Hypersonic boundary-layer stabilization by means of ultrasonically absorptive carbon-carbon material Part 2: Computational Analysis Viola Wartemann∗, Tobias Giese†, Thino Eggers‡ German Aerospace Center (DLR), Institute of Aerodynamics and Flow Technology, Braunschweig, Germany
Alexander Wagner§, Klaus Hannemann¶ German Aerospace Center (DLR), Institute of Aerodynamics and Flow Technology, G¨ ottingen, Germany The damping of acoustic second mode instabilities by passive porous surfaces is investigated numerically and compared with experiments for different hypersonic boundary-layer flows. The used geometry is a blunt 7◦ half-angle cone model with exchangeable nose. The cone surface consists partly of an ultrasonically absorptive material, which has a natural, random porosity. For the analyses of the second modes the DLR stability code NOLOT, NOnLocal Transition analysis code, is used. The influence of the nose radius and the unit Reynolds number on the second modes is investigated using a smooth surface. These results of the smooth surface are compared with those of the porous surface to study the damping effect on the second modes and the transition shift. The numerical results are compared with wind tunnel measurements, which were performed in the DLR High Enthalpy Shock Tunnel G¨ ottingen (HEG).
Nomenclature A B d J0 , J1 m Ma n Pr q r Rnose Re s T u, v, w x, y, z Z0
Admittance Thermal admittance Pore depth Bessel functions Propagation constant Mach number Porosity Prandtl number Flow/material quantity Pore radius Nose radius Reynolds number Surface coordinate Temperature Velocity in x-, y-, z-direction Coordinates Characteristic impedance
∗ Research
Scientist, Spacecraft Department, DLR Braunschweig,
[email protected], Member AIAA Student, Spacecraft Department, DLR Braunschweig,
[email protected] ‡ Head Spacecraft Department, DLR Braunschweig,
[email protected], Member AIAA § Research Scientist, Spacecraft Department, DLR G¨ ottingen,
[email protected] ¶ Head Spacecraft Department, DLR G¨ ottingen,
[email protected], Senior Member AIAA † Undergraduate
1 of 15 American Institute of Aeronautics and Astronautics
α, β −αi δ µ λ % ω
Wave numbers Growth rate Boundary-layer thickness Viscosity Wavelength Density Frequency
Indices i u w
Imaginary part unit Wall quantity
Superscripts −
˜ ˆ
Base flow quantity Disturbance flow quantity Eigenfunction
I.
Introduction
Hypersonic laminar flow control to achieve delayed laminar-turbulent transition is of high importance for re-entry vehicles since early transition results in an increase in the heat transfer by factors between 3 and 8, resulting in a cost and weight increase of thermal protection systems.16, 17 A lot of different strategies are used to delay or prevent the transition process. Classical flow control strategy began with natural laminar flow (NLF), which has been applied since the 1930s and implies delaying transition by modifying the body shape. By contrast, active laminar flow control (LFC), which started around the same time, uses suction or wall cooling/heating to influence the boundary-layer instability modes.8 In the present work the manipulation of the transition is performed in a passive way, which usually provides a lower cost compared to active systems. Also the hypersonic environmental conditions make the use of active LFC more difficult than in other flight regimes. In this paper the manipulation of the transition is performed by the use of porous surfaces to influence the growth of the second mode. The second mode, or so called Mack mode,11 is the dominant mode for the transition process at hypersonic Mach numbers. In an early work, Fedorov et al.3 have shown theoretically by using linear stability theory a strong stabilization effect of the second mode by a passive porous surface, which partially dissipates disturbance energy inside the pores. Using an analytical model of flow within blind, thin pores, typically with about 10-20 pores per wavelength of the Mack modes and a porosity about 25 %, the growth rate reduction is predicted to be a factor of two.3 Experiments of Rasheed et al.14 on a 5◦ half-angle sharp cone (Ma∞ ∼ 5-6) equipped with equally spaced, blind, cylindrically micropores, confirmed the theoretical predictions. Continuation activities, theoretical as well as experimental ones, around the working group of Fedorov have demonstrated robustness of this laminar flow control concept (see e.g.4, 5, 12 ). Direct numerical simulations confirmed the results of the linear stability theory for smooth and porous surfaces equally well (see e.g.2, 15 ). In this paper the stability code NOLOT of the Germany Aerospace Center (DLR) is used to predict the second modes for a hypersonic boundary-layer flow. The used geometry is a blunt 7◦ half-angle cone model with exchangeable nose. One third of the cone surface was equipped with C/C (carbon/carbon), a fibre reinforced ceramic material with a natural, random porosity. In the first part of this paper (section IV) the influence of the nose radius and the unit Reynolds number on the Mack modes of a cone with smooth surface is investigated numerically and compared with experimental results. In the second part (section V) the comparison of the measured and calculated damping of the Mack modes and the effect on the transition
2 of 15 American Institute of Aeronautics and Astronautics
by the ultrasonically absorptive ceramic material is analyzed.
II.
Numeric methods
The NOLOT code, NOnLocal Transition analysis code,7 was developed in cooperation between DLR and the Swedish Defence Research Agency (FOI) and can be used for local as well as non-local analyses. The equations are derived from the conservation equations of mass, momentum and energy, which govern the flow of a viscous, compressible, ideal gas. All flow and material quantities are decomposed into a steady laminar base flow q¯ and an unsteady disturbance flow q˜ q(x, y, z, t) = q¯(x, y) + q˜(x, y, z, t).
(1)
The laminar base-flow q¯ is calculated by the DLR FLOWer code.10 The FLOWer code solves the compressible Navier-Stokes equations for a perfect gas flow on block-structured grids, using second order finite volume techniques and cell-centered or cell vertex variables. The disturbance q˜ is represented as a harmonic wave q˜(x, y, z, t) = qˆ(x, y) exp[i(αx + βz − ωt)]
(2)
with the complex-valued amplitude function qˆ. In the following, the hat over a variable denotes an amplitude function. In this paper the local approach is used, which is a subset of the nonlocal stability equations. Since NOLOT is a spatial code, the wavenumbers α and β are complex quantities and the frequency ω is a real value. The growth rate −αi stands for the quantity of primary interest. The boundary conditions in NOLOT for a smooth wall (at y = 0) are: u ˆw , vˆw , w ˆw , Tˆw = 0. (3) The NOLOT code is validated by several test cases against published results, including DNS (direct numerical simulation), PSE (parabolized stability equations), multiple scales methods and LST (linear stability theory). A good summary of the validation is given by Hein et al.7 For the treatment of porous surfaces additional boundary conditions are implemented. The boundary conditions for porous walls are taken from Koslov et al.9 and Maslov et al.,12 a complete derivation can be found in these references. The conditions are given by u ˆw , vˆw , Tˆw = 0, w ˆw = Aˆ pw , (4) where a subscript w denotes a value at the wall. The admittance A is calculated by A=
n tanh(md). Z0
(5)
The pores of the boundary condition are equally spaced, blind, cylindrical pores with a depth d, a radius r, both normalized with the displacement thickness, and a porosity n. The characteristic impedance Z0 and the propagation constant m are q q ρˆ iωMa ρˆCˆ ˆ C √ , √ m= , (6) Z0 = − Ma Tw Tw where the dimensionless complex dynamic density ρˆ and dynamic compressibility Cˆ are expressed as ρˆ =
1 ˆ , Cˆ = 1 + (γ − 1)F2 (Λ). 1 − F1 (Λ)
(7)
The function F and the macroscopic parameter Λ depend on the pore shape: F (Λ) =
2J1 (Λ) , ΛJ0 (Λ)
ˆ where J0 and J1 are Bessel functions of the argument Λ and Λ: s √ iωρ0 ˆ = Λ P r. Λ=r , Λ µ 3 of 15 American Institute of Aeronautics and Astronautics
(8)
(9)
ω represents the dimensionless angular frequency. All dimensionless flow quantities are normalized by the values at the boundary-layer edge. The verification of the in NOLOT implemented boundary conditions of porous walls against the Southsampton LST code and direct numerical simulations is given in Wartemann et al.24, 25 The validation was based on experiments using a sharp cone, which has a smooth surface on one side and is covered with cylindrical, equally spaced, blind micropores on the other side.26, 27 For the calculation of the N-factors the most amplified wave angle is used and the frequency is kept constant: Z s a = −αi ds. (10) N = ln a0 s0 s0 denotes the streamwise position where the disturbances start to grow and a0 is the amplitude at this position.
III.
Wind tunnel and wind tunnel model
The experiments were performed in the DLR High Enthalpy Shock Tunnel G¨ottingen (HEG). The tunnel is a free driven shock tunnel, which allows to investigate flow past hypersonic flight configurations from low altitude Mach 6 up to Mach 10 at approximately 33 km altitude.6 The investigated test series in this paper has a free stream Mach number of 7.5 at different unit Reynolds numbers: Reu = 1.4 × 106 /m and Reu = 2.4 × 106 /m. The cone model is a 7◦ half-angle blunted cone with an overall length of 1, 100 mm and an exchangeable nose with nose tip radii of 2.5 mm and 5 mm. Figure 1 shows the model at the final stage of the manufacturing process at DLR. The tested model was equipped with 49 thermocouples, 8 pressure transducers of type KULITE XCL-100-100A and 12 piezoelectric PCB132A32 fast response pressure transducers. The model was supported by a sting system in HEG at nominal 0◦ angle of attack.
Figure 1. Photographic view of the model during manufacturing
The instrumentation was chosen such that the transition location on the cone was detectable by evaluating the surface heat-flux distribution obtained from the thermocouple readings. The flush-mounted KULITE pressure transducers were used to quantitatively measure the surface pressure so that the angle of attack and the yaw angle could be controlled. Twelve PCB132A32 fast response pressure transducers were grouped in pairs and flush mounted along the model at x = 650 mm, x = 785 mm, x = 950 mm. These pressure transducers feature a resonance frequency of 1 MHz or higher and were used to measure pressure fluctuations in the boundary-layer above the cone surface. One third of the surface was equipped with C/C (carbon/carbon), a fibre reinforced ceramic material with a natural, random porosity, which was manufactured by the DLR Institute of Structures and Design in Stuttgart. The ceramic insert has a minimum thickness of 5 mm. Using quicksilver-porosimetry, the hydraulic diameter and the porosity, which are needed for the calculations, can be measured/calculated. The porosity is about 15 % and the median hydraulic pore diameter is about 13 µm. A description of the quicksilver-porosimetry technique, the detailed analyzed pore side range and the model itself can be found
4 of 15 American Institute of Aeronautics and Astronautics
in Wagner et al.22, 23
IV.
Smooth surface
In the first part of this section the principal method of comparing measured and calculated results is introduced. The second part investigates the influence of the unit Reynolds number and the nose radius on the transition process, especially on the Mack modes. IV.A.
Mack mode comparison
In this section the principal method of the Mack mode comparison between experimental and numerical results is explained. Further, small differences are analyzed choosing a test case with a cone nose tip radius of 5 mm and a free stream condition at an unit Reynolds number of Reu = 1, 4 × 106 /m.
Figure 2. Smooth wall: N -factor as function of the normalized x-position - numerical results
In figure 2 the N-factors of the second mode calculated by NOLOT as a function of the normalized x-position are shown for different frequencies. Out of this diagram the N-factors can be extracted from the sensor positions for various frequencies, which can be seen in figure 3 for the three PCB sensor positions. The Mack mode is amplified in streamwise direction. Due to the increase of the boundary-layer thickness the frequencies of the distribution are shifted to lower values, which can be explained with the following relation between boundary-layer thickness δ and wavelength λ11, 20 : λ ≈ 2 δ.
(11)
For the Mack mode comparison of the numerical data with the measurements two requirements are necessary. First, the Mack mode at the PCB sensor position has to be strong enough, meaning it has to be higher than the background noise level. Second, the flow at the sensor position has to be laminar. Otherwise a breakup of the Mack mode due to the transition process is possible and most likely. In the current case, the measured Mack mode at the first sensor position (x = 650 mm) is still too weak and in the range of the background noise level. The last position (x = 950 mm) is in the transition region. Thus, for the comparison with the measured Mack mode, the second sensor position (x = 785 mm) is used. Figure 4a shows the measured power spectral density (PSD), which is given in pascal2 per hertz, as a function of the measured frequency marked as black symbols. As NOLOT result, the N-factor as a function of the frequency is shown as red line. Consequently, this is a comparison of the frequency range. For comparing the damping 5 of 15 American Institute of Aeronautics and Astronautics
Figure 3. Smooth wall: N -factor as function of the frequency at the different PCB sensor positions - numerical results
Figure 4. Smooth wall: Comparison NOLOT/PCB sensor
6 of 15 American Institute of Aeronautics and Astronautics
or increase of the Mack modes between measured and calculated amplitude in the following sections, eN instead of N has to be used. The distribution of eN is consequently smaller as of N (compare figure 4a and 4b). As visible in both figures the numerically predicted N -factor distribution is in good agreement with the measured data. The difference of the maximum frequency is lower than 10 % and therefore in a typical range for this kind of measurements.1 Wagner et al.21 investigated the influence of small deviations of the nose radius on the Mack mode. A change in the nose radius of 10 % results in a shift of the maximum frequency of about 10 %. Further, in Wagner et al.21 the influence of small variations of the angle of attack on the transition location and the Mack mode development are analyzed. In two subsequent tests with the same model as used here the angle of attack was varied in a range of ± 0.2◦ . The analysis shows a negligible effect on the transition process. In the test series of this paper the angle of attack of the model in HEG is adjusted using an inclinometer with a measuring inaccuracy below 0.03◦ . In Wartemann et al.26 the influence of usual measurement inaccuracies in the determination of the free stream conditions and the influence of the wall temperature on the maximum frequency of the Mack modes are investigated using a 3◦ half-angle sharp cone. It is shown that a change of the unit Reynolds number of 5 % results in a shift of the maximum frequency of about 2 %. To minimize these usual uncertainties an accurate determination of the named parameters is necessary and was performed. For each experimental test, e.g., the determination of the free stream was performed separately. In figure 5 the N-factor as a function of the frequency for the three sensor positions is shown to illustrate the influence on the Mack mode using the same wind tunnel condition in different experimental tests. The nose radius in these cases is 2.5 mm, the unit Reynolds number is Reu = 1, 4 × 106 /m or Reu = 1, 4 × 106 /m + 5 % at the same Mach number of 7.5. As visible, a change of the unit Reynolds number by 5 % results in a shift of the maximum frequency of about 10 %, depending on the position of the PCB sensor.
Figure 5. Smooth wall: Small changes of the unit Reynolds number - numerical results
IV.B.
Influence of the nose radius
For the analysis of the influence of the nose radius, it is necessary to specify the bluntness of the cone. The nose tip radii are 2.5 mm and 5 mm. According to Softley18 the bluntness of the cone can be identified with the nose Reynolds number Renose , which can be calculated with the nose radius Rnose and the unit Reynolds number Reu Renose = Reu Rnose .
7 of 15 American Institute of Aeronautics and Astronautics
(12)
Figure 6a (taken from Softley18 ) shows the transition Reynolds number as a function of the nose Reynolds number based on experiments with a blunt 5◦ half-angle cone with exchangeable nose. In the region where an increase of the nose Reynolds number stabilizes the boundary-layer flow the cone is defined as cone with small bluntness. In contrast, in the region with large bluntness an increase of the nose Reynolds number has a destabilizing effect. This behavior is called transition reversal. This paper is focused on cones with small bluntness. The HEG test cases, which are plotted in figure 6b together with the trend function of a cone with small bluntness following Softley,18 are for nose radii of 2.5 mm (black color) and 5 mm (orange color) at a unit Reynolds number of Reu = 1, 4 × 106 /m (x symbols) and Reu = 2, 4 × 106 /m (triangle symbols). All cases are in the small bluntness region and follow the expected transition trend of this region. The stability effect of the nose bluntness can be explained partially with the reduction of the local Reynolds number based on the boundary-layer edge quantities due to the entropy layer produced by the curved bow shock.13
Figure 6. Smooth wall: left: transition reversal - small and large bluntness18 and right: cone with small bluntness including HEG results
As described in the previous subsection, the Mack mode measured at the sensor position should provide a signal above the background noise level and a laminar flow is also needed for the Mack mode comparison of the numerical data with the measurements. Thus, the sensor at position x = 785 mm for a unit Reynolds number of Reu = 1.4 × 106 /m is analyzed in figure 7 for the comparison; the results of the nose tip radius of 2.5 mm are marked by black color and the 5 mm nose by orange color: lines are calculations of NOLOT and symbols are measurements. The numerical as well as experimental results confirmed the previous described trend: An increase of the nose radius has a stabilizing effect, visible in the decrease of the Mack modes. Further, an increase of the nose radius results in a shift to lower frequencies due to the increase of the boundary-layer thickness (see equation 11). Comparing the calculations with the measurements for both nose radii, the difference of the maximum frequency is under 10 %, which was discussed in the previous subsection. Furthermore a good agreement in the decrease of the Mack mode due to the increasing nose radius is visible. This confirms a comparison of N the ratio eN Rnose =2.5 /eRnose =5 = 3.2 and P SDRnose =2.5 /P SDRnose =5 = 3.3, using the corresponding values at the maxima. IV.C.
Influence of the unit Reynolds number
In this section the effect of the unit Reynolds number on the Mack modes is investigated using a cone tip radius of 5 mm.
8 of 15 American Institute of Aeronautics and Astronautics
Figure 7. Smooth wall: Variation of nose radius
Figure 8 shows numerical eN predictions as a function of the frequency for different unit Reynolds numbers: Reu = 1.4 × 106 /m (dash-dot lines) and Reu = 2.4 × 106 /m (solid lines) for all PCB sensor positions. The increase of the unit Reynolds number results in an increase of the eN -values and the frequencies as visible in figure 8. As test case for the comparison of calculated and measured Mack modes, the second sensor position x = 785 mm for both unit Reynolds numbers Reu = 1.4 × 106 /m (orange color) and Reu = 2.4 × 106 /m (black color) is chosen in figure 9. The NOLOT calculations are marked again as lines and the measurements as symbols. A good agreement between measurements and calculations is visible for the maximum frequencies as well as the increase of the Mack modes due to the increase of the unit Reynolds number. The ratios N eN Re=2.4 /eRe=1.4 = 2.8 and P SDRe=2.4 /P SDRe=1.4 = 2.9, using the corresponding values at the maxima, confirm the result. The same trends as described by Softley18 (see figure 6b) occur if the nose radius is kept constant and the unit Reynolds number is increased.19 The results analyzed here of the used cone with small bluntness confirmed an increase of the Mack modes, with increasing unit Reynolds number, and an upstream shift of the transition, in this case by about (xtrans,Re=2.4 − xtrans,Re=1.4 )/L = −0.1.
V.
Porous surface
The previous section shows a very good agreement of the calculated and measured Mack modes using a smooth surface. This is the basic requirement of the analyses of this section: a detailed study of the damping effect on the Mack modes and the resulting transition shift using the described porous surface of section III. As test case for the comparison a cone with a nose tip of 5 mm at different free stream conditions is chosen: Reu = 1.4 × 106 /m and Reu = 2.4 × 106 /m.
9 of 15 American Institute of Aeronautics and Astronautics
Figure 8. Smooth wall: Variation of the unit Reynolds number - numerical results
Figure 9. Smooth wall: Variation of the unit Reynolds number - comparison: NOLOT/PCB sensors
10 of 15 American Institute of Aeronautics and Astronautics
V.A.
Investigations at Reu = 1.4 × 106 /m
The first investigated test case has a unit Reynolds number of Reu = 1.4 × 106 /m.
Figure 10. Smooth/porous wall: Growth rate as function of the normalized x-position - Reu = 1.4 × 106 /m
Figure 11. Smooth/porous wall: Comparison NOLOT/PCB sensors - Reu = 1.4 × 106 /m
Figure 10 shows the calculated growth rate of the Mack mode. The black lines mark the results of the smooth cone side and the red lines the results of the porous cone side. The growth rates are completely damped by the porous surface, thus the N -factor and, consequently, the eN -values are zero. As a result, NOLOT predicts a laminar boundary-layer flow for the cone with porous surface, which is in agreement with the experiment. The thermocouple measurements show, that on the smooth cone side transition occurs at the downstream end of the cone, whereas on the porous side the boundary-layer flow is laminar. For a detailed comparison of the measured/calculated results, figure 11 shows the Mack mode comparison. As stated before, the Mack mode measured at the first sensor position is still too weak and in the range of the background noise level, thus the last two sensors are used for comparison. The NOLOT calculations for the smooth cone, marked as line, are compared with the wind tunnel measurements (symbols) for the solid cone, marked with black color, and the porous cone marked with red color. As already stated, the pressure
11 of 15 American Institute of Aeronautics and Astronautics
data can not be compared directly with the numerically calculated eN -values. It is nevertheless possible to compare the damping of the Mack mode. NOLOT predicts for the sensor position x = 765 mm a complete damping, which is in agreement with the measurements (figure 11a). The Mack mode is completely damped or in the range of the background noise level. The last sensor position x = 950 mm of the smooth side is in the transition region, thus it can not be excluded that the breakdown of the Mack modes due to the transition has already started. Hence it is not possible to give the measured damping using x = 950 mm. Since the transition in the experiment is in the range of the transition beginning at this position, it is used to make a worst case analysis to give the maximal possible difference between the measured Mack mode damping and the one predicted by NOLOT. For the calculation of the percentaged, measured damping of the Mack mode the PSD values at the maxima minus the background noise level, which is about 50000 Pa2 /Hz, are used: 100 % ·
P SDporous,max − P SDnoise ≈ 30 %. P SDsmooth,max − P SDnoise
(13)
Consequently, the measurements show a 70 % damping of the Mack mode using the porous surface instead of the smooth one. NOLOT predicts a 100 % damping. This results in a maximal possible difference of 30 % between measurements and calculations for this worst case analysis. V.B.
Investigations at Reu = 2.4 × 106 /m
The second investigated test case has a unit Reynolds number of Reu = 2.4 × 106 /m.
Figure 12. Smooth/porous wall: Growth rate as function of the normalized x-position - Reu = 2.4 × 106 /m
Figure 12 shows the calculated growth rate of the Mack mode. Again the black lines mark the results of the smooth cone side and the red lines the results of the porous cone side. As for the lower unit Reynolds number of the previous test case, the growth rates are completely damped by the porous surface, thus the N -factor and, consequently, the eN -values are zero. As a result, NOLOT predicts a laminar boundary-layer flow for the porous cone, which is in agreement with the experiment. The measurements show, that on the smooth cone side transition occurs at the downstream end of the cone, whereas on the porous side the boundary-layer flow is laminar. Figure 13 shows the Mack mode comparison for all PCB sensor positions. The NOLOT calculations for the smooth cone, marked as line, are compared with the wind tunnel measurements (symbols) for the solid cone, marked with black color, and the porous cone marked with red color. NOLOT predicts for all three sensor positions a complete damping of the Mack mode, which is in agreement with the measurements of the first two sensors (see figure 13a and 13b). There the measured Mack mode is completely
12 of 15 American Institute of Aeronautics and Astronautics
Figure 13. Smooth/porous wall: Comparison NOLOT/PCB sensors - Reu = 2.4 × 106 /m
damped or in the range of the background noise level. In the experiments the last sensor position of the smooth side is in the transition region, thus for this position it can not be excluded that the breakdown of the Mack modes due to the transition has already started. Hence it is not possible to evaluate the measured damping. In contrast to the lower unit Reynolds number case of the previous subsection, the sensor position is not longer at the transition beginning. Consequently, no further damping analysis is performed. But the measured Mack mode on the porous side is still visible in the experiments (see figure 13c), which indicates that the calculated damping is by trend slightly higher than the measured one.
VI.
Summary
The damping of acoustic second mode instabilities by passive porous surfaces is investigated numerically and compared successfully with experiments for a hypersonic boundary-layer flow. The used geometry is a blunt 7◦ half-angle cone model with exchangeable nose. One third of the cone surface was equipped with C/C (carbon/carbon), a fibre reinforced ceramic material with a natural, random porosity. For the analyses of the second modes the DLR stability code NOLOT, NOnLocal Transition analysis code, is used. In section IV Mack mode investigations using a smooth surface are shown. The influence of the nose radius and the unit Reynolds number on the Mack modes is analyzed. Since a cone with small bluntness has been chosen for the analysis, an increase of the nose radius stabilizes the boundary-layer flow, which results in a decrease of the Mack modes. An increase of the unit Reynolds numbers leads to an increase of the Mack modes. The calculated Mack modes are in very good agreement with the measurements. This conformity
13 of 15 American Institute of Aeronautics and Astronautics
for the smooth surface results is the basic requirement for the investigations using the porous cone surface. For different free stream conditions a significant damping of the Mack modes is demonstrated, using the porous surface with random, natural porosity instead of the smooth surface. The numerical predicted damping of the Mack modes is by trend slightly higher than the measured damping. The damping results in all cases in a downstream transition shift. NOLOT predicts a complete laminar flow for the cone with porous surface, which is confirmed by the thermocouple measurements.
References 1 Alba, C. R., Casper, K. M., Beresh, S. J., Schneider, S. P., “Comparison of Experimentally Measured and Computed Second-Mode Disturbances in Hypersonic Boundary-Layers”, 48th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition, AIAA2010-897, Orlando, Florida, 2010 2 Bres, G. A., Colonius, T., Fedorov, A. V., “Stability of Temporally-Evolving Supersonic Boundary Layers over MicroCavities for Ultrasonic Absorptive Coatings”, AIAA Paper, AIAA2008-4337, Seattle, Washington, 2008 3 Fedorov, A. V., Malmuth, N. D., Rasheed, A., Hornung, H. G., “Stabilization of hypersonic boundary-layers by porous coatings”, AIAA Journal, Vol. 39, No. 4, 2001, pp. 605-610 4 Fedorov, A., “Temporal Stability of Hypersonic Boundary Layer on Porous Wall: Comparison of Theory with DNS”, 48th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition, AIAA2010-1242, Orlando, Florida, 2010 5 Fedorov, A., “Instability of hypersonic boundary-layer on a wall with resonating micro-cavities”, 49th AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition, AIAA2011-373, Orlando, Florida, 2011 6 Hannemann, K., Martinez Schramm, J., Sebastian, K., “Recent extensions to the High Enthalpy shock tunnel Gttingen (HEG)”, 2nd International ARA Days, Arcachon, Frankreich, 2008 7 Hein, S., Bertolotti, F. P., Simen, M., Hanifi, A., Henningson, D., “Linear nonlocal instability analysis - the linear NOLOT code”, DLR-IB, 223-94 A56, 1994 8 Joslin, R., “Overview of Laminar Flow Control”, NASA, TB 1998-208705, 1998 9 Koslov, V. F., Fedorov, A. V., Malmuth, N. D., “Acoustic properties of rarefied gases inside pores of simple geometries”, J. Acoust. Soc. Am. 117 (6), June 2005, pp. 3402-3412 10 Kroll, N., Fassbender, J. K., “MEGAFLOW - Numerical Flow Simulation for Aircraft Design”, Springer, Berlin, 2005 11 Mack, L. M., “Boundary layer linear stability theory”, AGARD Special course on stability and transition of laminar flow, 1984 12 Maslov, A. et al., “Experimental and theoretical studies of hypersonic laminar flow control using ultrasonically absorptive coatings (UAC)”, Report ISTC, 2172-2001, 2003 13 Moeckel, W. E., “Some effects of bluntness on boundary-layer transition and heat transfer at supersonic speeds”, NACA Report, No. 1312, 1957 14 Rasheed, A., Hornung, H. G., Fedorov, A. V., Malmuth, N. D., “Experiments on passive hypervelocity boundary-layer control using an ultrasonically absorptive surface”, AIAA Journal, Vol. 40, No. 3, 2002, pp. 481-489 15 Sandham, N. D., L¨ udeke, H.,“A numerical study of Mach 6 boundary-layer stabilization by means of a porous surface”, AIAA Journal, Vol. 47. No. 9, 2009 16 Schneider, S., “Flight Data for Boundary Layer Transition at Hypersonic and Supersonic Speeds”, Journal of Spacecraft and Rockets, Vol. 36, No. 1, 1999, pp. 8-20, Rockets, Vol. 36, No. 1, 1999, pp. 820 17 Schneider, S., “Hypersonic Laminar-Turbulent Transition on Circular Cones and Scramjets Forebodies”, Process in Aerospace Sciences, Vol. 40, No. 1-2, 2004, pp. 1-50 18 Softley, E. J., “Boundary layer transition on hypersonic blunt, slender cones”, AIAA Paper AIAA69-705, 1969 19 Stetson, K. F., “Effect of bluntness and angle of attack on boundary layer transition on cones and biconic confgurations”, AIAA Paper, AIAA79-0269, 1979 20 Stetson, K. F., Thompson, E. R., Donaldson, J. C., Siler, L. G., “Laminar boundary-layer stability experiments on a cone at Mach 8, Part 2: Blunt cone”, AIAA Paper, AIAA84-0006, 1984 21 Wagner, A., Hannemann, K., Wartemann, V., L¨ udeke, H., Tanno, H., Ito, K., “Free piston driven shock tunnel hypersonic boundary-layer transition experiments on a cone configuration”, Hypersonic Laminar-Turbulent Transition Conference, AVT200-12, San Diego, 2012 22 Wagner, A., Hannemann, K., Kuhn, M., “Experimental investigation of hypersonic boundary-layer stabilization on a cone by means of ultrasonically absorptive carbon-carbon material”, 18th AIAA/3AF International Space Planes and Hypersonic Systems and Technologies Conference, France, 2012 23 Wagner, A., Hannemann, K., Wartemann, V., Giese, T., “Hypersonic boundary-layer stabilization by means of ultrasonically absorptive carbon-carbon material - Part 1: Experimental results”, AIAA Aerospace Sciences Meeting, Texas, 2013 24 Wartemann, V., L¨ udeke, H., Sandham, N. D., “Stability analysis of hypersonic boundary-layer flow over microporous surfaces”, 16th AIAA/DLR/DGLR International Space Planes and Hypersonic Systems and Technologies Conference, 20097202, Bremen, Germany, 2009 25 Wartemann, V., L¨ udeke, H., “Investigation of Slip Boundary Conditions of Hypersonic Flow over microporous Surfaces”, V European Conference on Computational Fluid Dynamics, ECCOMAS CFD, Lisbon, Portugal, 2010 26 Wartemann, V., L¨ udeke, H., Willems, S., G¨ ulhan, A., “Stability analyses and validation of a porous surface boundary condition by hypersonic experiments on a cone model”, 7th Aerothermodynamics Symposium, Brugge, Belgium, 2011
14 of 15 American Institute of Aeronautics and Astronautics
27 Wartemann, V., Willems, S., G¨ ulhan, A., “Mack mode damping by micropores on a cone in hypersonic flow”, Hypersonic Laminar-Turbulent Transition Conference, AVT-200-15, San Diego, 2012
15 of 15 American Institute of Aeronautics and Astronautics
