NASA-TM-112157
AerothermodynamicMeasurement
and Predictionfor Modified Orbiter
at Mach 6 and 10
John R. Micol
Reprinted from
JournalofSpacecraftandRockets
Volume32,Number5,Pages737-748
A publication of the
American Institute of Aeronautics and Astronautics, Inc.
370 L'Enfant Promenade,SW
Washington, DC20024-2518
JOURNAl. OF SPACEC[_.AI:I" AND ROCKETS
"4ol. 32, No. 5, September-October 1995
Aerothermodynamic Measurement and Prediction for
Modified Orbiter at Mach 6 and 10
John R. Micol*
NASA Langley Research Center, Hampton, Virginia 23681-0001
Detailed heat-transfer rate distributions measured laterally over the windward surface of an orbiter-like
configuration using thin-film resistance heat-transfer gauges and globally using the newly developed relative inten-
sity, two-color thermographic phosphor technique are presented for Mach 6 and 10 in air. The angle of attack was
varied from 0 to 40 deg, and the freestream Reynolds number based on the model length was varied from 4 x l0 s
to 6 × 106 at Math 6, corresponding to laminar, transitional, and turbulent boundary layers; the Reynolds number
at Mach 10 was 4 × l0 s, corresponding to laminar flow. The primary objective of the present study was to provide
detailed benchmark heat-transfer data for the calibration of computational fluid-dynamics codes. Predictions from
a Navier-Stokes solver referred to as the Langley aerothermodynamic upwind relaxation algorithm and an ap-
proximate boundary-layer solving method known as the axisymmetric analog three-dimensional boundary layer
code are compared with measurement. In general, predicted laminar heat-transfer rates are in good agreement
with measurements.
Nomenclature
C_, = heat-transfer coeflicient, lbm/ft2-s, q/(h,w - h,,,), where
h_,w : h,.z
C_,._i = reference heat-transfer coefficient, lbm/fl_-s
/t = enthalpy, Btu/lbm
L = model length in symmetry plane, in. (see Fig. 2)
M = Mach number
p = pressure, lb/in. 2
q = heat-transfer rate, Btu/ft2-sec
r = radius, in.
Re = unit Reynolds number, ft 1
T = temperature, R
t = time, s
V = velocity, ft/s
.r = longitudinal distance, in.
v = lateral or spanwise distance, in.
ot = angle of attack, deg
y = ratio of specific heats
Suh_cril_tX
aw
b
L
l,l
t,2
I1_
2
= adiabatic wall
= base
= based on model length in symmetry plane
= reservoir conditions
= stagnation conditions behind the normal shock
= model surface; wall surface
= static conditions behind the normal shock
= freestream static conditions
Introduction
OR many years, the Langley Research Center has been involved
in the study of Earth-to-orbit space transportation concepts that
fultill a variety of anticipated mission needs. Some examples are pro-
grams such as the Assured Crew Return Vehicle (ACRV), a vehicle
Presented as Paper 91 -1436 at the AIAA 26th Thermophysics Conference,
Honolulu, HI, June 24-27, 1991; received June 6, 1993; revision received
March 16, 1994; accepled Ior publication March 16, 1994. Copyright ©
1990 by the American Institute of Aeronautics and Astronautics, Ine, No
copyright is asserted in the United States under Title 17, U.S. Code. The
U.S. Government has a royalty-free license to exercise all rights under the
copyright claimed herein for Governmental purposes. All other rights are
reserved by the copyright owner.
*Aerospace Engineer, Aerothermodynamics Branch, Gas Dynamics Di-
vision. Member AIAA.
737
designed to return crew members from Space Station Freedom; the
Personnel Launch System (PLS), a personnel carrier to low Earth or-
bit and return; and Advanced Manned Launch Systems, a candidate
replacement for the current Space Shuttle Orbiter (see for example,
Refs. 1-4). From an aerodynamic/aerothermodynamic perspective.
these programs will benefit from knowledge gained as a result of
the comprehensive data base established for the Space Shuttle pro-
gram. Therefore, it is important that computational, ground-based,
and flight data bases be brought together in an effort to gain an accu-
rate knowledge of flow field phenomena associated with the space
transportation system. The impetus for this work is to provide im-
provements to current computational fluid-dynamic techniques and
to ground-to-llight extrapolation techniques that would be applied
to the next space transportation system.
The present study augments the well-established comprehensive
aerothermodynamic data base for the Shuttle Orbiter by providing
additional information concerning the complex three-dimensional
windward flowfield for an orbiter-like configuration. Areas of inter-
est include transition from laminar to turbulent boundary-layer
heating phenomena and shock-shock interaction phenomena on
windward-surface heating distributions. To date, most comparisons
of aerothermodynamic results obtained via ground-based measure-
ments, predictions, and orbiter flight data have focused on the wind-
ward symmetry plane; similar comparisons for systematic spanwise
heating are quite scarce. For the present study, detailed spanwise
heat-transfer distributions were measured over the windward sur-
face of a winged lifting entry configuration referred to herein as a
modified orbiter. The objective of the present study is to compare
predicted heating distributions from computational fluid-dynamics
(CFD) codes with these detailed spanwise heating measurements.
A set of high-fidelity modified orbiter models using the thin-film
resistance gauge technique have been designed, fabricated, instru-
mented, and tested in two wind tunnels of the Langley Hypersonic
Facilities Complex to obtain detailed aerothermodynamic data over
a wide range of test conditions. The discrete heat-transfer measure-
ments obtained using thin-film gauges are augmented by global ther-
mal mappings and qualitative heat-transfer mappings obtained with
the recently developed relative-intensity, two-color thermographic
phosphor technique. These measurement techniques nicely cap-
ture the previously documented streak-heating phenomena resulting
from the upstream effects of bow-shock-wing-shock interaction.
Experimental Method
Facilities
The Langley 20-Inch Mach 6 Tunnel is a blowdown wind tunnel
that uses dry air as the test gas. The air is heated to a maximum
738 MICOI.
temperatureof1088Rbyanelectricalresistanceheater,andthe
maximumreservoirpressureis525psia.Afixed-geometry,two-
dimensional,contourednozzlewithparallelsidewallsexpandsthe
flowtoMach6 atthe20-in.-squaretestsection.Thistunnelis
equippedwithabottom-mountedmodelinjection-retractionsystem
capableofinjectingheat-transfermodelsfromashelteredposition
tothenozzlecenterlineinlessthan0.6s.The run time lbr this
facility varies from 2 to 10 min. A description of this facility and
calibration results are presented in Ref. 5.
The Langley 31-Inch Mach 10 Tunnel, formerly known as the
Langley Continuous-Flow Hypersonic Tunnel, _ is a blowdown faci-
lity having a run time of approximately 60 s. The facility uses
a water-cooled, three-dimensional contoured nozzle to generate a
nominal Mach number of 10 at the 31-in.-square test section. Dry
air is used as the test gas and is heated to a maximum reservoir stag-
nation temperature of 1900_R. The maximum reservoir pressure is
1450 psia. The tunnel is equipped with a side-wall-mounted model
injection-retraction system. Models are sheltered in the injection
chamber on the side of the tunnel until the tunnel flow is started.
After establishing Mach 10 flow at the test section, heat-transfer
models are injected into the highly uniform stream within 0.5 s.
Models
Heat-transfer distributions were measured on the windward sur-
face of a 0.0075-scale (9.09-in. length) model of a Shuttle-like
PLAN VIEW _A
_ubstfate ..... __
Z Thin film
Steel upper Body _ gage \
_c -'_ A
Substrates • or lower body
Section A-A
Fig. 2 Sketch ofspanwise heating model.
1290 in. !
Section A-A _ A
a) Glass-ceramic model
b) Stainless-steel model with glass-ceramic inserts
Fig. 1 Photograph of modified orbiter.
a) Complete geometry b) Modified geometry
Fig. 3 Space Shuttle Orbiter geometry. Dashed lines represent fair-
ings to model leeward side in present study.
configuration. A photograph and a sketch of the model are presented
in Figs. 1 and 2, respectively. The model has the same lower shape as
the forward 93% of the Shuttle Orbiter. However, the upper surface
is defined by elliptical cross sections. As shown in Fig. 3, the orbiter
canopy has been faired and the vertical tail and OMS pods omitted.
Two 9.09-in.-length models, referred to herein as the modified
orbiters, were fabricated. Both models had solid steel upper sur-
faces; however the lower surfaces of the two models differed. One
model was machined entirely of steel and then slotted to accept
machinable glass-ceramic substrates (herein referred to as the slot-
ted stainless-steel model), whereas the second model was unique,
being machined entirely from the ceramic material and slotted to
accept the machinable glass-ceramic substrates (herein referred to
as the slotted ceramic model).
Both models were cut on a numerical milling machine using a
tape generated with the geometry program described in Ref. 7. Since
this program was also used to generate the geometry in the CFD
codes (to be discussed subsequently), differences between the ex-
perimental and computational models were within the machining
tolerance of ±0.003 in.
Test Conditions
For this study, the pitot pressure could not be conveniently mea-
sured when the model was positioned in the test-section of the
31-Inch Math 10 Tunnel. Thus, test-section flow conditions were
MICOL 739
Table1 Nominal reservoir and freestream flow conditions
Pt,I, Tt,I P_, T.x:, Re _,L Ch,ref,
Mzc psi ::'R psi R 10'5 T,,JT,.2 10 5 Btu_s/ft 4
5.82 29 872 0.022 112.2 4.16 0.62 0.276
5.94 127 915 0.086 113.6 17.21 0.59 0.556
5.98 253 911 0.163 I l 1.7 33.28 0.59 0.772
6.01 475 931 0.297 I I3.2 59.79 0.58 1.050
9.75 354 1822 0.009 94.9 4.17 0.30 0.326
)c'L
GAGE #
1-4
5-9
10- 17
18- 26
o. slIIII[]]IlIIl,
0.6 [II[l[III[1Ill
0.7 IIIIIIIIIIII' lll
y/L
Fig. 4 Layout of thin-film instrumentation.
27 - 37
38 - 49
50 - 62
63 - 77
78 - 97
98 - 122
123- 149
based on measured reservoir pressures and temperatures and a recent
unpublished calibration of the facility.
Flow conditions in the 20-Inch Mach 6 Tunnel were determined
from the measured reservoir pressure and temperature and the mea-
sured pitot pressure at the test section.
Nominal reservoir stagnation and corresponding freestream flow
conditions [or the present study are presented in Table 1. The value
of L used to determine Re_.L is 9.09 in.
The stainless-steel and ceramic models were tested over a range of
angles of attack from 0 to 40 deg in 10-deg increments for each flow
condition at Mach 6. Due to time constraints on testing in the 3l-
Inch Mach 10 Tunnel, only the stainless-steel model was tested at
Mach 10; this model was tested at angles of attack equal to 0, 10, 15,
20, and 30 deg. The sideslip angle was zero for all tests. The angle
of attack was measured relative to the model centerline (see Fig. 2).
Instrumentation and Testing Techniques
Thin-Fihn Gauges
Thin-film resistance heat-transfer gauges were used to mea-
sure surface temperature-time histories from which heat-transfer
rates were inferred. The technology of the thin-film gauges re-
mains unchanged from that developed for the Langley Expansion
Tube. s') The model surface contained 149 palladium gauges (0.04
by 0.05 in.), each approximately 1000 A thick. These gauges were
deposited on highly polished substrates that were precision-fitted
to the model. The model substrate thickness was sized to provide a
maximum run time of 1.5 s; that is, the substrate essentially behaves
as a semi-infinite slab for 1.5 s over most of the forebody. An alu-
minum oxide overlayer, approximately 5000 A thick, was deposited
over the sensing elements as a means of increasing the gauge dura-
bility. Detailed discussions of gauge construction, circuitry, and cal-
ibration procedures and the data acquisition system are discussed in
Refs. 8 and 9. The instrumentation layout is presented in Fig. 4.
77wrmographic Phosphor Technique
The relative-intensity two-color thermographic phosphor tech-
nique m t2 is rapidly becoming the most widely used heat-transfer
measurement technique in the Hypersonic Facilities Complex at
Langley.13 The model, generally MACOR, slycast, or ceramic (fused
silica), is coated with thermographic phosphors. When illuminated
with ultraviolet light, electrons are excited and emit visible light dur-
ing their subsequent relaxation to lower energy levels. The proba-
bility that the relaxation and subsequent fluorescence emission, will
occur is temperature-dependent, and the intensity of fluorescence
may be used to determine local temperatures. Phosphor materials
are selected for a two-color emission-band spectrum that matches
front-end filters on conventional color video cameras. A three-chip,
co-site sampling CCD camera, which has true color separation in the
red, green, and blue bands and spatially congruent detection arrays
is used. Early applications of this technique utilized thermographic
phosphors having secondary and primary emission bands at 450 and
520 nm, respectively (i.e., blue-green). Two materials were used, one
with a temperature sensitivity range of 520 to 71OR and the other
with a range of 560 to 810:R. Currently, a mixture is used con-
sisting of both a broadband phosphor and a narrowband rare-earth
phosphor. For this mixture, the green and red camera filters are used,
and a temperature sensitivity range of 480 to 860R is obtained. The
camera response is calibrated versus incident intensity. Measured in-
tensities from two of the camera color outputs are used to form the
intensity ratio from which quantitative temperature information may
be determined using digital processing. Heating rates are calculated
from surface temperature measurements using one-dimensional
semi-infinite solid heat-conduction equations, as is discussed in de-
tail in Refs. 11 and 12. Based on considerations presented in Refs. 12
and 14, the uncertainty in the heating coefficients is -t-15%.
For this study, the heat-transfer model was cast from a high-purity
fused-silica ceramic using an investment slip-casting technique de-
scribed in Refs. I 1 and 15. The slip casting forms a ceramic shell,
which is then heat-treated to 2600°R. A hydraulically setting mag-
nesia ceramic is used to backfill the ceramic shell, thus providing
strength and support to the sting structure. The phosphor mixture is
suspended in a colloidal silica binder and spray-coated on the model
surface. Typical coating thicknesses vary from 0.001 to 0.003 in.
Data Reduction and Uncertainty
The numerical method used to compute values of the heat-transfer
rate q from the output of the thin-film resistance gauges is discussed
in Refs. 9 and 16. A more accurate determination of the thermal
properties of the glass-ceramic material _7 was used to reduce the
data for the present study from those presented in Ref. 9. Sample
time histories for gauges along the windward centerline revealed
that the heat-transfer coefficient Ch was essentially constant (i.e.,
within -t-2%) over the time interval from 0.5 to 1.7 s; that is, the
substrate behaved one-dimensionally. Second-order least-squares
curve fits were applied to time histories of q for 0.5 < t < 1.7
s, and values of q (or Ch) presented herein were obtained from
these curve fits and generally correspond to t = I. 1 s. For gauges
along the planform leading edge, Ch was observed to increase with
time for t > 0.7 s, indicating that gauges in this region of small
surface radius do not behave one-dimensionally, as expected. How-
ever, for 0.5 s < t < 0.7 s, C, was essentially constant with time;
thus values of q (or C_ ) for gauges at the planform leading edge cor-
respond to t = 0.5 s. tteating distributions are presented in terms
of the ratio of heat-transfer coefficients C_,/Ct,.,.r, where Cj,.,,.r cor-
responds to the stagnation-point heat-transfer rate to a sphere with
a radius of 0.2067 in. For this 0.0075-scale model, the nose radius
(i.e., equivalent sphere radius) was determined to be 0.2067 in. This
value was determined by scaling the approximate equivalent sphere
radius for the full-scale Space Shuttle Orbiter nose, which is about