• Proceedings of the KSME Conference
• /
• 2003.11a
• /
• pp.550-555
• /
• 2003

When a shock wave arrives at an open end of tube, an impulse wave is discharged from the tube exit and complicated flow is formed near tube exit. The flow field is influenced by the cross-sectional geometry of tube exit, such as circular, square, rectangular, trapezoid and etc. In the current study, three-dimensional propagation characteristics of impulse wave discharged from the tube exit with non-circular cross section are numerically investigated using a CFD method. Total variation diminishing (TVD) scheme is used to solve the three-dimensional, unsteady, compressible Euler equations. Computations are performed for the Mach numbers of the incident shock wave $M_{s}$ below 1.5. The results obtained show that the peak pressure of the impulse wave and propagation directivity depends on the cross-sectional geometry of tube exit and the Mach number of incident shock wave.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2009.05a
• /
• pp.88-92
• /
• 2009

Total pressure recovery ratio in intake is crucial factor to the operational characteristics of supersonic propulsion system because it does not compress inlet air mechanically by compressor, but does compress inlet air by ram compression. As the result of that the dynamic characteristic analysis of engine was performed before the controller was designed, it could be ascertained when the AoA of flight vehicle increases, the buzz margin decreases so that the shock wave produced outside intake in the specified area according to flight operation's characteristics. Therefore the PID control algorithm was designed to be controlled buzz margin that the characteristic of shock wave could meet the requirement of performance in intake. The PID controller was designed that the buzz margin value is being positive number using the control variables; fuel flow and nozzle throat area.

• Proceedings of the KSME Conference
• /
• 2001.11b
• /
• pp.432-437
• /
• 2001

A computational work of the impulse wave which is discharged from the open end of a pipe is compared to the Lighthill's aeroacoustics theory. The second-order total variation diminishing(TVD) scheme is employed to solve the axisymmetric, compressible, unsteady Euler equations. The relationship between the initial compressure wave form and the resulting impulse wave is characterized in terms of the peak pressure. The overpressure, pressure gradient and wavelength of the initial compression wave are changed to investigate the influence of the initial compressure wave form on the peak pressure of impulse wave. The results obtained show that for the initial compression wave of a large wavelength and small pressure gradient the peak pressure of the impulse wave depends upon the wavelength and pressure gradient of compression wave, but for the initial compression wave of a short wavelength and large pressure gradient the peak pressure of the impulse wave is almost constant regardless of the wavelength and pressure gradient of compression wave. The peak pressure of the impulse wave is increased with an increase in the overpressure of the initial compression wave. The results from the numerical analysis are well compared to the results from the aeroacoutics theory with a good agreement.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.12 no.1
• /
• pp.48-56
• /
• 2002

A computational work of the impulse wave which is discharged from the open end of a pipe is compared to the Lighthill\`s aeroacoustics theory. The second-order total variation diminishing(TVD) scheme is employed to solve the axisymmetric, compressible, unsteady Euler equations. The relationship between the initial compressure wave form and the resulting impulse wave is characterized in terms of the peak pressure. The overpressure, pressure gradient and wavelength of the initial compression wave are changed to investigate the influence of the initial compressure wave form on the peak pressure of impulse wave. The results obtained show that for the initial compression wave of a large wavelength and small pressure gradient the peak pressure of the impulse wave depends upon the wavelength and pressure gradient of compression wave, but for the initial compression wave of a short wavelength and large pressure gradient the peak pressure of the impulse wave is almost constant regardless of the wavelength and pressure gradient of compression wave. The peak pressure of the impulse wave is increased with an increase in the overpressure of the initial compression wave. The results from the numerical ana1ysis are well compared to the results from the aeroacoutics theory with a food agreement.

• Journal of the KSME
• /
• v.29 no.4
• /
• pp.376-384
• /
• 1989

압축성 유동은 유동의 속도(Mach 수)에 따라, 아음속에서 초음속으로 바뀜에 따라 지배 방정 식의 형태가 바뀜은 물론이고 천음속에서는 아음속과 초음속이 공존하므로 지배방정식 자체도 두 가지 형태가 공존하는 어려움이 앞선다. 또한 압축성 유동장에서는 비압축성 유동장에서는 없는 충격파를 전후하여 유동변수들의 큰 불연속이 존재하게 되고, 이러한 불연속선을 유한한 크기의 계산격차를 사용하여 정확하게 해석한다는 것이 그렇게 쉬운 일이 아니다. 또한 저속의 비압 축성 유동에 비하여 일반적으로 보다 많은 독립변수들을 동시에 다루어야 하므로 대형컴 퓨터 용량과 빠른 계산속도를 요구하게 되며 국내에서는 2∼3년 전까지만 하더라도 실질적인 연구가 매우 어려운 실정이었다. 따라서 본 글에서는 최근 들어 대학실험실에 고성능 퍼스널 컴퓨터의 도입과 더불어 활발하게 진행되고 있는 압축성 유동의 수치해석 연구에 대하여 그 동 안의 국내연구 결과들을 모아 분류, 정리해 보고 앞으로의 연구에 대하여 간단히 언급해 보고자 한다.

• Proceedings of the KSME Conference
• /
• 2001.06e
• /
• pp.349-354
• /
• 2001

The present study addresses the open end correction associated with the reflection and discharge phenomena of a weak shock wave from an open end of a duct. The open end correction of the weak shock wave is investigated experimentally and by numerical computation. An experiment is made using a simple shock tube with an open end, and computation is performed to simulate the experimental flow field using the unsteady, axisymmetric, compressible, flow governing equations. The results obtained show that an open end correction should be involved for shock wave discharge and reflection problems generated from the exit of the duct with an open end baffle plate. With a baffle plate less than three times the duct diameter, it is found that the open end correction is a function of both the diameter of the baffle plate and normal shock wave magnitude. However, for a baffle plate larger than three times the duct diameter, it is independent of the baffle plate diameter. The present computations predict the results of shock tube experiment with good accuracy. A new empirical equation for prediction of the open end correction is found for the weak shock reflection and discharge phenomena occurring at the open end of the duct with and without a baffle plate.

• Journal of the Korean Society of Visualization
• /
• v.8 no.3
• /
• pp.56-62
• /
• 2010

The interaction patterns between shock wave and boundary layer in a rocket nozzle are mainly classified into two categories, FSS(Free Shock Separation) and RSS(Restricted Shock Separation), both of which are associated with the thrust characteristics as well as side loads of the engine. According to the previous investigations, strong side loads of the engine are produced during the period of transition from FSS to RSS or vice versa. The present work aims at investigating the unsteady behavior of the separation shock waves in a two-dimensional supersonic nozzle, using experimental method and CFD. Schlieren optical method was employed to visualize the time-mean and time-dependent shock motions in the nozzle. The unsteady, compressible N-S equations with SST K-$\omega$ turbulence closure were solved using a fully implicit finite volume scheme. The results obtained show the separation shock motions during the transition of the interaction pattern.

• Journal of computational fluids engineering
• /
• v.18 no.2
• /
• pp.56-65
• /
• 2013

Two compression ramp problems and an impinging shock problem are computed to investigate influence of turbulence models and eddy viscosity on the shock-wave / boundary layer interaction. A Navier-Stokes boundary layer generation code was applied to the generation of inflow boundary conditions. Computational results are validated well with the experimental data and effects of turbulence models are investigated. It is shown that the behavior of turbulence (eddy) viscosity directly affects both the extent of the separation and shock-wave positions over the separation.

• Proceedings of the KSME Conference
• /
• 2002.08a
• /
• pp.165-168
• /
• 2002

This paper describes the dynamics of the weak shock wave propagating inside some kinds of branched pipe bends. Computations are carried out by solving the two-dimensional, compressible, unsteady Euler Equations. The second-order TVD(Total Variation Diminishing) scheme is employed to discretize the governing equations. For computations, two types of branched pipe($90^{\circ}$ branch,$45^{\circ}$ branch) with a diameter of D are used. The incident normal shock wave is assumed at D upstream of the pipe bend entrance, and its Mach number is changed between 1.1 and 2.4. The flow fields are numerically visualized by using the pressure contours and computed schlieren images. The comparison with the experimental data performed for the purpose of validation of computational work. Reflection and diffraction of the propagating shock wave are clarified. The present computations predicted the experimented flow field with a good accuracy.

• Proceedings of the KSME Conference
• /
• 2001.11b
• /
• pp.417-422
• /
• 2001

The shock structure of supersonic, dual, coaxial jet is experimentally investigated. Eight different kinds of coaxial, dual nozzles are employed to observe the major features of the near field shock structure of the supersonic, coaxial, dual jets. Four convergent-divergent supersonic nozzles having the Mach number of 2.0 and 3.0, and are used to compare the coaxial jet flows discharging from two sonic nozzles. The primary pressure ratio is changed in the range between 4.0 and 10.0 and the assistant jet pressure ratio from 1.0 to 4.0. The results obtained show that the impinging angle, nozzle geometry and pressure ratio significantly affect the near field shock structure, Mach disk location and Mach disk diameter. The annular shock system is found depending the assistant and primary jet pressure ratios.