Validating LES predictions of the flow in a swirl-stabilized plasma torch geometry
ORAL
Abstract
Inductively coupled plasma torches have a variety of industrial and
academic uses, including as a plasma source for testing materials for
ablative thermal protective systems. However, the performance of
these devices can be sensitive to the driving flow. In this work, we
report on an effort to characterize the flow in a swirl-stabilized
plasma torch geometry with large eddy simulation and to validate the
simulations against experiments. The flow is fed by small,
compressible jets that induce a strongly swirling flow. This flow is
highly turbulent, but shortly above these entrance jets, the flow
becomes essentially incompressible and the initial jet turbulence
decays rapidly. Because of the large range of both length and
time scales induced by the geometry, the problem is split into two
subproblems to enable efficient simulation and thus, two sets of large
eddy simulations are conducted. The first simulations characterize
the compressible entrance jets and generate boundary condition data
for the second, which simulate the flow from just above the entrance
jets through the exit of the torch. Uncertainty in the inlet
conditions for the second simulations are propagated, and validation
comparisons against PIV measurements are shown.
academic uses, including as a plasma source for testing materials for
ablative thermal protective systems. However, the performance of
these devices can be sensitive to the driving flow. In this work, we
report on an effort to characterize the flow in a swirl-stabilized
plasma torch geometry with large eddy simulation and to validate the
simulations against experiments. The flow is fed by small,
compressible jets that induce a strongly swirling flow. This flow is
highly turbulent, but shortly above these entrance jets, the flow
becomes essentially incompressible and the initial jet turbulence
decays rapidly. Because of the large range of both length and
time scales induced by the geometry, the problem is split into two
subproblems to enable efficient simulation and thus, two sets of large
eddy simulations are conducted. The first simulations characterize
the compressible entrance jets and generate boundary condition data
for the second, which simulate the flow from just above the entrance
jets through the exit of the torch. Uncertainty in the inlet
conditions for the second simulations are propagated, and validation
comparisons against PIV measurements are shown.
–
Presenters
-
Todd Oliver
University of Texas at Austin, Oden Institute for Computational Engineering and Sciences
Authors
-
Todd Oliver
University of Texas at Austin, Oden Institute for Computational Engineering and Sciences
-
Sigfried W Haering
University of Texas at Austin
-
Dillon Ellender
UT Austin, University of Texas at Austin
-
Dan Fries
University of Texas at Austin
-
Noel T Clemens
The University of Texas at Austin, University of Texas at Austin
-
Robert D Moser
University of Texas at Austin