CSE513 Tall Building Structures:: Core Walls
CSE513 Tall Building Structures:: Core Walls
CSE513 Tall Building Structures:: Core Walls
html
• Introduction to Torsion
2
Introduction to Torsion
3
Introduction to Torsion
max max J
T = max dA = dA =
2
AR R A R
R
ρ
J = 2 dA Polar moment of inertia
A
TR T
max = , = = max
J J R
1
For a solid circular section of radius R, J = R 4 . For an annular cross section of exterior
2
1
( )
radius Re and interior radius Ri, J = Re4 − Ri4 .
2
4
Introduction to Torsion
R x max
x max
=
R
d max max T
= = =
dx R GR GJ
5
Introduction to Torsion
(1) Uniform torsion (or St. Venant’s torsion, or pure torsion): warpings in the cross
sections are not constrained;
(2) Non-uniform torsion (or warping torsion, or constrained torsion): warpings in the
cross sections are fully or partially constrained.
Introduction to Torsion
T q T q T
q = , = = , max = =
2 AE t 2tAE tmin 2tmin AE
d q ds T ds T 4 AE2
* = =
dx 2GAE t
=
4GAE2 =
t GJ
, J=
ds
t 7
*F. Beer, E. R. Johnston and J. T. DeWolf, Mechanics of Materials, McGraw-Hill, 4th ed., 2006
Introduction to Torsion
Characteristics:
•Cross sections do not remain plane but warp. This is actually the case for any non-
circular cross section;
•In a rectangle, the maximum shear stress occurs at the midpoint of the longer sides.
At the 4 rectangle corners, which are the points located at the greatest distance from
8
the center, shear stresses vanish.
Introduction to Torsion
Tt d T
max = , = =
J dx GJ
J = bt 3 and J = bt 3
If the rectangle qualifies as “thin-wall”, one can adopt:
1 1
= , J = J = J bt 3
3 3
If the cross section consists of several portions of different thicknesses:
1
J = bi ti3
3 i
See, e.g., S. P. Timoshenko and J. N. Goodier, Theory of Elasticity, 3rd ed., McGraw-Hill,
New York, 1969, Sec 109.
Torsional Behavior of Tall Buildings
• Reinforced concrete cores usually comprise an assembly of connected shear walls forming
a box section with openings that may be partially closed by beams or floor slabs.
• If a building is subjected to twist, as many are, the torsional stiffness of the core can be a
significant part of the total torsional resistance of the building.
• As the thickness of core walls is generally much smaller than the size of the cross section,
core walls can be taken as vertical thin-walled cantilever bars in the analysis.
• When the core twists, originally plane sections of the core warp.
• Thin-walled cores in tall buildings are generally free at the top end and are prevented from
warping at the bottom by the foundation (i.e., non-uniform torsion).
• The vertical warping stresses at the base of the core may be of the same order of
magnitude as the bending stresses.
T T ( z ) = T ( z ) + Ts ( z )
( s ) = rds
s
d = r ds
* *
0
s1 * ( s1 )
P s
AE
( s ) = 2 AE
* * ( s )
Profile of section
12
Sectorial properties of thin-walled cross-sections
Origin with
zero warping
displacement
(s)
Bending Warping
13
Sectorial properties of thin-walled cross-sections
Warping Function
s
It is defined as = r ds → d = r ds
0
ds
B
N
1
( s) s
2
P
1 1 rds d
Area of PNB = Base Height = ds r = =
2 2 2 2
= warping function, representing the longitudinal distribution of
displacement over a cross-section = sectorial coordinate
= double sectorial area (twice the sectorial area swept out by the
radius vector PN) 14
Sectorial properties of thin-walled cross-sections
Determination of the Principal Pole and Principal Origin
y
v
N
r
s
P(Pole)
Principal Axes x
15
Sectorial properties of thin-walled cross-sections
Determination of the Principal Pole and Principal Origin
y
N(x,y)
− 90
b
a
x
b = ( y − y p ) sin( − 90 ) = −( y − y p ) cos
0
16
Sectorial properties of thin-walled cross-sections
Determination of the Principal Pole and Principal Origin
y
N(x, y)
1
2
P(xp, yp) O*
1 *
O
2
P*(xp*, yp*)
17
Sectorial properties of thin-walled cross-sections
y
ds
sin ds = dy r = ( x − x p ) sin − ( y − y p ) cos
cos ds = dx r ds = ( x − x p )dy − ( y − y p )dx
x
Let a x = x p − x p* , a y = y p − y p*
(O*) = value of at O*
P and O can be determined once ax, ay and (o*) are found.
d = rds = ( x − x p )dy − ( y − y p )dx
d * = r * ds* = ( x − x p* )dy − ( y − y p* )dx
d − d * = −a x dy + a y dx and = * + a y x − a x y + c
where c is a constant.
The difference between and * is a linear function of x and y.
18
Sectorial properties of thin-walled cross-sections
With = * +a y x − a x y + c
( s )xdA = 0
A
( s ) ydA = 0
A
I x = A y 2 dA A * dA + cA = 0
A xydA = 0 ➔ A * ydA − a x I x = 0
I y = A x 2 dA A * xdA + a y I y = 0
19
Sectorial properties of thin-walled cross-sections
1
Hence c = − A * dA
A
1
a x = A * ydA
Ix
1
a y = − A * xdA
Iy
Principal pole: x p = a x + x p*
y p = a y + y p*
Principal origin:
P*
x
Proof: Assume P* and O*
* = anti-symmetrical
21
Sectorial properties of thin-walled cross-sections
1
c = − A *dA = 0
A ➔ x p = ax + x p* = 0 ➔ P on y
1
a x = A y dA= 0
Ix
Since c = ax = xo* = 0
(O *) = * (O *) + ay x − ax y + c = 0 + 0 − 0 + 0=0
22
Theory of non-uniform torsion
Governing Differential Equation for Non-Uniform Torsion
T ( z ) = T ( z ) + Ts ( z )
T = Q1 x1 + Q2 x2 Ts = GJ ' ( z )
(Warping torque) (St. Venant torque)
Theory of non-uniform torsion
Assumptions
•Rigid section contour (i.e., the section rotates as a rigid body about the central
axis, although it is allowed to warp in the z direction);
•Shear strain of the middle surface is zero, i.e.,
v w
sz = + = 0
z s
y
v
v
w
w
z x
24
Theory of non-uniform torsion
Axial displacement w
v w w v
Since sz = + =0 =−
z s s z
v
= [( r ( s ) ( z )] = r ( s ) ( z )
z z
Note: ' ( z ) = ( ( z ))
z
Integration to find w (s)
v
s s
w = − ds + w0 ( z ) = − ' ( z ) r ( s )ds + w0 ( z )
0 z 0
w0 ( z ) = w( z ) |s = 0
M x = A y dA = 0 ➔ A ( s ) ydA = 0
M y = A xdA = 0 ➔ A ( s ) xdA = 0
Note: A
w0 ( z)xdA = w0 ( z) ydA = 0
A
26
Theory of non-uniform torsion
w = − ( z ) ( s ) = − E ( z ) ( s )
A ( s ) xdA = 0
Determines the principal pole P
A ( s ) ydA = 0
27
Theory of non-uniform torsion
Warping Moment (Bi-Moment)
Since = − E ( z ) ( s )
Define: M = A dA = − E " A dA = − EI "
2
M
Since = −
EI
M M
= − E " ( z ) ( s) = − E ( − ) ( s ) =
EI I
M= summation of the moments of the elemental force
ωtds when is taken as the “moment arm” 28
Theory of non-uniform torsion
29
Theory of non-uniform torsion
Shear Stress
+ ds
s
ds ω
dz
t
+ dz
z
z
30
Theory of non-uniform torsion
31
Theory of non-uniform torsion
Warping Torque
T = A ( tds) r = E ( z ) A S d
where
tds = elemental shear force
r = distance to the principal pole
32
Theory of non-uniform torsion
→ S d = − 2 dA = − I
A A
EI = warping rigidity
Thus the warping shear stress
T
t = E ( z ) S ( s) = − S ( s)
I
Compare with the formula for flexural shear stress
QS
= Q Q
Ib
b
33
Theory of non-uniform torsion
Governing Differential Equation for Non-Uniform Torsion
dT x
= − EI IV ( z ) + GJ ( z )
dz
m T
EI IV ( z ) − GJ ( z ) = m( z )
dT
m=−
dz
Primary unknown: (z ) z T+dT
EI l
Let d 2 = , k = l = l GJ , then d =
GJ d EI k
EI IV − GJ = m ( z )
GJ m(z)
IV
− =
EI EI
1 m(z)
IV
− 2 =
d EI
Closely analogous to the
governing equation of
k2 m( z )
IV − 2 ( z ) = wall-frame structures.
l EI
35
Theory of non-uniform torsion
k is non-dimensional
d has the dimension of length = characteristic length for non-
uniform torsion
Solution:
kz kz
(z ) = C1 sinh + C2 cosh + C3 z + C4 + p (z )
l l
where C1, C2, C3, C4 are to be found using boundary conditions,
and p is a particular solution.
Solution = Complementary Part + Particular Solution
e x − e− x
sinh( x ) =
2
e x + e− x
cosh( x ) = 36
2
Theory of non-uniform torsion
Boundary Conditions
Generalized force T M
Associated displacement ’
Pair I Pair II
k kz k kz
'( z ) = C1 cosh + C2 sinh + C3 + p ( z )
l l l l
kz kz
M ( z ) = − EI ( z ) = −GJ C1 sinh + C2 cosh + d p ( z )
2
l l
T ( z ) = − EI ( z )
k
kz k
kz l 2
= −GJ C1 cosh + C 2 sinh + 2 p ( z )
l l l l k
39
Theory of non-uniform torsion
T s ( z ) = GJ ( z )
k kz k kz
= GJ C1 cosh + C 2 sinh + C 3 + p ( z )
l l l l
T ( z ) = Ts + Tw = GJ [C3 + p ( z ) + d 2 p( z )]
( s)
Warping normal stress ( s, z ) = M ( z)
I
S ( s)
Warping shear stress ( s, z )t = − T ( z )
I
where I = A 2dA
S ( s ) = A ( s )dA
o 40
Examples
Example 1
Pure torsion as a special case of non-uniform torsion
y
Solution: T0
x
m = 0 → p = 0
B.C.s:
z = 0: T = To , M = 0 l
z=l: T = To , M = 0 z
T0
To
T = To = GJC3 → C3 =
GJ
M (0) = 0 → C2 = 0
M (l ) = 0 → C1 = 0
41
Examples
To
( z) = z + C4
GJ
To
( z ) =
GJ
M ( z) = 0
Ts ( z ) = To
T ( z ) = 0
T ( z ) = To
42
Examples
Example 2
43
Examples
z
q = q0 (1 − )
l
l
B t A
a
4a
a
3a
44
Examples
Solution:
First, draw * diagram with both the origin and the pole at
the middle of the web.
* = r *ds
3a
B* = 2ads1 = 6a 2 , * = 2as1 , between A and B
0
a
s1
= 6a + 3ads 2 = 6a + 3a = 9a
* 2 2 2 2
c 6a2
0
s2 B A
y P*, O* 2a
9a2 x
2
9a P 2a
ax
6a2
3a
45
Examples
1 2 3a a
ax = A ydA = I x 0 2ads1 t + 0 ( 2a − s2 ) tds2
* * *
Ix
1 1 3
2
I x = t (4a ) + 2 (3at ) ( 2a ) + 2 a t + at a = 34a 3t
3 2 3
12 12 2
3a a a
A ydA = 2 2at 0 ds1 + 2at 0 ds2 − t 0 s2ds2
* * * *
1 9 a 2
+ 6 a 2
= 2 2at 3a 6a + 2at
2
a
2 2
2 a 1 2
− t 6a a + 3a a a
2
2 2 3
4
1 58a t
ax = A ydA = 34a3t = 1.706a
*
Ix
3.4a2
= + ay x − ax y + c
*
1 7.3a2
with c = − A * dA = 0
A 7.3a2
1
a y = − A * xdA = 0
Iy
2.6a2 3.4a2
= * −a x y
Diagram of
47
Examples
3 z
m = a x + a q = 3.206qo a 1 −
2 l
The governing equation for non-uniform torsion of open
section bars is:
k2 m( z )
or IV
− 2 =
l EI
where
l GJ EI
k = =l d =
2
d EI GJ
48
Examples
Solution of the Differential Equation
= h + p
kz kz 3.206qo a z 3 z 2
= C1 sinh + C2 cosh + C3 z + C4 + −
l l GJ 6l 2
k kz k kz 3.206qo a z 2
= C1 cosh + C2 sinh + C3 + − z
l l l l GJ 2l
2 2
k kz k kz 3.206qo a z
= C1 sinh + C2 cosh + − 1
l l l l GJ l
3 3
k kz k kz 3.206qo a
= C1 cosh + C2 sinh +
l l l l GJl
49
Examples
M ( z ) = −EI ( z )
kz kz 3.206q0ad z
2
= −GJ C1 sinh + C2 cosh + − 1 (1)
l l GJ l
Ts = GJ 2
l
T = − EI = −GJd 2 = −GJ
k
k kz k kz 3.206q0ad
2
= −GJ C1 cosh + C2 sinh + (2)
l l l l GJl
k kz k kz 3.206q0 a z 2
T = GJ C1 cosh + C2 sinh + C3 + − z
l l l l GJ 2l
k kz k kz 3.206q0 ad
2
− GJ C1 cosh + C2 sinh +
l l l l GJl
3.206q0a z 2 3.206q0ad 2
= GJ C3 + − z − 50
GJ 2 l GJl
Examples
Boundary Conditions:
z = 0: M = 0 , T=0
z=l: = 0, = 0
(I) T ( 0) = 0 3.206qo ad 2
GJ C3 − =0
GJl
3.206qo ad 2
C3 =
GJl
(2) M (0) = 0
3.206qad 2
− GJ C2 − =0
GJ
3.206qad 2
C2 =
GJ
51
Examples
(III) (l ) = 0
k k 3.206q0a l
C1 cosh k + C2 sinh k + C3 + − l =0
l l GJ 2
1.603q0 al / (GJ ) − C2 (k / l ) sinh k − C3
C1 =
(k / l ) cosh k
(IV) (l ) = 0
3.206q0 a l 3 l 2
C1 sinh k + C2 cosh k + C3l + C4 + − =0
GJ 6l 2
3.206q0 al 2
C4 = − C1 sinh k + C2 cosh k + C3l −
3GJ
The expressions for M and T are given by equations (1)
and (2) with the constants given above.
52
Examples
Example 3 2b
Locate the principal pole, plot the
warping function, and determine
t
t
h
the warping constant for the mono-
symmetric I-section shown on the
t
right.
b
Solution:
y
P*,O* x
Since y axis is the axis of symmetry,
the principal pole and the principal
origin are on y axis.
53
Examples
= * + a y x − ax y + c
1
c = − A dA = 0
A
1
a x = A ydA = 0
Ix
1 bh
a y = − A xdA +
2
Iy −
bh -
2
1 3 3
I y t[( 2b) + b ] = tb
3 3
diagram
12 4
3 b/2 3 3 3
thx th b b thb
A
xdA = −b / 2 ( hx ) xtdx =
b/2
= ( + )=
3 −b / 2 3 8 8 12
ydA thb 3
4 h
ay = − A
=− 3 =−
Iy 12 3tb 9
54
Examples
h bh
= + ay x = − x 9 P
h
−
bh
9
9 h 9
b b/2 2 − x
I = dA = tdx + tdx
2 2 9
−b −b / 2
b b/2 4
= 2 tdx + 2 tdx
2 2 bh
4 9
0 0 − bh 8
9 hx
h2 2
b b / 2 64 2 2 9
= 2 x tdx + 2 h x tdx
0 81 0 81 ωdiagram
diagram
2h t x23 b
64 x 3 b/2
2h 2t b3 8b3
= + = +
81 3 0 3 0 81 3 3
2 2 3
= h tb
27
55
Examples
2 2
Note: I = I yof top flange 1 h + I y of bottom flange 8 h
9 9
1 1 1 64
= t ( 2b)3 h 2 + tb3 h 2
12 81 12 81
1 8 64 1 72 2
= th 2b3 + = th 2b3 = h 2tb3
12 81 81 12 81 27
56