DAY8
BUCKLING OF AIRCRAFT STRUCTURES
BUCKLING Buckling is a failure mode characterized by a sudden failure of a structural member subjected to high compressive stresses, where the actual compressive stresses at failure are smaller than the ultimate compressive stresses that the material is capable of withstanding Buckling is also described as failure due to elastic instability
BUCKLING TYPES Stable or Gentle Buckling is a buckling in which the displacements increase in a controlled fashion as loads are increased, ie. the structure's ability to sustain loads is maintained Unstable or violent Buckling is a buckling in which the displacements increase instantaneously, the load carrying capacity nose- dives and the structure collapses catastrophically
STRUCTURAL MEMBERS •
•
•
•
Column – A structural member which transmits the load of the structure above it through compression to other members Strut – A structural member designed to resist longitudinal compression Plate / Panel – A structural member whose third dimension is small compared to the other two dimensions Shell – A thin shell is defined as a shell with a thickness which is relatively small compared to its other dimensions and in which deformations are not large compared to thickness. A primary difference between a shell structure and a plate structure is that, in the unstressed state, the shell structure has curvature as opposed to plates structures which are flat
COLUMN BUCKLING • Column buckling – Buckling is defined as an instance of lateral bending or bowing of the column shape due to a compressive load on a column.
π 2 EI Pcr = k L2
S.No
Type
k
1
Pinned
1
2
Fixed-fixed
4
3
cantilever
1/4
COLUMN BUCKLING (Contd..) • Column buckling can be classified as – Primary instability • Cross sections are translated or rotated but not distorted
– Secondary instability • Cross sections are distorted but not translated or rotated
LOAD VS DEFLECTION
Load P
Pcr
Perfect structures
p im
ct u tr s ct e f er
e ur
s
Lateral Deflection
COLUMN CLASSIFICATION Type
Short Intermediate Long
Material Structural Steel
Aluminium alloy (6000)
Aluminium alloy (2000)
Wood
SR<40
SR<9.5
SR<12
SR<11
40<SR<150
9.5<SR<66
12<SR<55
SR>150
SR>66
SR>55
Slenderness Ratio (SR) =Leff /ρ
Long Intermediate
Short
Column Theory Johnson
11<SR<30 Inelastic SR>30
Euler
BUCKLING SHAPES
BEAM BENDING EQUATION From Flexure formula
M 1 = ... (1) EI R
Radius of curvature
d2 y
2 1 dx = ... (2) 2 3/ 2 R dy 1 + dx
Ignoring higher order terms 1 d2 y = 2 ... (3) R dx
From (1) & (3)
d2 y EI 2 = M... (4) dx
EULER BUCKLING FORMULA d 2v
M = dx 2 EI
Beam deflection equation
d 2v
Applying the column load
dx 2
Actual solution Substituting in equation
P v=0 EI
v = 0 at
x = 0 ⇒ 0 = C2
v = 0 at
x = L ⇒ 0 = C1 sin(λL)
x v = C1 sin(nπ ) L
P − λ2C1 sin(λx) + C1 sin(λx ) = 0 EI
(
P − λ2 )C1 sin(λx) = 0 EI
P
B
B Spring
L
A
v = C1 sin(λ x) + C 2 cos(λ x)
Solution to the above equation Boundary conditions
+
P
λ=
A
Fully Aligned
nπ L
P
P
B
v
M= -Pv
x
A
P P
B
P = EIλ 2
Pcr =
n 2π 2 EI L2
A n=1
n=2
PLASTICITY REDUCTION FACTOR • If the elastic buckling stress is more than the yield stress, plasticity reduction factor has to be applied. For columns, plasticity reduction factor is applied through tangent modulus
INELASTIC BUCKLING • For a column with intermediate length, buckling occurs after the stress in the column exceeds the proportional limit of the column material and before the stress reaches the ultimate strength. This kind of situation is called inelastic buckling
Euler Engesser
Pcr =
Reduced modulus theory Pcr =
π2 Er I
(L ) eff
2
Er =
π2 Et A Leff ρ
(
2
4 EE t E + Et
)
2
REDUCED MODULUS FORMULA
d1
As load remains constant
∫σ 0
Moment equilibrium
d1
∫σ ( y x
0
1
x
dA = ∫ σ v dA
………… (1)
0
d2
+ e )dA + ∫ σ v ( y 2 − e ) dA = − Pv
σ1 σx = y1 d1 Change in slope
d2
………… (2)
0
σv =
σ2 y2 d2
σ1 σ2 d 2v = = dz 2 Ed 1 E t d 2
………… (3) ………… (4)
REDUCED MODULUS FORMULA d 2v E 2 dz
Equation (1) becomes
d1
d 2v ∫0 y1dA − E t dz 2
d2
∫ y dA = 0 2
0
………… (5)
Equation (2) becomes d d2 d1 d2 2 d 2 v 1 2 d v 2 = − Pv E y dA + E y dA + e E y dA + E y dA 1 t ∫ 2 1 t ∫ 2 2 2 ∫ ∫ dz 0 dz 0 0 0
………… (6)
Equation (5) in (6) gives d 2v ( EI1 + E t I 2 ) = − Pv dz 2
………… (7)
Equation (7) can be rewritten as
d 2v E r I 2 + Pv = 0 dz
………… (8)
Solving (8) we get
π2 Er I Pcr = l e2
………… (9)
EULER ENGESSER FORMULA Equation (9) gives
But
π2 Er I Pcr = l e2
E r I = EI1 + E t I 2
………… (10) ………… (11)
If there is no strain reversal, then entire region gets compressive stress
E r I = E t I1 + E t I 2 = E t I
………… (12)
Now Equation (10) becomes
π2 Et I Pcr = l e2
………… (13)
EULER ENGESSER CURVE Euler Engesser curve is divided in to three regions 1) Block compression 2) Short column range ( Plasticity effects) 3) Long column range (Euler buckling)
TANGENT & SECANT MODULUS Tangent modulus
Et =
Et =
E 3 F 1 + n 7 F0.7
n −1
n = 1+
E 0.002 En σ n−1 1 + σ σ cy cy
Secant modulus
Es =
E n 0 . 002 E σ 1 + σ σ cy
log e ( 17 / 7 ) F0.7 F0.85
COLUMNS ON ELASTIC FOUNDATION
The stiffness of elastic foundation increases the buckling load and reduces the buckling length Critical column load
π 2 EI Pcr = 2 L
βL4 2 4 m π EI
4 β L m 2 ( m + 1 )2 = 4 π EI
µ β= a
CRIPPLING •Crippling is defined as the post-buckling failure of a axial section that is comprised of plate elements joined together at their boundaries •All the members subjected to axial load are to be checked for crippling •When local buckling takes around 0.7 - 0.8 Fcy , The crippling stress will be same as buckling stress
Web and Flange elements
CRIPPLING -ASSUMPTIONS •Material is isotropic •Material is ductile •b/t ratio is less than 3.0 •Transverse shear is ignored • Webs are assumed to have constant thickness Crippling stress
STRESS DISTRIBUTION •As the buckling takes place, the increasing load is transferred to the corners. •Stress build up at the corner after the buckling is not well understood •Boundary restraint between flange and plate element is unknown
FLANGE CRIPPLING The crippling stress is defined by dividing the failure load at which the flange collapses by the area of the flange.
Pre-Buckled
Post-Buckled
Stress distribution in a flange
WEB CRIPPLING The crippling of a web is similar to flange. The stress is uniform before buckling and increases near the edge after buckling
Stress distribution in a Post-Buckled web
INPLANE WARPING The post-buckled stress distribution in a flange or web is affected by the presence of restraints for in-plane lateral deflections at the un-loaded edges
Unrestrained
Restrained
WEB CRIPPLING STRESS DISTRIBUTION
Straight unloaded edges
Unloaded edges free to warp in the plane of plate
POST BUCKLING OF PLATES Thick plates crippling will take place in plastic range and for thin plates in elastic range
Stress distribution
PREDICTION OF CRIPPLING STRESS
1 Angle method (Needham method) • Member is divided into number of angles • Crippling strength is obtained by summation of individual crippling strength
Fcs
(F E )
1
cy
= n
ce
( t) b′
0.75
c e = 0.316
(for two edges free)
c e = 0.342
(for one edge free)
c e = 0.366
(for no edge free)
b′ =
a+b 2
Crippling load
For other sections
for an angle section
Pcs = Fcs A Fcs =
∑ crippling loads of angles ∑ Area of angles
PREDICTION OF CRIPPLING STRESS (Contd…) 2
Gerard method
For angles, V-groove plates, stiffened panels with distorted unloaded edges 0.85
2 Fcs gt = 0.56 A Fcy
E Fcy
1
2
For T, H, cruciform, plates with undistorted edges
2 Fcs gt = 0.67 A Fcy
E Fcy
1
2
0.40
For Z,J and channel
2 t E Fcs = 3.2 A Fcy Fcy
1
3
0.75
CORRECTION FOR CLADDING Maximum crippling stress
σ cl f 1 + 3 σ cr η= (1 + 3 f ) f=Cladding thickness / Total thickness =0.1 for 2024-T3 = 0.08 for 7075-T3
S.No
Section
Value
1
Angles
0.7Fcy
2
V-groove plates
Fcy
3
Multi-corner sections
0.8Fcy
4
Stiffened panels
Fcy
5
H, T, Cruciform
0.8Fcy
6
Z,J, Channel
0.9Fcy
RESTRAINT BY LIPS & BULB •
•
•
Compressive buckling coefficient of a element can be increased the presence of lip or bulb Compressive buckling coefficient for a – Plate element is 4 – Flange is 0.43 To provide a simple support the lip and bulb dimensions should satisfy
IL
i.e
Dmin t f
4
AL 2.73 − ≥5 3 bf t bf t
− 1.6 D min t f
bL b ≤ 0.328 f tL tf 3
− 0.374 Dmin t f
2
For lip
b = 7.44 f t f
For Bulb
PREDICTION OF CRIPPLING STRESS (BOEING) • Divide the section into segments. • For each segment, determine the ratio of the width to the thickness (b/t) • For each segment, determine the boundary conditions (1EF or NEF) • For each segment, determine the crippling stress and crippling load based on the method of analysis appropriate for the b/t region • Add the contribution of the crippling load from each segment to obtain the total crippling load • The crippling stress is obtained by dividing the crippling load by the calculated area
SECTION DETAILS If t > t other t other b=b + 2 If t other > t t b=b + 2
Extruded / Machined
Formed section
b = b + 0.57 R
SECTION DETAILS
t 0 + t1 t= 2 Tapered section Stepped section
SECTION DETAILS
t thick ,cor
bthick = 3t thin bthin
Thin / thick section
3 1 t f b f Rθ = 20 b w t w3
Rα
tf 2 Rα = cos α + b f
2
sin 2 α
Adjoining flange
SECTION DETAILS Dmin t f
4
Bulb section
− 1.6 D min t f
3
− 0.374 D min t f
2
b = 7.44 f t f
The bulb will fall into one of the following three categories: 1) Case 1: Diameter large enough (D > Dmin ) The flange may be considered a web (NEF) for the purposes of the crippling analysis 2) Case 2: Intermediate diameter (2 tf < D < Dmin ) The flange may be considered as a web (NEF), but the crippling stress for this segment (including the bulb) will be adjusted to 70% of the stress calculated assuming the no-edge-free condition 3) Case 3: Diameter too small (D < 2 tf ) Consider the flange as one edge-free. The area of the bulb should be added to the flange area, and bf should be measured to the tip of the bulb
SHEET EFFECTIVE WIDTH •
Aircraft structures consists of sheet and stringers together
•
Sheet and stringer deform together. Hence, the effective sheet width has to be taken into account in calculating the crippling stress
•
Ignoring the sheet will be over conservative design
kc π 2 E t Fcr = 2 12 1 − ν b
(
t = 3.6E b
)
2
2
SHEET EFFECTIVE WIDTH (Contd..) Von-Karman Sechler method
E w = 1.9t F cy NASA Structures manual 2w e = Kt
( E s ) skin ( E s ) stiff
1 f stiff
Es 2w e = Kt f stiff
f stiff =
P Astiff
For skin & stiffener different material
For skin & stiffener same material
K = 1.3 1.7
for one edge free for no edge free
CURVED PANEL
Pcr = P flat + Pcurved
Pcr = ( Fc ) stiff ( Ast + 4t s w e ) + ( Fc ) curved ( b − 2 w e ) t s
REGIONS OF CRIPPLING CURVE For calculating the crippling stress, the crippling curve is divided into three regions based on the value of b/t: 1) Stress cutoff 2) Plastic plate buckling 3) Empirical crippling curve
JOHNSON-EULER FORMULA
COLUMN BUCKLING CURVE
INCREASING CRIPPLING STRESS • Select a material with higher E and yield stress • Reduce b/t ratio • Add appropriate lip or bulb to change the edge conditions
PLATE STRUCTURES DEFINITION : PLATE IS A STRUCTURAL MEMBER WHOSE THIRD DIMENSION IS COMPARATIVELY SMALLER TO THE OTHER TWO DIMENSIONS AND SUBJECTED TO NORMAL LOAD / INPLANE LOAD
CLASSIFICATION OF PLATES: a) Thick plate : δ < 0.5 Load resisted by bending t δ 0 . 5 < < 5 Load is resisted by bending b) Thin plate : and inplane actiont c) Membrane : δ > 5 t
Load is resisted by tension
PLATE THEORIES • KIRCHHOFF PLATE THEORY – Shear deformation is ignored w = w( x, y ) ∂w u = −z ∂x ∂w v = −z ∂y
ε x = −z
∂θ y
∂x ∂θ ε y = −z x ∂y ∂θ ∂θ γ xy = z y − x ∂x ∂y
γ yz = γ xz = 0
• MINDLIN PLATE THEORY – Shear deformation is considered ∂θ w = w( x, y ) εx = z y ∂x ∂w ∂θ x u=z ε y = −z ∂x ∂y ∂w ∂θ y ∂θ x v = −z γ xy = z − ∂y ∂y ∂x
∂w −θx ∂y ∂w γ xz = −θ y ∂x
γ yz =
PLATE BENDING Qyz
Myx
My Qxz Mxy
Mx dy
Mxy
z Mx
y
Qxz
dx dz
x Myx
My Qyz
PLATE BENDING (Contd….) •
Plate bending equation is derived based on the following a) Strain-displacement relation ∂ 2 w0 2 ∂2x εx θx ε y = z θ y = − z ∂ w20 ∂y γ 2θ xy xy ∂ 2 w0 ∂ x ∂ y
………….. (1)
b) Stress-strain relation εx 1 1 ε y = − ν γ E 0 xy
−ν 1 0
0 σ x 0 σ y 2(1 + ν ) τ xy
………….. (2)
PLATE BENDING (Contd….) c) Moment & Force resultants M x t / 2 σ x M y = ∫ z σ y dz M −t / 2 τ xy xy Qxz = t / 2 τ xz dz Q yz −t∫/ 2 τ yz
………….. (3)
………….. (4)
d) Equilibrium equations ∂M xy ∂M y Q yz = + ∂x ∂y ∂M x ∂M yx Qxz = + ∂x ∂y
∂Qxz ∂Q yz + = −p ∂x ∂y
………….. (5) ………….. (6)
………….. (7)
PLATE BENDING (Contd….) (5) & (6) in (7) gives
∂ 2 M xy ∂ 2 M y ∂2M x +2 + = −p 2 2 ∂x ∂x∂y ∂y
………….. (8)
(3) in (8) gives
∂ 2σ xy ∂ 2σ y ∂ 2σ x dz = − p + ∫ z 2 + 2 2 −t / 2 ∂x∂y ∂y ∂x t/2
………….. (9)
Rearranging (2), we get
σ x 1 ν E σ y = ν 1 2 τ (1 − ν ) xy 0 0
0 ε x 0 ε y (1 − ν ) γ xy 2
∂ 2 w0 2 ∂2x σ x 1 ν 0 ∂ w σ y = − Ez 2 ν 1 0 20 ( 1 − ν ) 0 0 (1 − ν ) ∂y τ xy ∂ 2 w0 ∂x∂y
………….. (10)
(1) in (10) gives
………….. (11)
PLATE BENDING (Contd….) (11) in (9) gives
∂ 4 w0 ∂ 4 w0 ∂ 4 w0 ∂ 4 w0 Ez 2 ∂ 4 w0 4 + ν 2 2 + 2(1 − ν ) 2 2 + 4 + ν 2 2 dz = p ∫ 2 −t / 2 1 − ν ∂x ∂y ∂x ∂y ∂y ∂x ∂y ∂x t/2
∂ 4 w0 ∂ 4 w0 ∂ 4 w0 Ez 2 dz 4 + 2 2 2 + = p ∫ 2 4 −t / 2 1 − ν ∂x ∂y ∂y ∂x t/2
But
3 E Et 2 D= ∫ z dz = 2 −t / 2 1 − ν 12(1 − ν 2 )
………….. (12)
t/2
………….. (13)
(13) in (12) gives
∂4 ∂4 ∂4 D 4 + 2 2 2 + 4 w0 = p ∂x ∂y ∂y ∂x
p ∇ w= D 4
………….. (14)
PLATE SOLUTION Plate equation is given as
∂4w ∂4w ∂ 4 w p ( x, y ) 4 + 2 2 2 + 4 = ∂x ∂y ∂y D ∂x
………….. (14)
Assume
mπx nπy w( x, y ) = ∑ ∑ Amn sin sin m =1 n =1 a b ∞ ∞ mπx nπy q ( x, y ) = ∑ ∑ amn sin sin m =1 n =1 a b ∞
∞
………….. (15) ………….. (16)
4
∂4w ∞ ∞ mπ mπx nπy = A sin ∑ ∑ mn sin 4 m =1 n =1 ∂x a a b 4
∂ w nπ mπx nπy = A sin ∑ ∑ sin mn 4 m =1 n =1 ∂y b a b 4
∞
∞
………….. (17)
2
………….. (18)
2
∞ ∞ ∂4w mπ nπ mπx nπy = ∑ ∑ Amn sin sin 2 2 m = 1 n = 1 ∂x ∂y a b a b
………….. (19)
PLATE SOLUTION (17), (18) & (19) in (14) gives 2 2 4 ∞ ∞ mπ 4 a mn mπx nπy m π n π n π sin ∑ ∑ Amn + 2 + − sin = 0 ………….. (20) m =1 n =1 a b b D a b a From (20), we get 2 2 4 mπ 4 amn m π n π n π Amn =0 + 2 + − ………….. (21) a a b b D
Amn =
amn
(
2 m π D 4
1 ∞ ∞ w( x, y ) = 4 ∑ ∑ 2 π D m =1 n =1 m
(
a
2 n 2 +
a
) b
amn 2 n + 2
………….. (22)
2
2
b2
)
mπx nπy sin sin 2 a b
For uniformly distributed load (q ( x, y ) = q0 ) mπ nπ sin sin ∞ 16 q0 ∞ 2 2 w( x, y ) = 6 ∑ ∑ 2 π D m =1, 3, 5,... n =1, 3, 5,... m 2 n 2 mn + a b For concentrated load mπa nπb mπx nπy sin sin sin L sin L L L 4P x y x y w( x, y ) = 4 ∑ ∑ 2 π DL x L y m =1 n =1 m 2 n 2 + Lx L y ∞
………….. (23)
………….. (24)
∞
………….. (25)
PLATE SOLUTION Moment
2 m 2 n +ν a ∞ b 16q0 ∞ mπx nπy Mx = 4 ∑ sin ∑ sin π m =1,3, 5,...n =1,3, 5,... m 2 n 2 2 a b mn + a b m 2 n 2 ν + a b ∞ 16q0 ∞ sin mπx sin nπy My = 4 ∑ ∑ π m =1, 3, 5,...n =1, 3, 5,... m 2 n 2 2 a b mn + a b
Stress
12M x z σx = t3
12M y z σy = 3 t
………….. (26)
………….. (27)
………….. (28)
………….. (29)
BUCKLING OF PLATES Nx
Nx
b
a Displacement
mπx nπy w( x, y ) = ∑ ∑ Amn sin sin m =1 n =1 a b ∞
∞
………….. (1)
Potential energy 2 2 ∂ 2 w ∂ 2 w ∂ 2 w 2 1a b ∂w ∂ 2 w ∂ 2 w U + V = ∫ ∫ D 2 + 2 − 2(1 −ν ) 2 − − N x dxdy 2 0 0 2 ∂x ∂y ∂x ∂x∂y ∂x ∂y ………….. (2) 2
∞ ∞ π abD ∞ ∞ 2 m n π 2b 2 U +V = N x ∑ ∑ m 2 Amn ∑ ∑ Amn + − m =1 n =1 8 m=1 n=1 a b 8 4
2
2
………….. (3)
BUCKLING OF PLATES (Contd..) Differentiating 2
m n π 2b ∂ (U + V ) π abD = Amn + − N x m 2 Amn ∂Amn 4 4a a b ………….. (4) 2
4
2
Critical buckling load
π a D m n N Cr = + 2 m a b 2
kπ 2 D N Cr = b2 Critical buckling stress
σ Cr
2
2
2
2
mb a k = + a mb
where
t = kπ E b 2
………….. (5)
2
………….. (6)
2
………….. (7)
END FIXITY COEFFICIENT
BUCKLING OF PLATES (from column formula) π 2 EI Pcr = k L2 σy σ εx = x − ν E E σy σ εy = −ν x E E
• Euler column formula • Stress-strain relation
………….. (1)
………….. (2)
• Assuming that the plate has no curvature in y direction ε y = 0 σ y = ν σx gives εx =
σx 1− ν 2 E
(
)
………….. (3)
• Flat plate has a smaller elongation compared to a column 2 Pcr =
π EI
(1 − ν )( L ) 2
eff
• For a rectangular plate • (5) in (4) gives
bt 3 L = a, I = 12
π2 E t σ cr = 2 12 1 − ν a
(
)
2
………….. (4)
and σ cr = Pcr tb ………….. (5) 2
………….. (6)
SHEAR BUCKLING OF PLATES
SHEAR BUCKLING OF PLATES • Shear buckling formula
Fcs π t =k E 2 η 12 1 − ν b 2
(
2
)
METHODOLOGY • Calculate a/b from plate dimensions measured between panel supports • Determine edge restraint fixity 3) Select the buckling coefficient curve for the edge condition most nearly representing the support conditions existing, enter curve with a/b from (1) and obtain "K" (or "k"). If the support condition is believed to be between two conditions represented by curves, obtain "K" for both, calculate average value or interpolate as desired. 4) Determine buckling stress from equation 12. If this stress is in the elastic range, η = 1.0 (skip to step (5)) 5) If the stress is in the plastic range, obtain the proper plasticity reduction factor η 6) If the material is Alclad material, calculate the cladding reduction factor
INCREASING THE BUCKLING LOAD OF PANELS Fcs π t =k E 2 η 12 1 − ν b 2
(
2
)
There are three primary effective ways to increase the buckling load of a panel: 1) Decrease the "b" dimension of the panel 2) Increase the thickness of the panel 3) Increase the fixity of the panel supports
LINEAR BUCKLING ANALYSIS
[ K ]{ X } = λ [ K G ]{ X } [K] – Stiffness matrix [KG] – Geometric Stiffness matrix {X} – Buckling shape λ - Buckling load factor
Solution method Lanczos method Subspace iteration Backward iteration
NONLINEAR STATIC ANALYSIS
[ K L + K NL ]{ X } = [ P] [KL] – Linear Stiffness matrix [KNL ] – Nonlinear Stiffness matrix {X} – Deflection vector [P} - Load vector
Solution method Newton-Raphson Method