NASA Technical Reports Server (NTRS) 19930021674: A comparison of classical mechanics models and finite element simulation of elastically tailored wing boxes

8 Pages·0.14 MB·English

by  NASA Technical Reports Server (NTRS)

N93-30863
A COMPARISON OF CLASSICAL MECHANICS MODELS
AND FINITE ELEMENT SIMULATION
/,/
OF ELASTICALLY TAILORED WING BOXES
/
• /
Lawrence W. Rehfield, Richard D. Pickings,
Stephen Chang and Michael Holl
University of California, Davis
SUMMARY
Structural tailoring concepts have been developed by Rehfield to create wings with elastically produced
camber for the purpose of increasing lift during take-off conditions.
Simple models based upon enhancements to the thin-walled composite beam theory of Rehfield have been
developed to investigate prospects for elastic tailoring of the chordwise deformation of wing structures.
Previous correlation studies have shown that the thin-wall beam theory is very accurate for predicting
bending and twisting deformation of thin-wall composite box beams, structural tubes, and rotor blades
with both single and multiple cells. However, earlier work has not emphasized chordwise deformations.
Application of the modified theory to real-world problems must be executed with confidence and minimum
risk. Consequently, it is imperative that the theory is of high enough fidelity to accurately predict not only
the intentionally produced tailoring mechanisms, but also the actual overall behavior of wing box
structures. Therefore, extensive correlation studies must be performed to establish that the theory is valid
for predicting chordwise deformations.
Validation will be carded out in two steps: finite element correlation and experimental correlation.
Extensive testing using scale models of wing box structures is planned. However, the purpose of present
work is to provide a comparison of the theoretical results with a finite element model for the bending
method of producing camber.
At present, finite element correlation studies have been completed for two cases: a bonded unstiffened
structural box, and a bolted unstiffened structural box. Results from these studies show an error of less
than one percent for the bonded case and less that six percent for the bolted case in predicting camber
curvature for the structural box. Examination of the results shows that the theory is very accurate for the
cases studied and therefore, will provide an excellent basis for conducting further tailoring studies.
809
ELASTIC TAILORING
THE USE OF
• STRUCTURAL CONCEPT
• FIBER ORIENTATION
• STACKING SEQUENCE
• BLEND OF MATERIALS
TO ACHIEVE SPECIFIC PERFORMANCE GOALS
CRUISE
MAX. LIFT
ELASTICALLY TAILORED
METHODS OF INCREASING AIRFOIL LIFT
810
C - 130 CENTER WING BOX MODEL
80
"J _ 66 D.[
T
30
A_
192
................"..'-........0...5..0..0.." 't.000" - - 1
o.o5o" 'rtlK (TYI'). ,--
I / 1
8 PI,Y .044" TItICK COMPOSITE A /
PANEL BONDED 3'0 ALUMINUM CtlANNELS. /
J
PRISMATIC BEAM WITH LENGTIt = 20.00"
STRUCTURAL BOX CROSS SECTION
811
4.000"
-- 0.500"
0.050" THK (TYP).---'-
0"
8 PLY, 0.044" THK COMPOSITE LAMINATE
PRISMATIC BEAM I_ITH LENGTH = 20.00"
}_'ITH #10 BOLTS @ 1" SPACING
CROSS SECTION OF BOLTED STRUCTURAL BOX
MATERIAL PROPERTIES
AS4/3501-6 GRAPHITE EPOXY
Ell = 20.0 MSI
E 22 -- 1.7 MSI
G12 = 0.85 MSI
V 12 = 0.3
ALUMINUM CHANNEL
• E = 10.0 MSI
• V = 0.33
812
DEFORMED SHAPE OF STRUCTURAL BOX
ESTABLISHING A TECHNOLOGY BASE
GOAL: UTILIZATION OF TECHNOLOGY WITH
CONFIDENCE AND LIMITED RISK
• PROOF OF CONCEPT
• DESIGN ANALYSIS METHODOLOGY
• FIDELITY OFMODELING
- FINITE ELEMENT CORRELATION
- EXPERIMENTAL CORRELATION
• CAUSE & EFFECT RELATIONSHIPS
813
PLANES OF SYMMETRY
ANSYS STIF 43
SIIELL ELEMENTS
UPPER & LOWER
ALUMINUM CIIANNEL Z LAMINATE COVERS
Y
FINITE ELEMENT MODEl.
RESULTS OF CORRELATION STUDY
CHORDWISE CAMBER CURVATURE (IN -1)
MODEL TYPE BOLTED BONDED
CLASSICAL 0.01040 0.00954
FINITE ELEMENT 0.00982 0.00947
PERCENT DIFFERENCE -5.6 -0.7
814
CONCLUSIONS
• CORRELATION EXCELLENT
• CLASSICAL MODEL MAY BE USED
FOR DESIGN
815