiple planes.

The results thus reveals that the change in angle of injection from 90° reduce the

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (21)

  • Y. You et al.

    Injection and mixing in a scramjet combustor: DES and RANS studies

    Proc. Combust. Inst.

    (2013)
  • Y.S. Chen et al.

    Mixing effectiveness study in scramjet combustion

    Procedia Eng.

    (2013)
  • K. Kumaran et al.

    Investigation of the effect of chemistry models on the numerical predictions of the supersonic combustion of hydrogen

    Combust. Flame

    (2009)
  • Sadatake Tomioka et al.

    Combustion tests of staged supersonic combustor with a strut

    J. Propul. Power

    (2001)
  • Nobuo Chinzei, Tomoyuki Komuro, Kenji Kudou, “Effects of injector geometry on scramjet combustor performance”, J....
  • K.M. Pandey et al.

    Recent advances in scramjet fuel injection - A

    Int. J. Chem. Eng. Appl.

    (2010)
  • Abu-Farahn et al.

    Numerical simulations of single and multi-staged injection of H2 in a supersonic scramjet combustor

    Propul. Power Res.

    (2014)
  • Juntao Chang et al.

    Research progress on strut-equipped supersonic combustors for scramjet application

    Progr. Aerospace Sci.

    (2018)
  • J. Naidu, A. Bhargav,“Computational study on effects of shockwave on combustion and injected fuel by unconventional...
  • William H.Heiser, David T. Pratt, Hypersonic Air-breathing Propulsion, Second Printing ed., Washington, DC 20024 -...
There are more references available in the full text version of this article.

Cited by (0)

View full text