This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Design of a “Hinged” Delta
Wing to Control Vortex Flow
Wing
By: Jake Mitchell
12/13/03 I. Introduction
I. Delta wings have
Delta
leadingedge vortices
over the surface
over It may be desirable to
It
control the size and
strength of these
vortices
vortices I. Introduction
I. Changing the size and
Changing
strength of a vortex may
change how it breaks
down
down Early vortex breakdown
Early
can be desirable because
an aircraft can follow in its
wake more closely
wake Early vortex breakdown
Early
can also be undesirable
because if it breaks down
over the wing, it can
decrease lift
decrease I. Introduction
I. Some devices used to control vortex breakdown:
– Mechanical Devices: strakes, canards, fillets, leadingedge extensions, flaps, vortex fences
– Pneumatic devices: suction applied downstream of
Pneumatic
vortex breakdown stabilizes vortex; blowing also shifts
the vortex breakdown location downstream
the Delaying vortex breakdown causes aircraft to be
Delaying
able to perform at higher angles of attack
able II. Proposed Design
II. Does creating a ‘hinge’ near the vortex line affect the
Does
vortex size, strength, or breakdown?
vortex Does the hinged part affect wing performance? Bend outboard section up and down, and see what the
Bend
overall affect is
overall II. Proposed Design
II.
• Modeling an actual wing with a hinge proved
Modeling
to be very difficult.
to
• An approximation was used: very simplified
An
delta wing design
delta
• The overall effects on the simplified delta
The
wing will likely happen on a normal delta
wing
wing II. Proposed Design
II. Model was created in
Model
Gambit by defining
vertices: a ‘triangular’
airfoil was used
airfoil A tapered triangular mesh
tapered
was used
was II. Proposed Design
II. To capture vortex flow
To
across surface of wing,
a nonlinear code must
be used
be In this design, Fluent
In
6.0 was used. Fluent
solves the full nonsolves
llinear NavierStokes
inear
equations
equations III. Solution
III. Solution methods:
– Pressure: standard algorithm
– Pressurevelocity coupling: SIMPLE algorithm
Pressurevelocity
(3rd order accurate)
(3
– Momentum: 2nd order upwind
– Modified turbulent viscosity: 2nd order upwind 1st order upwind is too diffusive for an
accurate answer
accurate Most grids had ~117,000 cells in them III. Solution
III. Turbulence model: SpalartAllmaras
– Oneequation model
– An equation is derived for the eddy viscosity
An
from scratch and then taylored for aerodynamic
flows
flows
– SA model is diffusive, but it is better (and
SA
quicker) than kε model
quicker)
model III. Solution
III. Geometry settings:
–
–
–
– Root chord: 15 meters
30degree sweep
Hinge line is at 20 degrees of sweep
Thickness at chord trailing edge: 1 meter Velocity: 150 m/s at 15 degrees angle of attack Standard atmosphere temperature and pressure III. Solution
III. Base wing with no hinge (to compare others with) Solution was allowed to converge for a first order accurate solution;
Solution
then was continued until convergence with a secondorder solution
then Residuals show how quickly solutions converged Baseline converged after ~210 iterations total All wings have same geometry except for the hinge angle III. Solution
III. 15 degree case Converged after ~290
Converged
iterations
iterations III. Solution
III. 30 degree case Did not converge even after
Did
almost 1,700 iterations
almost Came close enough that we can
Came
assume it is fairly accurate
assume III. Solution
III. +15 degree case Converged after ~245
Converged
iterations
iterations III. Solution
III. +30 degrees case Solution converged
Solution
after ~215 iterations
after III. Solution
III. Problem: base wing
Problem:
produces negative lift!
produces Contours of pressure
Contours
confirm this
confirm Pressure is negative on
Pressure
lower surface, and positive
on upper surface
on Shown is a case run at 30
Shown
degrees angle of attack,
with the base wing
with IV. Analysis
IV. Graphs of pressure
Graphs
coefficient on wing section
also confirm that we have
downforce instead of lift Shown are 30 deg, 0 deg,
Shown
and + 30 deg; the pressure
distribution ‘closes’ as the
angle increases
angle Using an actual airfoil
Using
section would probably
have prevented this
have IV. Analysis
IV. Baseline case is shown above
First image is vortex at 10 meters aft of the leading edge
Second image is vortex at 10 meters aft of the trailing edge, which can
Second
be considered the ‘near wake’
be Images are contours of relative zvelocity, or velocity in the spanwise
Images
direction
direction All zvelocity graphs have the same range of values for comparison IV. Analysis
IV. Wing with 30 degree hinge
Vortices are very large on the wing
In the wake, vortices are taking a long time to
In
break down
break IV. Analysis
IV. Wing with 15 degree hinge Vortices decrease on the wing as the angle is
Vortices
increased
increased In the wake, the vortex is breaking down quicker IV. Analysis
IV. Baseline (0 degree) case again (for visualization
Baseline
continuity purposes)
continuity IV. Analysis
IV. Wing with +15 degree hinge
Vortex loses strength and size on surface of wing
Vortex breaks down rapidly behind wing IV. Analysis
IV. Wing with +30 degree hinge
Flow over the wing surface is complex
In the wake, vortices are lasting longer IV. Analysis
IV.
70000 60000 50000 Drag Force [n] 40000 30000 20000 10000 0
40 30 20 10 0 10 20 30 40 10 20 30 40 Tip Angle [deg] Lift vs. Tip Angle
150000 100000 50000 0
40
Lift Force [n] What about the forces
What
on the wing?
on Minimum drag is when
Minimum
the angle is zero
the Maximum lift is
Maximum
obtained when the
angle is the highest
angle We get positive lift
We
when the angle is
sufficiently high
sufficiently Drag vs. Tip Angle 30 20 10 0 50000 100000 150000 200000 250000
Tip Angle [deg] IV. Analysis
IV. Lift over Drag Ratio
Lift
also increases as the
hinge angle increases
hinge L/D for different tip angles
2 1 0 L/D 40 30 20 10 0 1 2 3 4
Tip Angle [deg] 10 20 30 40 IV. Analysis
IV. The cause for the
The
increase in
performance: the
hinged part of the wing
gets loaded up more
gets This may be bad from
This
a structural point of
view
view IV. Analysis
IV. Possible sources of error
– Grid may not be refined enough – possible to do a grid
Grid
refinement study
refinement
– Turbulence modeling used – only way to get around this
Turbulence
is to do a Direct Numerical Simulation (DNS), which is
extremely costly. Possible to use a more accurate
turbulence model, however.
turbulence
– Convergence criterion set relatively high, to allow for
Convergence
quick convergence. For a full design study, the
convergence criterion would be much lower.
convergence V. Conclusion
V. Decreasing the hinge angle causes poorer
Decreasing
performance and stronger vortices in the
wake
wake Increasing the hinge angle increases
Increasing
performance and causes the vortices in the
wake to break down quicker, to a point.
wake The hinged portion of the wing receives high
The
aerodynamic loads, which can destroy the
hinges if not taken into consideration
hinges V. Conclusion
V. Possible uses for a hinged delta wing:
– Allow the wake behind an aircraft to diminish
Allow
more quickly so that other aircraft can take off
after it or fly in formation behind it
after
– Increase drag while braking
– Increase L/D while taking off
– By hinging one wing, a rolling moment is
By
created
created ...
View
Full
Document
This note was uploaded on 01/15/2012 for the course AAE 490 taught by Professor Andrisani during the Fall '09 term at Purdue UniversityWest Lafayette.
 Fall '09
 ANDRISANI

Click to edit the document details