ThreeDimensional Analysis of PowerTakeOff Regions

1
AIAA-2008-4330
Three-Dimensional Analysis of Power-Take-Off
Regions of Experimental Scramjet Driven MHD
Generator
T. Takahashi, T. Fujino and M.
Ishikawa University of Tsukuba, JAPAN and J. T.
Lineberry LyTec LLC
2
Back Ground of Research
One of concept of scramjet driven MHD power
generation
Concept of scramjet driven magnetohydrodynamics
(MHD) power generation has been proposed.

• Hypersonic Vehicle Electric Power System
  (HVEPS) program
• was started as a five-year research and
  development plan.

• In 2006, HVEPS scramjet MHD power demonstration
  tests
• were successfully concluded by Lytec LLC.

3
Experimental Scramjet Driven DCW-MHD Generator
View of experimental scramjet-driven DCW-MHD
generator ( DCW Diagonal Conducting Wall )
4
Result of Power Demonstration Test
Distribution of electrode potential obtained by
one of power demonstration tests (test no. 3)
5
Objectives
To examine generator performance of experimental
scramjet driven DCW-MHD generator by mean of
three-dimensional numerical analysis
1)
To propose optimum electrode configuration in
power-take-off regions from viewpoint of
maximizing electrical power output by mean of
three-dimensional numerical analysis
2)
6
Basic Equations of Gasdynamics
Mass conservation equation
Momentum conservation equations
7
Basic Equations of Gasdynamics
Total energy conservation equation
For convection terms
Harten-Yee implicit TVD scheme
For viscous terms
Second order central difference scheme
Algebraic turbulence model proposed by Cebeci and
Smith and improved by Stock and Haase is adopted.
8
Basic Equations of Electrodynamics
Steady Maxwell equations
Generalized Ohms law
Second-order elliptic partial differential
equation on electric potential
Galerkin finite element method (FEM) with first
order tetrahedron elements
9
Analytic Conditions
Numerical condition
Generator channel shape in x-y plane and x-z plane
10
Experimental Electrode Configuration
3
3
Electrode potentials are calculated so that
conservation of electric current is satisfied.
Electrode potentials are given to equipotential
values.
Distribution of fixed current value
11
Thermodynamical Properties
Dependency of electrical conductivity on the
temperature and the pressure
Minimum value is fixed to be 0.1 S/m when
electrical conductivity becomes less than 0.1 S/m
as approximate arcing model.
Electrical conductivity is reduced to consider
the nonuniformity of plasma.
electrical conductivity deficit factor
?0.8
12
Magnetic Flux Density
z-component of magnetic flux density (measured)

x-component of magnetic flux density (calculated)
Magnetic field was applied to generator channel
by split-coil superconducting magnet that NASA
provided.
We assumed that magnetic flux density has x- and
z-component, which varies in all directions.
13
Numerical Results
Behavior of plasma flow
Comparison of experiment and numerical results
Effects of x-component of Magnetic Flux Density
Proposal for Improvement of Generator Performance
14
Distribution of Static Pressure
Compression and expansion waves arise turning of
duct upper wall.
Distribution of static pressure on x-y plane with
z 0
Small influence of MHD interaction on flow field
is anticipated because MHD interaction is weak in
this MHD power generator.
Distribution of static pressure along center
line (with/without MHD)
15
Current Density and Current Flow
Distribution of current density and current flow
on x-y plane with z 0 (Lines illustrate a
projection of current on x-y plane)
Eddy currents are induced in both power-take-off
regions.
Rather large axial component of current is
induced in active generator region.
Loading condition is not optimized.
16
Numerical Results
Behavior of plasma flow
Comparison of experiment and numerical results
Effects of x-component of Magnetic Flux Density
Proposal for Improvement of Generator Performance
17
Electrode Potential with Experimental Electrode
Configuration
Distribution of electrode potential obtained by
numerical analysis and power demonstration
experiment
Discrepancy between numerical analysis and
experimental result is caused mainly by large
temperature nonuniformities probably induced in
scramjet combustion.
18
Electrode Potential with Experimental Electrode
Configuration
Numerical results approximately agree with
experimental results.
Distribution of electrode potential obtained by
numerical analysis and power demonstration
experiment
Discrepancy between numerical analysis and
experimental result is caused mainly by large
temperature nonuniformities probably induced in
scramjet combustion.
19
Electrode Potential with Experimental Electrode
Configuration
Voltage drop obtained by numerical analysis is
about 67 lower than experimental value.
Distribution of electrode potential obtained by
numerical analysis and power demonstration
experiment
Discrepancy between numerical analysis and
experimental result is caused mainly by large
temperature nonuniformities probably induced in
scramjet combustion.
20
Numerical Results
Behavior of plasma flow
Comparison of experiment and numerical results
Effects of x-component of Magnetic Flux Density
Proposal for Improvement of Generator Performance
21
Effects of x-component of Magnetic Flux Density
Voltage drop with Bx is about 1.08 times larger
than voltage drop without Bx.
Voltage drop with Bx is about 1.11 times larger
than voltage drop without Bx.
Distributions of electrode potential with and
without x-component of magnetic flux density
In this experimental generator, effects of Bx are
rather small.
This effect might become large for larger
generators.
22
Numerical Results
Behavior of plasma flow
Comparison of experiment and numerical results
Effects of x-component of Magnetic Flux Density
Proposal for Improvement of Generator Performance
Magnetic flux density near power-take-off regions
is low.
Electromotive force becomes small.
Number of power-take-off electrodes may have
large effects on generator performance.
23
Optimization of Number of PTO Electrodes
Number of electrodes in outlet power-take-off
region is given to be 3 (fixed)
Number of electrodes in inlet power-take-off
region is given to be 3 (fixed)
Relationships between number of electrodes in
power-take-off regions and load voltage
Optimum number of inlet and outlet power-take-off
electrodes is determined to be five and nine,
respectively.
24
Optimum Electrode Configuration
5
9
Current value is fixed across MHD generator
section included power-take-off regions.
25
Generator Performance
Outlet PTO region
Inlet PTO region
Distributions of electrode potential under
optimum electrode configuration and experimental
electrode configuration
Optimized power-take-off electrode configuration
can improve MHD generator performance with about
30 increase of electric power.
26
Concluding Remarks (1/2)
Present three-dimensional numerical analysis has
examined generator performance of experimental
scramjet driven DCW-MHD generator implemented in
HVEPS project, and also has proposed optimum
electrode configuration in power-take-off regions
from viewpoint of maximizing electrical power
output.
Numerical results approximately agree with
experimental results in active generator region
and outlet power-take-off region.
Discrepancy of voltage drop near inlet
power-take-off region between numerical and
experimental results still exists. There is
possibility that discrepancy is caused by rather
large temperature nonuniformities induced in
scramjet combustion.
27
Concluding Remarks (2/2)
Effect of the x-component of magnetic flux
density is rather small in present generator.
Optimum number of inlet and outlet power-take-off
electrodes is five and nine, respectively,
resulting in improvement of the electric power of
30 .