374
SHANG, HAYES, AND MENART
deposition of the plasma. Another feature of the shock wave and
plasma counterow jet interaction lies in the unusual con guration
of the complex shock waves. The resultant shock wave conforma-
tion at a lower mass ow rate does not follow the trend of the
room-temperature air injection. The thermal diffusion can play a
(
)
signi cant role in this phenomenon see Fig. 8 .
The charge separation across the bow shock leading to a steep
electrical eld gradientisanotherphysicalpossibility.Themeanfree
path of the oncoming stream under the tested condition is 3.35 ¹m
¡6
(
)
10 m , whereas the debye lengthof the plasma is about 11.0 ¹m.
The debye length is, thus, more than a factor of three greater than
the mean free path of the freestream, which is of the same order of
magnitude of the shock wave thickness.As a point of reference, the
electron mean free path is about 0.1 mm at the freestream pressure
of 5 torr. This phenomenon remains as a key element for future
investigation.
VIII. Conclusions
The drag reduction by a plasma counterow jet is investigated
in a Mach 6 wind tunnel at a freestream pressure of 344.7 kPa and
temperature of 79 K. The plasma with a vibrionic temperature of
(
)
.
ref
Fig. 13 Drag data of plasma injection vs m/ m
§
4400 400 K, an electron temperature about 20,000 K, and an es-
room-temperature air injection in a single data sweep. This con-
tinuous data distribution serves as a reference for individual drag
measurements.In the present data collectionprocess,the drag mea-
suring procedure starts with room-temperature air injection. The
plasma is then introduced by igniting the torch, which is sustained
for a durationof 15 s. In a typicaltest, the piezoelectricforce sensors
yield consistentoutput just a fraction of a second after the transient
electromagnetic pulse has passed. The drag of an individual test
by the room-temperature air injection registers a value close to the
data of reference. The maximum difference is generally con ned
within the data scattering band of about 5%. The measured drags
rise when the plasma is ignited. According to the earlier computa-
tional analyses of similar simulations, this behavior is mostly due
to reduced mass injection ow rate. There is little doubt that the
plasma injection has produced a greater drag reduction than the
room-temperatureair injection at the identical mass ow rate.
Identical experimental data are presented in terms of mass ow
rate in Fig. 13. The mass ow rate is obtained by solving the mass-
12
3
£
timated electron density greater than 3 10 /cm is injected from
a hemispherical cylinder. At a given stagnation pressure, the mea-
sureddragis nearly10% higherthanthe room-temperatureairinjec-
tion. However, if the decreased injection mass ow by the elevated
plasma temperature is taken into consideration, the drag reduction
by plasma injection is in fact greater.
Based on the equilibrium chemical composition at the measured
plasma condition and at the identical injecting mass ow rate, the
plasma injectionwould yield a rangefrom 6.1 to 13.4%greaterdrag
reductionby the plasma thermal deposition.From the experimental
data, the effects of chemical nonequilibrium and electromagnetic
force are most likely negligible.
In addition to a greater drag reduction, the plasma injection re-
duces the amplitude of shock wave oscillation associated with the
counterow jet and bow shock interaction. In the injection pres-
sure range lower than the shock bifurcation point, the plasma
injection signi cantly reduces the oscillatory amplitude of drag
measurements by 10 dB.
–
averaged Navier Stokes equations using the the chemically equi-
librium composition. The calculated mass ow rate yields values
of 0.13, 0.19, 0.23, and 0.26 g/s, which correspond to the injecting
Acknowledgments
The computing resource was supported by a grant from the De-
partment of Defense High Performance Computing Shared Re-
P
P
of 1.4, 2.0, 2.5, and 2.8, respec-
0
stagnation pressure ratios
=
j
tively.The additionaldragreductionby the thermaldepositionof the
plasma injectionis clearly demonstrated.The computed magnitude
of the reduced drag is as high as 13.4%.
–
source Center at Wright Patterson Air Force Base. The research
team deeply appreciates the sponsorship by S. Walker of the Air
Force Of ce of Scienti c Research.The invaluablecontributionsby
James M. Williamson,Dean Emmer, and the wind-tunnelcrew, Tom
Norris, Ray Raber, and Michael Greene are duly acknowledged.
In an attempt to further de ne the electromagnetic effect for the
plasma counterow jet, an applied magnetic eld was imposed by
(
)
a set of neodymiumrare earth NeFeB magnets around the plasma
torch chamber. The polarization of the applied magnetic eld is
aligned with the axis of the nozzle to enhance the plasma pinch
effect.21 The magnethas a maximum magnetic ux densityof0.47T
at the pole, but the eld strengthdiminishesrapidly toward the noz-
zle axis. The estimatedvalue is about 0.17 T locally,and the plasma
References
1Shang, J. S., Hayes, J., Wurtzler, K., and Strang, W., “Jet-Spike Bi-
furcation in High-Speed Flows,” AIAA Journal, Vol. 37, No. 6, 2001, pp.
–
1159 1165.
2Shang, J. S., Hayes, J., Harris, S., Umstattd, and Ganguly, B., “Exper-
imental Simulation of Magneto-Aerodynamic Hypersonics,” AIAA Paper
2000-2258, June 2000.
S
B2 R
u
=½ / is much less than unity. Un-
interaction parameter .¾
der this circumstance, the effects of Lorentz force is dif cult to
detect. As anticipated, the only success was achieved at the low-
est plasma stagnationpressure. The differencebetween plasma and
room-temperatureair injectionincludingthe applied magnetic eld
is detectable. The difference in drag measurement with and with-
out an applied magnetic eld is not consistent,but always less than
4%. The plasma also failed to ignite at all higher plasma stagna-
tion pressures.Note that this magnetic eld is applied only near the
electrodes, where the electrical charge separation may take place.
From this observation, one recognizes that the induced magnetic
eld by plasma injectionis most likely negligible.14;15;31 Therefore,
a signi cantelectromagneticeffectby plasmainjectionfor ow eld
modi cation would be highly doubtful.
3Shang, J. S., Ganguly, B., Umstattd, R., Hayes, J., Arman, M., and
Bletzinger, P., “Developing a Facility for Magneto-Aerodynamic Experi-
–
ments,” Journal of Aircraft, Vol. 17, No. 6, 2000, pp. 1065 1072.
4Josyula, E., “Computational Study of Vibrationally Relaxing Gas Past
BluntBodyin HypersonicFlow,” Journalof Thermophysics andHeat Trans-
–
fer, Vol. 14, No. 1, 2000, pp. 18 26.
5Wegner, P. P., and Buzyna, G., “Experiments on ShockStandoffDistance
in Non-EquilibriumFlow,” Journal of Fluid Mechanics, Vol. 32, Pt. 2, 1969,
–
pp. 325 335.
6Adamson, T. C., and Nicholls, J. A., “On the Structureof Jet from Highly
Under Expanded Nozzles into Still Air,” Journal of the Aeronautical Sci-
–
ences, Vol. 26, Jan. 1959, pp. 16 21.
7Warren, C. H. E.,“AnExperimentalInvestigationoftheEffect ofEjecting
a Coolant Gas at the Nose of a Bluff Body,” Journal of Fluid Mechanics,
Vol. 8, 1960, p. 400.
From the present research results at an identical mass ow rate,
the plasma injection indeed produces a greater drag reduction than
its room-temperaturecounterpart.However, the major portionof the
reduced drag of the present experiment is derived from the thermal
8Baron, J. R., and Alzner, E., “An Experimental Investigation of a Two
Layer Inviscid Shock Cap Due to Blunt Body Nose Injection,” Journal of
Fluid Mechanics, Vol. 15, 1963, p. 442.