control and stability in gnat aircraft conceptual design … · aircraft conceptual design ......

9/30/05

Control and Stabilityin

Aircraft Conceptual Design

Based on AIAA Paper 93-3968,“Control Authority Assessment in Aircraft Conceptual Design,” byJacob Kay, W. H. Mason, W. Durham, and F. Lutze, Virginia Techand: VPI-Aero-200, November 1993, available on the web as apdf file, see the reference on the class web page.

Aerospace andOcean Engineering

W. H. Mason

Gnat

YF-17

T-45

F-104graphics fromJoe Chambers

9/30/05

The Problem In Conceptual Design

The Flight Controls Guys(if they’re even there, and worse, they may be EEs!):“We need a complete 6 DOF, with an aero math modelfrom -90° to + 90° or else forget it”

The Conceptual Designers:“Just Use the Usual Tail Volume Coefficient”

Exaggerated? —Not That Much!

This class requires a reasonable middle groundbetween these extreme views

9/30/05

What’s a Tail Volume Coefficient?(Hopefully a review)

cHT = LHTSHTcwSW

, cVT = LVT SVTbWSW

See Raymer, pages 123-125, typical values: cHT: 0.5 to 1, cVT: 0.04 - 0.09

LHT

quarter chord

CL

cW

SW - wing areaSHT - tail area

9/30/05

What you need to know and do• Control and Stability are distinctly different• You have to develop a policy for each axis:

- stable or unstable? Why?• You have to decide how you want to control the vehicle

• including the control system concept design • You have to establish the criteria to determine

the amount of control needed• You have to have an assessment plan:

-How do you know you have adequate control power?

The story for each design is different, there are no universal cookbook answers

9/30/05

To do this you need

• stability derivatives• control derivatives• weight and mass properties

- the cg range- the moments of inertia

• flight envelope- where are the critical conditions?

9/30/05

Some Guidance

• FAR Part 25 (commercial) and Part 23 (general aviation)– tell you where and under what conditions you have to

demonstrate adequate trim and control

• MIL STD 1797 (replacing the MIL SPEC 8785)– provides quantitative guidance for handling qualities

requirements

• Some control requirements are performance based– rotation at takeoff (trim for seaplanes)

• We have some programs to estimate some control andstability derivatives, and a spreadsheet to assess

9/30/05

Typical Conceptual & PD Considerations IEquilibrium/Performance Considerations

Normal Trimmed Flight:• Classical 1G trim• Longitudinal Maneuvering Flight• Steady Sideslip• Engine-Out Trim• Crosswind landing

9/30/05

Typical Conceptual & PD Considerations IIDynamic Considerations

• Takeoff and Landing Rotation• Time-to-Bank• Inertia Coupling

- Pitch Due to Roll and Yaw Due to Loaded Roll• Coordinated Velocity Axis Roll • Short Period and CAP Requirements• High Angle-of-Attack/Departure

9/30/05

Typical Conceptual & PD Considerations IIIOther Considerations:

• Gust• Non-linear Aerodynamics

- High angle of attack• Aeroelasticity• Control Allocation for multiple controls• Special Requirements: weapons separation,

stealth, etc.

9/30/05

Control Authority Assessment Sequence

Design Concept:• Geometry• Mass Properties

AerodynamicsDATCOMComputational Aero(i.e., vortex lattice)

Control Power Evaluation

Requirements:Pass/Fail and why?

Flight Conditions:weight, cg locationspeed, altitudethrust, load factor

9/30/05

Some PC Tools

• simple flight condition definitions tool• a vlm code: JKayVLM

– longitudinal (& poor lat/dir) S & C derivatives

• three surface & two surf + thrust vector trim code• a first-cut spreadsheet evaluation of control power.• New: Drela’s AVL Extended VLM code

9/30/05

JKayVLMthe Vortex Lattice Method

• follows Katz and Plotkin: vortex rings• includes ground effects• define longitudinal & lateral surface separately

- lateral is very crude approximation• define config as a collection of panels, each with a

constant % chord LE & TE device• puts 40-50 rings on each panel• let code step using finite differences to estimate both

stability and control derivatives.

9/30/05

Single panel:

Note: panel does not have to be in a coordinate plane

Point 1

Point 4

Hinge Line

slat

flap

Point 2

Point 3

%c from LE

%c from TE

9/30/05

Combine Panel to Model Plane

23

1

4

Section No.Wing1 2Wing2 3Tail1 4Tail2 4

Longitudinal Model(Top View)

Y

X X

Z

Two Tails (m Y Dir.)Total of Four Sections

Lateral/Directional Model(Side View)

9/30/05

JKayVLM Validation

The following bar charts show the predictionsof JKayVLM with DATCOM and actual valuesfor an F-18 type configuration

9/30/05

VLM Code Accuracy: α Derivatives

0.00

2.00

4.00

6.00

Mach .2 Mach .6

CLαper rad.

Datalift curve slope

VLM DATCOM

-0.40

-0.20

0.00

0.20

0.40

Mach .2 Mach .6

Cmαper rad.

pitching moment slopeData VLM DATCOM

0.0

2.0

4.0

6.0

8.0

Mach .2 Mach .6

staticmargin

static margin, % mean chordData VLM DATCOM

9/30/05

VLM Code Accuracy: Pitch Rate Derivatives

-10.00

-5.00

0.00

5.00

10.00

Mach .2 Mach .6

Cm q

Pitch DampingData VLM DATCOM

0.00

2.00

4.00

6.00

8.00

10.00

Mach .2 Mach .6

CL q

Data VLMLift Due to Pitch Rate

DATCOM

9/30/05

VLM Code Accuracy: β Derivatives

-0.10-0.08-0.06-0.04-0.020.000.02

Mach .2

Cl β

Mach .6

Datarolling moment due to sideslip

VLM DATCOM

0.00

0.05

0.10

0.15

Mach .2 Mach .6

Cn β

Datayawing moment due to sideslip

VLM DATCOM

-1.50

-1.00

-0.50

0.00

0.50

Mach .2 Mach .6

CY β

side force due to sideslipData VLM DATCOM

9/30/05

VLM Code Accuracy: Roll Rate Derivatives

-0.15

-0.10

-0.05

0.00

0.05

Mach .2 Mach .6

Data

Cn p

VLM DATCOM

Yaw Due to Roll Rate

-0.50-0.40-0.30-0.20-0.100.000.10

Mach .2

Cl p

Mach .6

DataRoll DampingVLM DATCOM

9/30/05

VLM Code Accuracy: Yaw Rate Derivatives

-0.40

-0.20

0.00

0.20

Mach .2 Mach .6

Cn r

Datayaw damping

VLM DATCOM

0.00

0.02

0.04

0.06

0.08

Mach .2 Mach .6

Cl r

roll due to yaw rateData VLM DATCOM

0.00

0.20

0.40

0.60

Mach .2 Mach .6

CY r

Dataside force due to yaw rate

VLM DATCOM

9/30/05

VLM Code Accuracy: Elevator Effectiveness

0.00

0.05

0.10

0.15

Mach .2 Mach .6

Cl δ e

Dataroll due to differential elevator

VLM

0.000.200.400.600.801.00

Mach .2 Mach .6

CL δ e

Data VLMlift due to elevator

DATCOM

-1.50

-1.00

-0.50

0.00

0.50

Mach .2 Mach .6

Cm δ e

pitching moment due to elevatorData VLM DATCOM

9/30/05

VLM Code Accuracy: Flap Effectiveness

0.00

0.50

1.00

1.50

2.00

Mach .2 Mach .6

CL δ flp

flap effect on liftData VLM DATCOM

0.000.050.100.150.200.25

Mach .2 Mach .6

Cm δ flp

Data VLM

flap effect on pitching moment

0.00

0.05

0.10

0.15

0.20

Mach .2 Mach .6

Data

Cl δ flp

VLMrolling moment due to differential flap

9/30/05

VLM Code Accuracy: Aileron Effectiveness

0.00

0.05

0.10

0.15

0.20

Mach .2 Mach .6

Data

Cl δ ail

VLM DATCOM

9/30/05

VLM Code Accuracy: Rudder Effectiveness

0.00

0.05

0.10

0.15

0.20

Mach .2 Mach .6

CY δ r

Data VLM DATCOMside force due to rudder

0.000

0.005

0.010

0.015

0.020

Mach .2 Mach .6

Data

Cl δ r

VLM DATCOMrolling moment due to rudder

-0.05-0.04-0.03-0.02-0.010.000.01

Mach .2

Cn δ r

Mach .6

Data VLM DATCOMyawing moment due to rudder

9/30/05

A Similar Evaluation of AVL Required

• Can’t trust that you know how to use the code• Can’t understand code limitations☛ Until you do a complete evaluation as shown

above for JKayVLM

http://raphael.mit.edu/avl/

9/30/05

Aircraft Assessment SpreadsheetFor several typical flight situations, a spreadsheet containing11 different cases has been put together. The spreadsheetactually computes the required control deflection or timerequired to do the maneuver.To use it you need to enter:

• the flight conditions and the mass properties, both atheavily loaded and lightly loaded conditions, the fullrange of cg locations and the inertias

• the stability and control derivatives- corresponding to the each cg position

To do the assessment:• check that the required control deflection is acceptable• check that the time required meets the requirement

9/30/05

EXCEL Spreadsheets, 11 Worksheets:

1. Nose-wheel Lift-off2. Nose-down Rotation During Landing Rollout3. Trimmed 1-G Flight4. Maneuvering Flight (Pull-up)5. Short Period & Control Anticipation Parameter (CAP)6. Pitch Due to Roll Inertial Coupling7. Time-to-Bank Performance8. Steady Sideslip Flights (Aileron & Rudder Deflections)9. Engine-out Trim (Aileron & Rudder Deflections)10. Roll Pullout11. Initiate & Maintain Coordinated Velocity Axis Roll

9/30/05

Just One Simple Example:***********************************************************************Trimmed 1-G Flight***********************************************************************Input: Weight (lbs) 51900

Reference Area (ft^2) 400Speed (ft/s) 400Air Density (slug/ft^3) 0.002376C-m-0 0.0181C-m-delta E (/rad) -1.117C-L-0 -0.0685C-L-delta E (/rad) 0.8688C-m / C-L (-Static Margin) -0.1C-L-alpha (/rad) 4

Output: C-L Required for 1-g trim 0.6826073Elevator Deflection for Trim (deg) -4.53912205AOA Required for 1-g Trim (deg) 11.744717

9/30/05

Tested Against Existing Airplane

• methodology applied to a known airplane• results generally good

9/30/05

Review: The Pitching Moment Trim Story

9/30/05

Sizing Control Surfaces: the X-plot A rational basis for H-Tail sizing:

simplified for illustration: critical aft limit depends on design

TailSize

cg location, usually given as %macaftfwd

nose uplimit

nose down limit(unstable a/c aft cg limit)

tip back limit

requiredcg range

static stability limit(stable a/c aft cg limit)

MinimumTail Size

bigger

9/30/05

High Angle of Attack

• Aerodynamics are nonlinear- complicated component interactions (vortices)- depends heavily on WT data for analysis

• Motion is highly dynamic• Exact requirements are still being developed• Keys issues:

- adequate nose down pitching moment to recover- roll rate at high alpha- departure avoidance- adequate yaw control power- the role of thrust vectoring

9/30/05

The Hi-α StoryLongitudinal

Typical unstable modern fighter

α 90°

+

-

Cm

Max nose up moment

Max nose down moment

Cm*

Minimum Cm* allowable is an open questionIssue of including credit for thrust vectoring

Pinch often around α ~ 30°-40°

0°

9/30/05

Typical Swept Wing Cm Characteristics

-0.06

-0.04

-0.02

0.00

0.02

0.04

0.06

0.08

-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2

Cm

CL

AR = 10, Λc/4 = 35°, λ = 0.5

data from NACA RM A50K27

Re = 10 million

x ref = c/4

VLMpc calculation

9/30/05

Real World PitchupDC-9 F-16

Shevell and Schaufele,, Journal of Aircraft,Vol. 3, No. 6, pp. 515-523 (1966)

W. H. Mason, “Stability and Control in Computational Simulations for Conceptual andPreliminary Design of Aircraft: the past, today, and future?,” NASA/CP-2004-213028/PT1, April 2004, pp. 309-340

Max NoseDown

Max NoseUp

9/30/05

Design Chart to Avoid Pitchup:A starting point

0.0

2.0

4.0

6.0

8.0

10.0

0° 10° 20° 30° 40° 50° 60°

historical trends from early wind tunnel data

AspectRatio

Probably OK

Probably Pitchup Prone

NASA TM X-26

Quarter Chord Sweep

Fighters Transports

Note: DATCOM has a more detailed chart

Pitching moment characteristics as separation occurs must becontrollable. Requires careful aero design.

Horizontal tail location is critical

9/30/05

The Hi-α StoryDirectional

+ Stable

α 90°

- Unstable

00°

w/ Vertical Tail

ForebodyEffect

Directional problem oftenaround 30° for fighters

Config w/o V-Tail

V-Tail in wakeCnβ

9/30/05

The Hi-α StoryLateral

Clβ

+

-

α 90°0°“Stable”

“Unstable”

0

Dihedral Effect(also due to sweep) Flow Separates

on wing

Drooping LE deviceshelps controlseverity

9/30/05

Comment on Thrust Vectoring

• Thrust vectoring mainly provides moments at high thrust- a problem if you don’t want lots of thrust!

• Thrust vectoring moment is near constant with speed• Ratio of aero to propulsive moments

- propulsion dominates at low q- aero dominates at high q

9/30/05

Special Issues for Supersonic Flightand Related Planforms

• Pitchup– See Alex Benoliel’s Thesis for a survey and estimation method

• Aerodynamic Center Shift– See Paul Crisafulli’s Thesis. One chapter addresses ac shift

Both available at Mason’s web site under: Thesis/Dissertation Titles and Placement.

http://www.aoe.vt.edu/aoe/faculty/Mason_f/MRthesis.html

Alex’s Thesis is also under “design related reports” on our web site

9/30/05

Conclusion

Aerospace andOcean Engineering

• We’ve outlined the issues, typical criteria and procedures• We’ve established some useful tools.• Each project is different in the details• Each individual controls person has to develop the detailed

approach for their particular design

F-15

EA-6B

control and stability in gnat aircraft conceptual design … · aircraft conceptual design ......

Documents