Asa

IS RAL2005001 1707
Plane Frame and Grid Equations
Introduction
Many structures, such as buildings and bridges, are composed of frames and/or grids. This chapter develops the equations and
methods for solution of plane frames and grids. First, we will develop the stiffness matrix for a beam element arbitrarily oriented in a
plane. We will then include the axial nodal displacement degree of freedom in the local beam element stiffness matrix. Then we will
combine these results to develop the stiffness matrix, including axial deformation effects, for an arbitrarily oriented beam element. We
will also consider frames with inclined or skewed supports.
Two-Dimensional Arbitrarily Oriented Beam Element
We can derive the stiffness matrix for an arbitrarily oriented beam element,
shown in the figure below, in a manner similar to that used for the bar element. The local axes x and y are located along the beam
element and transverse to
the beam element, respectively, and the global axes x and y are located to be convenient for the total structure.
The transformation from local displacements to global displacements is given in matrix form as:
IS RAL2005001 1707
S dx C = cos
d C
x = d =
d S
C
y y S sin
Using the second equation for the beam element, we can relate local nodal degrees of freedom to global degree of freedom:
d1X

d1y
d1y S C 0 0 0 0
1

d2y 0 0 1 0 0 0 1
2 = d y = Sd x +Cdy
0 0 0 S C 0 d2 X
d2y
0 0 0 0 0 1
For a beam we will define the following as the transformation matrix:
S C 0 0 0 0
0 0 1 0 0 0
T= 0 0 0 S C 0
0 0 0 0 0 1
Notice that the rotations are not affected by the orientation of the beam. Substituting the above transformation into the general form of
the stiffness matrix
T
k =T kT gives:
12S2 12SC 6 LS 12S2 12SC 6LS
12SC 12C2 6LC 12SC 12C2 6LC

EI 6 LS 6LC 4L2 6LS 6LC 2L2
k= 3 2 2
L 12S 12SC 6LS 12S 12SC 6LS
12SC 12C2 6 LC 12SC 12C2 6LC
6 LS 6LC 2L2 6LS 6LC 4L2
Lets know consider the effects of an axial force in the general beam transformation.
IS RAL2005001 1707
Recall the simple axial deformation, define in the spring element:
f 1 1
1x d1x
f = AE
2x L 1 1
d2 x
Combining the axial effects with the shear force and bending moment effects, in local coordinates,
gives:
f1x
0 0 C1 0 0 d1x
C1
f 1 y
m 0 12C 6 LC 0 12C 6LC d1x
1 f 2 2 2 2
2 x
f 2 6 LC 2 4 C 2 L2 0 6 LC 2 2C 2 L2 1
0
y = 0 0 C 0 0
m C 1 1 d2x
2
0 12C 2 6 LC 2 0 12C 2 6LC2
d 2y
0 6 LC 2 2C 2 L2 0 6 LC 2 4C 2L2
where
C2
C1 = AE = EI
L L3
IS RAL2005001 1707
Therefore:
C1 0 0 C1 0 0
0 12C2 6LC2 0 12C2 6LC2
0 6LC2 4C2L2 0 6LC2 2C2L2

k =
C1 0 0 C1 0 0
0 12C2 6LC2 0 12C2 6LC2
0 6LC2 2C2L2 0 6LC2 4C2L2
The above stiffness matrix include the effects of axial force in the x direction, shear force in the y , and bending moment about the z
axis. The local degrees of
freedom may be related to the global degrees of freedom by:
S 0 0 0 0 d1x
d1x C
C 0 0 0 0
d1x
d1x S
0 1 0 0 0 1
0
1 = 0 0 C S 0 d2 x

d2 x 0
0 0 S C 0 d2 y
d2 y 0
0
1
2 0 0 0 0 2
where the transformation matrix, including axial effects is:
C S 0 0 0 0
S C 0 0 0 0
T= 0 0 1 0 0 0
0 0 0 C S 0
0 0 0 S C 0
0 0 0 0 0 1
IS RAL2005001 1707
Substituting the above transformation into the general form of the stiffness matrix
T
k =T kT gives:
k = EL
The analysis of a rigid plane frame can be undertaken by applying stiffness matrix. A rigid plane frame is defined here as a series of
beam elements rigidly connected to each other; that is, the original angles made between elements at their joints remain unchanged
after the deformation. Furthermore, moments are transmitted from one element to another at the joints. Hence, moment continuity
exists at the rigid joints. In addition, the element centroids, as well as the applied loads, lie in a common plane. We observe that the
element stiffnesses of a frame are functions of E, A, L, I, and the angle of orientation of the element with respect to the global-
coordinate axes.
IS RAL2005001 1707
Rigid Plane Frame Example
Consider the frame shown in the figure below.
The frame is fixed at nodes 1 and 4 and subjected to a positive horizontal force of 10,000 lb applied at node 2 and to a positive moment
of 5,000 lb-in. applied at node 3. Let E = 30 x 106 psi and A = 10 in.2 for all elements, and let I = 200 in.4 for elements 1 and 3, and I = 100
in.4 for element 2.
Element 1: The angle between x and x is 90
C=0 S =1
where
12I 12( 200) 2 6I 6(200) 3
L2 = (120)2 = 0.167 in L = 120 =10.0 in
E 30 106
= = 250,000 lb / in3
L 120
IS RAL2005001 1707
Therefore, for element 1:
d1x d1y 1 d 2x d2 y 2
0.167 0 10 0.167 0 10
0 10 0 0 10 0
k(1) = 250,000 10 0 800 10 0 400 lb

0.167 0 10 0.167 0 10 in
0 10 0 0 10 0
10 0 400 10 0 800

C =1 S=0
12I 12(100) 2 6I 6(100) 3
L2 = (120)2 = 0.0835 in L = 120 = 5.0 in
d 2x d 2y 2 d3x d3y 3
10 0 0 10 0 0
0 0.0835 5 0 0.0835 5
k( 2) = 250,000 0 5 400 0 5 200 lb

10 0 0 10 0 0 in
0 0.0835 5 0 0.0835 5
0 5 200 0 5 400

C=0 S = 1
12I 12( 200) 2 6I 6(200) 3
L2 = (120)2 = 0.167 in L= 120 =10.0 in
30 106
E= = 250,000 lb / in3
L 120
IS RAL2005001 1707
d3x d3y 3 d 4x d 4y 4
0.167 0 10 0.167 0 10
0 10 0 0 10 0
k (3 ) = 250,000 10 0 800 10 0 400 lb

0.167 0 10 0.167 0 10 in
0 10 0 0 10 0
10 0 400 10 0 800
The boundary conditions for this problem are:
d1x = d1y = d4 x = d4 y = 1 = 4 = 0
After applying the boundary conditions the global beam equations reduce to:
2
10,000 10.167 0 10 10 0 0 d x
0 0 10.0835 5 0 0.0835 5 d 2y
2
0 = 2.5 105 10 5 1200 0 5 200
3
0 10 0 0 10.167 0 10 d x
3
0 0 0.0835 5 0 10.0835 5 d y
5,000 0 5 200 10 5 1200 3
Solving the above equations gives:
d2x 0.211 in
d2y 0.00148 in

2 = 0.00153 rad
d3 x 0.209 in
d3y
0.00148 in
3 0.00149 rad
IS RAL2005001
1707
Element 1: The element force-displacement equations can be obtained using
f = kTd . Therefore, Td is:
0 1 0 0 0 0 d1x = 0 0
1 0 0 0 0 0 d1y = 0 0
Td = 0 0 1 0 0 0 1 = 0 = 0
0 0 0 0 1 0 d2 x = 0.211 in 0.00148 in
0 0 0 1 0 0 d2y = 0.00148 in 0.211 in
0 0 0 0 0 1 2 = 0.00153 rad 0.00153 rad

Recall the elemental stiffness matrix is:
C1 0 0 C1 0 0
6LC2
0 12C2 6LC2 0 12C2
0 6LC2 4C2L2 0 6LC2 2C2L2

k =
C1 0 0 C1
0 0
0 12C2 6LC2 0 12C2 6LC2 4C2L2
0 6LC2 2C2L2 0 6LC2

IS
RAL2005001 1707
Therefore, the
local force-
displacement 0 0
equations are:
10 10 00 0 10 0
400 0
0 0.167 10 0 0.167
10 0.00148 in
(1) 5 0 10 800 0 10
10 0.211 in
f = kTd = 2.5 10
10 0 0 10 0
800 0.00153 rad
Element 2: 0 0.167 10 0 0.167
100
0 10 400 0 10
Td =
Simplifying the above equations gives:0 0
f1x 3,700 lb
0
f1y 4,990 lb
376 k in
m1 =
f2 x 3,700 lb

f2y 4,990 lb
223 k in
m2
IS
RAL2005001 1707
0.209 in
The element force-displacement equations are:
0 0 0 1 0 d3 y = 0.00148 in 0.00148 in
0 0 0 0 0 d2 x = 0.211 in 0.211 in
1 0 0 0 0 d2y = 0.00148 in 0.00148 in

1 3
0 1 0 00 0 02 = 0.00153
0 0 rad
= 0.00149 rad0.00153 rad 0.00149 rad
0 0 1 0 0 d3 x = 0.209 in =
Therefore, the local force-displacement equations are:
10 0 0 10 0
0 0.211 in
0 0.0833 5 0 0.0833
5 0.00148 in
( 2) 200 0.00153 rad

5 0 5 400 0 5
f = kTd = 2.5 10
0 0.209 in
10 0 0 10 0
5 0.00148 in
0 0.0833 5 0 0.0833
0 5 200 0 5
400 0.00149 rad
Simplifying the above equations gives:
f2 x 5,010 lb
f2y 3,700 lb
m 223 k in
2 =
f 3 x 5,010 lb

f3 y 3,700 lb
221 k in
m3
Element 3: The element force-displacement equations are:
0 1 0 0 0 0 d3 x = 0.209 in 0.00148 in
1
0 0 0 0 0 d3 y = 0.00148 in 0.209 in
0
0 1 0 0 0 3 = 0.00149 rad 0.00149 rad
=
Td =
0
0 0 0 1 0 d4 x = 0 0
0
0 0 1 0 0 d4 y = 0 0
0
1
0 0 0 0 4 = 0 0
IS RAL2005001 1707

10 0 0 10 0
0 0.00148 in
0 0.167 10 0 0.167
10 0.209 in
400 0.00149 rad
(3 ) 5 0 10 800 0 10
f = kTd = 2.5 10
10 0 0 10 0 10 0
10 0
0 0.167 10 0 0.167
800 0
0 10 400 0 10
f3 x 3,700 lb
f 3 y 5,010 lb
226 k in
m
3 =
f 4x 3,700 lb

f4 y 5,010 lb
375 k in
m4
The frame is fixed at nodes 1 and 3 and subjected to a positive distributed load of 1,000 lb/ft applied along element 2. Let E = 30 x 106 psi
and A = 100 in.2 for all elements, and let I = 1,000 in.4 for all elements.
IS RAL2005001 1707
First we need to replace the distributed load with a set of equivalent nodal forces and moments acting at nodes 2 and 3. For a
beam with both end fixed, subjected to a uniform distributed load, w, the nodal forces and moments are:
wL (1,000)40
f2 y =f
3y = = = 20k
2 2
m2 = m3 = wL2 = (1,000)402 = 133,33 lb ft = 1,600 k in 12 12
If we consider only the parts of the stiffness matrix associated with the three degrees of freedom at node 2, we get:

C = 0.707 S = 0.707
where
E 30 106 3 12I 12(1,000) 2
L= 509 = 58.93 lb / in L2
= (12 30 2)2 = 0.0463 in
6I = 6(1,000) =11.78551 in3
L 12 30 2
d2 x d2 y 2
50.02 49.98 8.33

(1
)
k = 58.93 49.98 50.02 8.33
8.33 8.33 4000
Simplifying the above equation:
d2 x d2 y 2
2,948 2,945 491
(1)
k = 2,945 2,948 491
491 491 235,700

IS RAL2005001 1707
C=1 S=0
where
E 30 106 3 12I 12(1,000) 2
L = 480 = 62.5 lb / in L2 = (12 40)2 = 0.0521 in
6I 6(1,000)
= =12.5 in3
L 12 40
d2 x d2y
2
100 0 0
k ( 2) = 62.50 0 0.052 12.5 lb

in
0 12.5 4,000
Simplifying the above equation:
d2 x d2 y 2
6,250 0 0
k ( 2) = 0 3.25 781.25 lb
in
0 781.25 250,000
The global beam equations reduce to:
0 9,198 2,945 491 d 2x
20 k = 2,945 2,951 290 d 2y
2
1,600 k in 491 290 485,700
d 2x 0.0033 in
d 2y = 0.0097 in
2
0.0033 rad
IS RAL2005001 1707

0.707 0.707 0 0 0 0 0 0
0.707 0.707 0 0 0 0 0 0
0 0 1 0 0 0 0 0
Td = =
0 0 0 0.707 0.707 0 0.0033 in 0.00452 in
0 0 0 0.707 0.707 0 0.0097 in 0.0092 in
0 0 0 0 0 1 0.0033 rad 0.0033 rad
Recall the elemental stiffness matrix is a function of values C1, C2, and L
C1 = AE = (100)30 106 = 5,893 k in C2 = EI 30 106 (1,000)

= = 0.2273 k in
L 12 30 2 L3 (12 30 2)3
5,893 0 10 5,893 0 0 0
0 2.730 694.8 0 2.730 694.8 0
10 694.8 117,900 0 694.8 117,000 0

f(1) = kTd =
5,893 0 0 5,983 0 0 0.00452 in
0 2 .730 694.8 0 2.730 694.8 0.0092 in
0 694.8 117,000 0 694.8 235,800 0.0033 rad

f
1x
26.64 k
f1y
m
1 f 2.268 k
2x f
2y 389.1 k in
m =
2 26.64 k
2.268 k
778.2 k in
IS RAL2005001 1707
IS RAL2005001 1707
Element 2: The element force-displacement equations are:
1 0 0 0 0 0 0.0033 in 0.0033 in
0
1 0 0 0 0 0.0097 in 0.0097 in
0
0 1 0 0 0 0.0033 rad = 0.0033 rad
Td =
0
0 0 1 0 0 0 0
0
0 0 0 1 0 0 0
0 0 0 0 1 0 0
0
Recall the elemental stiffness matrix is a function of values C 1, C2, and L
C1 = AE = (100)30 106 = 6,250 k in C2 = EI = 30 106 (1,000) = 0.2713 k in
L 12 40 L3 (12 40)3
6,250 0 0.0033 in
0 0 6,250 0
0
0 6,250 0 3.25 781.1 0 3 .25 781.1 0.0097 in
0 125,000 0.0033 rad
781.1 250,000 0 781.1

f(2) = kTd = 0 0 6,250 0 0 0
781.1 0
3 .25 781.1 0 3.25
781.1 125,000 0 781.1 250,00 0
20.63 k
2.58 k
832.57 k in
kd =
20.63 k
2.58 k
412.50 k in
IS RAL2005001
1707
To obtain the actual element local forces, we must subtract the equivalent nodal forces.
f2x
0 20.63 k
20.63 k
f 2y
m 20 k 17.42 k
2.58 k
2 f
in 1600 k in = 767.4 k in
3 x
f
832.57 k 0 20.63 k
= 20 k 22.58 k
3y
20.63 k
m
3 2.58 k in 1600 k in 2,013 k in
412.50 k
Consider the frame shown in the figure below. In this example will illustrate the equivalent joint force replacement method for a
frame subjected to a load acting on an element instead of at one of the joints of the structure. Since no distributed loads are present, the
point of application of the concentrated load could be treated as an extra joint in the analysis.
This approach has the disadvantage of increasing the total number of joints, as well as the size of the total structure stiffness
matrix K. For small structures solved by computer, this does not pose a problem. However, for very large structures, this might
reduce the maximum size of the structure that could be analyzed.
IS RAL2005001 1707
The frame is fixed at nodes 1, 2, and 3 and subjected to a concentrated load of 15 k applied at mid-length of element 1. Let E = 30 x 106
psi, A = 8 in2, and let I = 800 in4 for all elements.
Solution Procedure
1. Express the applied load in the element 1 local coordinate system (here x is directed from node 1 to node 4).
IS RAL2005001 1707
2. Next, determine the equivalent joint forces at each end of element 1, using the table in Appendix D (see figure
below).
3. Then transform the equivalent joint forces from the local coordinate system forces into the global coordinate
system forces, using the equation f =T T f . These global joint forces are shown below.
IS RAL2005001 1707
4. Then we analyze the structure, using the equivalent joint forces (plus actual joint forces, if any) in the usual manner.
5. The final internal forces developed at the ends of each element may be obtained by subtracting Step 2 joint forces from
Step 4 joint forces.
Element 1: The angle between x and x is 63.43
C = 0.447 S = 0.895
where
12I 12(800) 6I 6(800)

= = 0.0334 in 2 = = 8.95 in3
)2
L2 ( 44.7 12 L 44.7 12
E 30 106
= = 55.9 lb / in3
L 44.7 12
d4x d4 y 4
90.0 178 448
(1)
k = 178 359 244
448 244 179,000

IS RAL2005001 1707

Element 2: The angle between x and x is 116.57
C = 0.447 S = 0.895
where
12I 12(800) 6I 6(800)

= = 0.0334 in2 = = 8.95 in3
)2
L2 ( 44.7 12 L 44.7 12
E 30 106
= =55.9 lb / in3
L 44.7 12
d 4x d4 y 4
90.0 178 448
(2)
k = 178 359 244
448 244 179,000

Element 3: The angle between x and x is 0 (The author of your textbook directed the element from node 4 to 3. In general, as we have
discussed in class, we usually number the element numerically or from 3 to 4. In this case the angle between x and x is 180)
C=1 S=0 E = 30 106 = 50 lb / in3
L 50 12
12I 12(800) 6(800)

= = 0.0267 in2 6I = = 8.0 in3
L2 ( 50 12)2 L 50 12
d 4x d4 y 4
400 0 0
(2)
k = 0 1.334 400
0 400 160,000
IS RAL2005001 1707
The global beam equations reduce to:
7.5 k 582 0 896 d4 x
0 = 0 719 400 d4y

900 k in 896 400 518,000 4
d4 x 0.0103 in
d4y = 0.000956 in
4
0.00172 rad
C S 0 0 0 0
S C 0 0 0 0
T= 0 0 1 0 0 0 C = 0.447 S = 0.895
0 0 0 C S 0
0 0 0 S C 0
0 0 0 0 0 1
0.447 0.895 0 0 0 0 0 0
0.895 0.447 0 0 0 0 0 0
Td = 0 0 1 0 0 0 0 = 0
0 0 0 0.447 0.895 0 0 .0103 in 0.00374 in
0 0 0 0.895 0.447 0 0.000956 in 0.00963 in
0 0 0 0 0 1 0 .00172 rad 0.00172 rad

IS RAL2005001 1707
Recall the elemental stiffness matrix is:
C1 0 0 C1 0 0
6LC2
0 12C2 6LC2 0 12C2
0 6LC2 4C2L2 0 6LC2 2C2L2

k =
C1 0 0 C1 0 0
0 12C2 6LC2 0 12C2 6LC2 4C2L2
0 6LC2 2C2L2 0 6LC2
C1 = AE = (8)30 106 = 447.2 k in C2 = EI = 30 106 (800) = 0.155 k in
L 12 44.72 L3 (12 44.72)3

447 0 0 447 0 0 0
0 1.868 500.5 0 1.868 500.5 0
0 500.5 179,000 0 500.5 89,490 0

f(1) = kTd =
447 0 0 447 0 0 0.00374 in
0 1.868 500.5 0 1.868 500.5 0.00963 in
0 500.5 89,490 0 500.5 179,000 0.00172 rad

1.67 k
0.88 k
158 k in
f(1) = kd =
1.67 k
0.88 k
311 k in
To obtain the actual element local forces, we must subtract the equivalent nodal forces.
IS RAL2005001 1707
f
1x
1.67 k 3.36 k 5.03 k
f1y
m
1 f 0.88 k 6.71 k 7.59 k
4 x
f 158 k in 900 k in 1,058 k in
4y = =
m 1 .67 k 3.36 k 1.68 k
4 0.88 k 6.71 k 5.83 k
311 k in 900 k in 589 k in
C S 0 0 0 0
S C 0 0 0 0
T= 0 0 1 0 0 0 C = 0.447 S = 0.895
0 0 0 C S 0
0 0 0 S C 0
0 0 0 0 0 1
0.447 0.895 0 0 0 0 0 0
0.895 0.447 0 0 0 0 0 0
0 0 1 0 0 0 0 0
Td = =
0 0 0 0.447 0.895 0 0.0103 in 0.00546 in
0 0 0 0.895 0.447 0 0.000956 in 0.00879 in
0 0 0 0 0 1 0.00172 rad 0.00172 rad
C1 = AE = (8)30 106 = 447.2 k in C2 = EI = 30 106 (800) = 0.155 k in
L 12 44.72 L3 (12 44.72)3

IS RAL2005001 1707
IS RAL2005001 1707
447 0 0 447 0 0 0
0 1.868 500.5 0 1.868 500.5 0

500.5 179,000 0 500.5 89,490 0
0
f( 2) = kTd =
0 0 447 0 0 0.00546 in
447
1.868 500.5 0 1.868 500.5 0.00879 in
0
0
500.5 89,490 0 500.5 179,000 0.00172 rad
2.44 k
0.877 k
158 k in
f(2) = kd =
2.44 k
0.877 k
312 k in
Since there are no applied loads on element 2, there are no equivalent nodal forces to account for. Therefore, the above
equations are the final local nodal forces

1 0 0 0 0 0 0.0103 in 0.0103 in
0 1 0 0 0 0 0.000956 in 0.000956 in
0 0 1 0 0 0 0.00172 rad 0.00172 rad

Td = =
0 0 0 1 0 0 0 0
0 0 0 0 1 0 0 0
0 0 0 0 0 1 0 0
IS
RAL2005001 1707
C1 = AE = (8)30 106 = 400 k in C2 = EI = 30 106 (800) = 0.111 k in
L 12 50 L3 (12 50)3
400 0 0 400 0 0 0.0103 in
0 1.335 400 0 1.335 400 0.000956 in

400 160,000 0 400 80,000 0.00172 rad
0
f(3) = kTd =
0 0 400 0 0 0
400
1.335 400 0 1.335 400 0
0
400 80,000 0 400 160,000 0

0
4.12 k
0.687 k
275 k in
f(3) = kd =
4.12 k
0.687 k
137 k in
Since there are no applied loads on element 3, there are no equivalent nodal forces to account for. Therefore, the above
equations are the final local nodal forces. The free-body diagrams are shown below.
IS RAL2005001 1707
500 kN
The frame is fixed at nodes 2 and 3 and subjected to a concentrated load of 500 kN applied at node 1. For the bar, A = 1 x 10-3 m2, for the
beam, A = 2 x 10-3 m2, I = 5 x 10-5 m4, and L = 3 m. Let E = 210 GPa for both elements.

Beam Element 1: The angle between x and x is 0
C =1 S=0
where
12I = 12(5 105 ) = 6.67 105 m2 6I = 6(5 105 ) =104 m3
L2 (3)2 L 3
E = 210 106 = 70 106 kN / m3
L 3
d1x d1y 1
2 0 0
k(1) = 70 103 0 0.067 0.10 kN

m
0 0.10 0.20
IS RAL2005001 1707
Bar Element 2: The angle between x and x is 45
C = 0.707 S = 0.707 where

d1x d1y
( 2) 103 m2 (210 106 kN / m2 ) 0.5 0.5

kN
k = 4.24 m 0.5 0.5 m
d1x d1y
0.354 0.354
kN
k ( 2) = 70 103 0.354 0.354 m
Assembling the elemental stiffness matrices we obtain the global stiffness matrix
2.354 0.354 0
K = 70 103 0.354 0.421 0.10 kN

m
0 0.10 0.20
The global equations are:
0 2.354 0.354 0 d1x
500 kN = 70 103 kN 0.354 0.421 0.10 d

1y
m
0 0 0.10 1
0.20
d1x 0.00388 m
d1y = 0.0225 m
1
0.0113 rad
Bar Element: The bar element force-displacement equations can be obtained
= using f .
kTd
d1x
f = AE f1x L 3 x 1 1 C S 0 0 d1y
d
1 1 0 0 C S 3x
d
3y
IS RAL2005001 1707
Therefore, the forces in the bar element are:
f =
AE (Cd
+ Sd ) = 670 kN
1
x 1x 1y
L
f = AE (Cd + Sd ) = 670 kN
3
x 1x 1y
L
Beam Element: The beam element force-displacement equations can be

obtained using f = kd . Since the local axis coincides with the global coordinate
system, and the displacements at node 2 are zero. Therefore, the local force-
displacement equations are:
C1 0 0 C1 0 0
0 12C2 6LC2 0 12C2 6LC2 AE

C1
=
0 6LC2 4C2 L2 0 6LC2 2C2L2 L

k =
C1 0 0 C1 0 0
0 12C2 6LC2 0 12C2 6LC2 C2 = EI3

L
0 6LC2 2C2 L2 0 6LC2 4C2L2
2 0 0 2 0 0 0.00388 m
0 0.067 0.10 0 0.067 0.10 0.0225 m

3 0 0.10 0.20 0 0.10 0.10
f(1) = kd = 70 10
2 0 0 2 0 0 0
0 0.067 0.10 0 0.067 0.10 0
0 0.10 0.10 0 0.10 0.20 0
Substituting numerical values into the above equations gives:
f1x 473 kN
f1y 26.5 kN
0.0
m1 =
f2 x 473 kN

f2y 26.5 kN
IS RAL2005001 1707
78.3 kN m
m2
IS RAL2005001 1707
The frame is fixed at nodes 1 and 3 and subjected to a moment of 20 kN-m applied at node 2. Assume A = 2 x 10-2 m2, I = 2 x 10-4 m4,
and E = 210 GPa for all elements.
C=0 S=1
where
12I = 12(2 104 ) = 1.5 104 m2 6I = 6(2 104 ) = 3 104 m3
L2 (4)2 L 4
E = 210 106 = 5.25 107 kN / m3
L 4
Therefore, the stiffness matrix for element 1, considering only the parts associated with node 2, is:
d 2x d 2y 2
0.015 0 0.03
k (1) = 5.25 105 0 2 0 kN

m
0.03 0 0.08
IS RAL2005001 1707

C=1 S=0
where
12I = 12(2 104 ) = 9.6 105 m2 6I = 6(2 104 ) = 2.4 104 m3
L2 (5)2 L 5
E = 210 106 = 4.2 107 kN / m3
L 5
Therefore, the stiffness matrix for element 2, considering only the parts associated with node 2, is:
d 2x d 2y 2
2 0 0
k ( 2) = 4.2 105 0 0.0096 0.024 kN

m
0 0.024 0.08
Assembling the elemental stiffness matrices we obtain the global stiffness matrix:
0.8480 0 0.0158
6
K= 10 0 1.0500 0.0101 kN
m
0.0158 0.0101 0.0756
0 0.8480 0 0.0158d 2x
6
0 =10 0 1.0500 0.0101 d 2y

20 kN m 0.0158 0.0101 0.0756 2
d 2x 4.95 106 m
d = 2.56 106 m
2y
4
2 2.66 10 rad
IS RAL2005001 1707
Element 1: The beam element force-displacement equations can be obtained

= using f
kTd .
0 1 0 0 0 0 0 0
1 0 0 0 0 0 0 0
Td = 0 0 1 0 0 0 0 = 0
0 0 0 0 1 0 4.95 10 6 m 2.56 106 m
0 0 0 1 0 0 2.56 10 6 m 4.95 106 m
0 0 0 0 0 1 2.66 10 4 rad 2.66 104 rad
C1 0 0 C1 0 0
0 12C2 6LC2 0 12C2 6LC2
0 6LC2 4C2L2 0 6LC2 2C2L2

k=
C1 C1
0 0 0 0
0 12C2 6LC2 0 12C2 6LC2
0 6LC2 2C2L2 0 6LC2 4C2L2
C1 = AE = (2 102 )210 106 =1.05 106 kN m

L 4
6 4
C2 = EI = 210 10 (2 10 ) = 656.25 kN m
L3 (4 )3
200 0 0 200 0 0 0
0 1.5 3 0 1.5 3 0
3 0 3 8 0 3 4 0
f(1) = kTd = 5.25 10 6
200 0 0 200 0 0 2.56 10 m
0 1.5 3 0 1.5 3 4.95 10 6 m
0 3 4 0 3 8 2.66 10 4 rad
IS RAL2005001
1707
Solving for the forces and moments gives:
f1x 2.69 kN
f1y 4.2 kN
5.59 kN m
m1 =
f2 x 2.69 kN

f2y 4.2 kN
11.17 kN m
m2
Element 2: The beam element force-displacement equations can be obtained

= using f
kTd .
1 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0
0 0 1 0 0 0 0 0
Td = 6 = 6
0 0 0 1 0 0 4.95 10 m 4.95 10 m
6 6
0 0 0 0 1 0 2.56 10 m 2.56 10 m
0 0 0 0 0 1 2.66 104 rad 2.66 104 rad
C1
= AE = (2 102 )210 106 = 0.84 106 kN
m
C2
L 5
EI 210 106 (2 104 )
= L3 = (5 )3 = 336 kN m
200 0 0 200 0 0 4.95 106 m
0 0.96 2.40 0 0.96 2.40 2.56 106 m
3 0 2.40 8 0 2.40 4 2.66 104 rad

f( 2) = kTd = 4.2 10
200 0 0 200 0 0 0
0 0.96 2.40 0 0.96 2.40 0
0 2.40 4 0 2.40 8 0
IS RAL2005001 1707
IS RAL2005001 1707
f2x 4.16 kN
f 2.69 kN
2y

m2 8.92 kN m
=
f 3 x 4.16 kN
2.69 kN
f3y
4.47 kN m

m3
Inclined or Skewed Supports
If a support is inclined, or skewed, at some angle for the global x axis, as shown below, the boundary conditions on the
displacements are not in the global x-y directions but in the x-y directions.
We must transform the local boundary condition of d3y = 0 (in local coordinates) into the global x-y system. Therefore, the relationship
between of the components of the displacement in the local and the global coordinate systems at node 3 is:
IS RAL2005001 1707
d '3 x d cos sin 0 d3x

' 3 y
= sin cos 0 d3y
' 3
0
3
0 1
We can rewrite the above expression as:
cos sin 0
{d ' 3 } = [t3 ]{d 3 } [t 3 ]= sin cos 0

0 0
We can apply this sort of transformation to the entire displacement vector as:
{d '} = [Ti ]{d } or {d }= [Ti ]T {d'}
where the matrix [Ti] is:
[I] [0] [0]
[T ] = [0] [I] [0]

i
]
[0] [0] [t 3
Both the identity matrix [I] and the matrix [t3] are 3 x 3 matrices.
The force vector can be transformed by using the same transformation.
{f '}= [Ti ]{}f
In global coordinates, the force-displacement equations are:
{}f = [K ]{d}
Applying the skewed support transformation to both sides of the force-displacement equation gives:
[Ti ]{}f = [Ti ][K ]{d}
IS RAL2005001
1707
By using the relationship between the local and the global displacements, the force-displacement equations become:
[Ti ]{f } = [Ti ][K ][Ti ]T {d '} {f '}= [Ti ][K ][Ti ]T {d'}
F1x d1x
F1y d1y
M1 1
F2 x d2 x
F2 y
Therefore the global equations become: = [Ti ][K ][Ti ]T d2y
2
M2
F'3 x d'3 x
F' d'
3y 3y

M3 1
Grid Equations
A grid is a structure on which the loads are applied perpendicular to the plane of the structure, as opposed to a plane frame where
loads are applied in the plane of the structure. Both torsional and bending moment continuity are maintained at each node in a grid
element. Examples of a grid structure are floors and bridge deck systems. A typical grid structure is shown in the figure below.
IS RAL2005001 1707
A representation of the grid element is shown below:

The degrees of freedom for a grid element are: a vertical displacement diy (normal to the grid), a torsional rotation ix about the x
axis, and a bending rotation iz about the z axis. The nodal forces are: a transverse force fiy a torsional moment mix about the x
axis, and a bending moment miz about the z axis.
Lets derive the torsional rotation components of the element stiffness matrix. Consider the sign convention for nodal torque and
angle of twist shown the figure below.
A linear displacement function is assumed.

= a1 + a2 x
Applying the boundary conditions and solving for the unknown coefficients gives:

2 x 1x
= x +
1
x
L
IS RAL2005001 1707
Or in matrix form:
1x
= [N N ]=
1 2

2x
where N1 and N2 are the interpolation functions gives as:
x x
N = 1 N =
1 2
L L
To obtain the relationship between the shear strain and the angle of twist consider the torsional deformation of the bar as
shown below.
If we assume that all radial lines, such as OA, remain straight during twisting or torsional deformation, then the arc length AB is:
AB = maxdx = Rd
Therefore;
max = Rddx
At any radial position, r, we have, from similar triangles OAB and OCD:
IS RAL2005001 1707
= r d = r (2 x 1x ) dx L
The relationship between shear stress and shear strain is:
= G
where G is the shear modulus of the material. From elementary mechanics of materials, we get:
mx = RJ
where J is the polar moment of inertia for a circular cross section or the torsional constant for non-circular cross sections. Rewriting the
above equation we get:
mx = GJL ( 2 x 1x )
The nodal torque sign convention gives:
= =
m1x m2 x
mx mx
Therefore;
m1x = GJL ( 1x 2 x ) m2 x = GJL ( 2 x 1x )
In matrix form the above equations are:
m1x
GJ 1 1 1x
m 2 =
x
L 1 1 2 x
IS RAL2005001
1707
Combining the torsional effects with shear and bending effects, we obtain the local stiffness matrix equations for a grid element.
f1y 12EI 6EI 6EI

L3 0 2 123EI 0 L2 d 1y

m1x L L 1x
0 GJ GJL 0
m L 0 0 2EI
1z
6EI 4EI 6EI 0 L 1z d2
f
2 0 L L2 6EI
y
2y = L 6E 0 2
12L3EI I 12EI 2 x
L
m 2 GJ 2z
0 L2 L3
0
x m L 0
GJL 0 0 4EI
2z 6EI
2EI 6EI 0 L
2
0 L L2
L
The transformation matrix relating local to global degrees of freedom for a grid is:
1 0 0 0 0 0
0 C S 0 0 0
0 S C 0 0 0
TG =
0 0 0 1 0 0
0 0 0 0 C S
0 0 0 0 S C
where is now positive taken counterclockwise from x to x in the x-z plane: therefore;
x j xi zj zi
C = cos = S = sin q =
L L
The global stiffness matrix for a grid element arbitrary oriented in the x-z plane is given by:
T k T
= G G
kGTG
IS RAL2005001 1707
Grid Example
The frame is fixed at nodes 2, 3, and 4, and is subjected to a load of 100 kips applied at node 1. Assume I = 400 in.4, J = 110 in.4, G = 12 x
10 3 ksi, and E = 30 x 10 3 ksi for all elements.
To facilitate a timely solution, the boundary conditions at nodes 2, 3, and 4 are applied to the local stiffness matrices at the
beginning of the solution.
d2 y = 2 x = 2z = 0 d3 y = 3 x
= 3 z = 0 d4 y = 4 x = 4 z = 0
Beam Element 1:
C = cos = x2L(1) x1 = 022.3620 = 0.894
S = sin = z2L(1) z1 = 2022.3610 = 0.447

where
12EI = 12(30 103 )(400) = 7.45 k
L3 (22.36 12)3 in 6EI 6(30 103 )(400)

2 = 2 = 1,000 k
L (22.36 12)
IS RAL2005001
1707
4EI 4(30 10 3 )(400) GJ (12 10 3 )(110)

= = 179,000 k in = = 4,920 k in
L (22.36 12) L (22.36 12)
The global stiffness matrix for element 1, considering only the parts associated with node 1, and the following relationship:
kGTG
kG =TG
1 0 0 1 0 0
T
T = 0 0.894 0.447 T = 0 0.894 0.447
G G
0 0.447 0.894 0 0.447 0.894
d1y 1x 1z
7.45 0 1,000
(1) k
k = 0 4,920 0 in
1,000 0 179,000
Therefore, the global stiffness matrix is
d1y 1x 1z
7.45 447 894
k (1) = 447 39,700 69,600 k

in
894 69,600 144,000
Beam Element 2:
C = cos = x3L(2) x1 = 022.3620 = 0.894

S = sin = z3L(2)z1 = 022.3610 = 0.447
where
12EI = 12(30 103 )(400) = 7.45 k

6EI 6(30 103 )(400)
2 = 2 = 1,000 k
L3 (22.36 12)3 in
L (22.36 12)
IS RAL2005001 1707
4EI 4(30 10 3 )(400) GJ (12 10 3 )(110)

= = 179,000 k in = = 4,920 k in
L (22.36 12) L (22.36 12)
T
k GT G
kG =TG
1 0 0 7.45 0 1,000 1 0 0
( 2)
k = 0 0.894 0.447 0 4,920 0 0 0.894 0.447
0 0.447 0.894 0 179,000 0.447 0.894
d1y 1x 1z
7.45 447 894
k (2) = 447 39,700 69,600 k

in
894 69,600 144 ,000
Beam Element 3:
x1
x 4 20 20 z4 z 1 0 10
C = cos = = =0 S = sin = = = 1
L(3) 10 L(3) 10
where
12EI 12(30 103 )(400) 6EI 6(30 103 )(400)

= = 83.3 k / in = = 5,000 k
L3 (10 12)3 L2 (10 12)2
4EI = 4(30 103 )(400) = 400,000 k in GJ = (12 103 )(110) = 11,000 k in
L (10 12) L (10 12)
T k T k T
= G G G G
IS RAL2005001 1707
1 0 0 83.3 0 5,000 1 0 0
(3)
k = 0 0 1 0 11,000 0 0 0 1
0 1 0 5,000 0 400,000 0 1 0
d1y 1x 1z
83.3 5,000 0
k(3 ) = 5,000 400,000 0
0 0 11,000
Superimposing the three elemental stiffness matrices gives:
d1y 1x 1z
98.2 5,000 1,790
K= 5,000 479,000 0
1,790 0 299,000
1,790 0
F1y = 100 k 98.2 5,000 d1y
299,000
M1x
=0 = 5,000 479,000 1x

1z
M1z =0 1,790 0
d1y 2.83 in
1x = 0.0295 rad
1z
0.0169 rad
IS RAL2005001 1707
IS RAL2005001 1707
Element 1: The grid element force-displacement equations can be obtained
using f = kGTG d .
1 0 0 0 0 0 2.83 in 2.83 in
0 0.894 0.447 0 0 0 0.0295 rad 0.0339 rad
0 0.447 0.894 0 0 0 0.0169 rad 0.00192 rad

TG d = =
0 0 0 1 0 0 0 0
0 0 0 0 0.894 0.447 0 0
0 0 0 0 0.447 0.894 0 0
7.45 0 1,000 7.45 0

1,000 2.83 in
0 4,920 0 0 4,920
0 0.0339 rad
1,000 0 179,000 1,000 0 89,500 0.00192 rad

f(1) = kTd =
7.45 0 1,000 7.45 0 1,000 0
0 0
0 4,920 0 0 4,920
1,000 0 89,500 1,000 0 179,000 0
f 19.2 k
1y

m1x 167 k in
m 2,480 k in
1z =
f2y 19.2 k
167 k in

m2x
2,260 k in
m
2z
IS RAL2005001 1707
IS RAL2005001 1707
using f = kGTG d .
1 0 0 0 0 0 2.83 in 2.83 in
0 0.894 0.447 0 0 0 0.0295 rad 0.0188 rad
0 0.447 0.894 0 0 0 0.0169 rad 0.0283 rad

TG d = =
0 0 0 1 0 0 0 0
0 0 0 0 0.894 0.447 0 0
0 0 0 0 0.447 0.894 0 0

7.45 0 1,000 7.45 0
1,000 2.83 in
0 4,920 0 0 4,920
0 0.0188 rad
1,000 0 179,000 1,000 0 89,500 0.0283 rad

f( 2
) = kTd = 1,000 0
7.45 0 1,000 7.45 0 0 0
0 4,920 0 0 4,920
179,000 0
1,000 0 89,500 1,000 0
f 7.23 k
1y
92.5 k in
m1x
2,240 k in
m1z
=
7.23 k
f
92.5 k in
3y

m3
295 k in
x

m
3z
IS RAL2005001 1707
IS RAL2005001
1707
using f = kGTG d .
1 0 0 0 0 0 2.83 in 2.83 in
0 0 1 0 0 0 0.0295 rad 0.0169 rad
0 1 0 0 0 0 0.0169 rad 0.0295 rad

TG d = =
0 0 0 1 0 0 0 0
0 0 0 0 0 1 0 0
0 0 0 0 1 0 0 0
83.3 0 5,000 83.3 0

5,000 2.83 in
0 11,000 0 0 11,000 0 0.0169 rad
5,000 0 400,000 5,000 0

200,000 0.0295 rad
f( 3 ) = kTd =
83.3 0 5,000 83.3 0
5,000 0
0 11,000 0 0 11,000
0 0
5,000 0 200,000 5,000 0 400,000 0
f 88.1 k
1y

m1x 186 k in

m 2,340 k in
1z =
f4y 88.1 k
186 k in

m4 x
8,240 k in
m
4z
IS RAL2005001 1707
IS RAL2005001 1707
To check the equilibrium of node 1 the local forces and moments for each element need to be transformed to global
coordinates. Recall, that:
f =Tf f =T T f T T =T 1
Since we are only checking the forces and moments at node 1, we need only the upper-left-hand portion of the transformation matrix
TG.
Therefore; for Element 1:
f1y 1 0 0 19.2 k 19.2 k
m1x = 0 0.894 0.447 167 k in = 1,260 k in
m1z
0
0.447 0.894 2,480 k in 2,150 k in
IS RAL2005001 1707
f1y 1 0 0 7.23 k 7.23 k
m1x = 0 0.894 0.447 92.5 k in = 1,080 k in
m1z
0
0.447 0.894 2,240 k in 1,960 k in
f1y 1 0 0 88.1 k 88.1 k
m1x = 0 0 1 2,340 k in = 2,340 k in
m1z
0
1 0 186 k in 186 k in
The forces and moments that are applied to node 1 by each element are equal in magnitude and opposite direction. Therefore the sum
of the forces and moments acting on node 1 are:
F1y = 100 7.23 +19.2 + 88.1 = 0.07 k
M1x = 1,260 1,080 + 2,340 = 0.0 k in
M1z = 2,150 +1,060 +186 = 4.0 k in
The forces and moments accurately satisfy equilibrium considering the amount of truncation error inherent in results of the calculations
presented in this example.
IS RAL2005001 1707
Grid Example
The frame is fixed at nodes 1 and 3, and is subjected to a load of 22 kN applied at node 2. Assume I = 16.6 x 10-5 m4, J = 4.6 x 10-5 m4, G =
84 GPa, and E = 210 GPa for all elements.
To facilitate a timely solution, the boundary conditions at nodes 1 and 3 are applied to the local stiffness matrices at the
beginning of the solution.
d1y = 1x = 1z = 0 d3 y = 3 x
= 3 z = 0
Beam Element 1: the local x axis coincides the global x axis
2 x1
x 3 z2 z1 0
C = cos = = =1 S = sin = = =0
L(1) 3 L(1) 3
where
12EI = 12(210 106 )(16.6 105 ) =1.55 104 kN / m
L3 (3)3
6EI = 6(210 106 )(16.6 105 ) = 2.32 104 kN
L2 (3)2
4EI = 4(210 106 )(16.6 105 ) = 4.65 104 kNm
L 3
GJ =(84 106 )(4.6 105 ) = 0.128 104 kNm
L 3
IS RAL2005001 1707
The global stiffness matrix for element 1, considering only the parts associated with node 2, may be obtained from the following
relationship:
kG =TG kGTG
1 0 0 1.55 0 2.32 1 0 0
(1) 4
k =10 0 1 0 0 0.128 0 0 1 0 kN
m
0 0 1 2.32 0 4.65 0 0 1
d2y 2x 2z
1.55 0 2.32
k(1)
= 104 0 0.128 0 kN
m
2.32 0 4.65

Beam Element 2: the local x axis is located from node 2 to node 3
x3 x2 z3 z2 3
C = cos = =0 =0 S = sinq = = = 1
L( 2) 3 L(1) 3
The global stiffness matrix for element 2, considering only the parts associated with node 2, may be obtained using:
kG =TG kGTG
1 0 0 1.55 0 2.32 1 0 0
( 2) 4
k = 10 0 0 1 0 0.128 0 0 0 1 kN
m
0 1 0 2.32 0 4.65 0 1 0
d2y 2x 2z
1.55 2.32 0
k( 2) = 104 2.32 4.65 0 kN
m
0 0 0.128
IS RAL2005001 1707
Superimposing the two elemental stiffness matrices gives:
d 2y 2 x 2 z
3.10 2.32 2.32
K = 10 4 2.32 4.78 0 kN
m
2.32 0 4.78
F2y = 22 kN 3.10 2.32 2.32 d2y
M =0 =104 2.32 4.78 0

2x 2x
2 x
M2z =0 2.32 0 4.78
d2y 0.00259 m
2x = 0.00126 rad
2z
0.00126 rad
using f = k GT Gd .
1 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0
0 0 1 0 0 0 0 0
TGd = =
0 0 0 1 0 0 0.00259 m 0.00259 m
0 0 0 0 1 0 0.00126 rad 0.00126 rad
0 0 0 0 0 1 0.00126 rad 0.00126 rad

IS RAL2005001 1707
IS RAL2005001 1707

1.55 0 2.32 1.55 0
2.32 0
0 0.128 0 0 0.128
0 0
2.33 0
4 2.32 0 4.65 2.32 0
f(1) = kTd = 10
1.55 0 2.32 1.55 0 2.32 0.00259 m
0 0.128 0 0 0.128 0 0.00126 rad
2.32 0 2.33 2.32 0
4.65 0.00126 rad

f 11.0 kN
1y

m1x 1 .50 kN m
m 31.0 kN m
1z =
f2y 11.0 kN
1.50 kN m

m2x
1.50 kN m
m
2z
using f = k GT Gd .
1 0 0 0 0 0 0.00259 m 0.00259 m
0 0 1 0 0 0 0.00126 rad 0.00126 rad
0 1 0 0 0 0 0.00126 rad 0.00126 rad

TGd = =
0 0 0 1 0 0 0 0
0 0 0 0 0 1 0 0
0 0 0 0 1 0 0 0
IS RAL2005001 1707
IS RAL2005001 1707

1.55 0 2.32 1.55 0
2.32 0.00259 m
0 0.128 0 0 0.128
0 0.00126 rad
2.33 0.00126 rad
4 2.32 0 4.65 2.32 0
f( 2) = kTd =10
1.55 0 2.32 1.55 0 2.32 0
0 0.128 0 0 0.128 0 0
2.32 0 2.33 2.32 0 4.65 0
11.0 kN f2y
1.50 kN m m

1 .50 kN m = 2
11.0 kN
x
1 .50 kN m
31.0 kN m m

2
z
3
y
m

3
z
IS RAL2005001 1707
CIVL 7117 Finite Elements Methods in Structural Mechanics Page 233

IS RAL2005001 1707
Beam Element Arbitrarily Oriented in Space
In this section we will develop a beam element that is arbitrarily oriented in three-dimensions. This element can be used to
analyze three-dimensional frames. Let consider bending about axes, as shown below.
The axis y is the principle axis for which the moment of inertia is minimum, Iy. The right-hand rule is used to establish the z axis and
the maximum moment of inertia, Iz.
Bending in the x z plane: The bending in the x z plane is defined by my . The stiffness matrix for bending the in the x-z plane
is:
12L 6L2 12L 6L2

2 3 2 3
EIy 6L 4L 6L 2L
kY = L4 12L 6L2 12L 6L2
6L2 2L3 6L2 4L3
where Iy is the moment of inertia about the y axis (the weak axis).

IS RAL2005001 1707

Bending in the x y plane: The bending in the x y plane is defined by mz .
The stiffness matrix for bending the in the x y plane is:
12L 6L2 12L 6L2

2 3 2 3
EIz 6L 4L 6L 2L
kz = L4 12L 6L2 12L 6L2
6L2 2L3 6L2 4L3
where Iz is the moment of inertia about the z axis (the strong axis).
Direct superposition of the bending stiffness matrices with the effects of axial forces and torsional rotation give:

d1x d1y d1z 1x 1y 1z d 2x d 2y d 2z 2x 2y 2z

IS RAL2005001 1707
The global stiffness matrix may be obtained using:
T
k =T kT
where
3 x3
3 x3
T=
3 x3
3 x3
where
Cxx Cyx Czx
= C C yy C zy
3x3 xy
C C C
yz
xz zz
where the direction cosines, Ci j , are defined as shown below
The direction cosines of the x axis are:
x = cos x x i + cos y x j + cos zx k
where
x2 x 1 y2 y 1 z2 z1
cosxx = =l cosyx = =m coszx = =n
L L L
IS RAL2005001 1707

IS RAL2005001 1707
The y axis is selected to be perpendicular to the x and the z axes is such a way that the cross product of global z with x results in the y
axis as shown in the figure below.
z x = y
i j k
1 m l
z x = y = 0
0 1 = i + j
D l D D
m n
where
D= l 2 + m2
The z axis is determined by the condition that z = x y
1 i j k ln mn
z = x y = l m n = i j + Dk
D m l 0 D D
Therefore, the transformation matrix
Cxx
Cyx Czx
= C C C zy
3x xy yy
3
C C C
xz yz zz
becomes

IS RAL2005001 1707
l m n
l
=m 0
3x
3
D D
ln mn
D
D D
There are two exceptions that arise when using the above expressions for mapping the local coordinates to the global system: (1) when
the positive x coincides with z ; and (2) when the positive x is in the opposite direction as z. For the first case, it is assumed that y is y.
0 0 1
= 0 0 0
1 0 0
In case two, it is assumed that y is y.
0 0 1
= 0 0 0
1 0 0

IS RAL2005001 1707
If the effects of axial force, both shear forces, twisting moment, and both bending moments are considered, the stiffness matrix for a
frame element is:

IS RAL2005001 1707
In this case the symbol are:
12EIy 12EIz
y = 2 z = 2
GA L GA L
s s
where As is the effective beam cross- section in shear. Recall the shear modulus of elasticity or the modulus of rigidity, G, is related to
the modulus of elasticity and the Poissons ratio, as:
G= 2(1E+ )
If y and z are set to zero, the stiffness matrix reduces the that shown previously on page 235. This is the form of the stiffness matrix
used by SAP2000 for its frame element.

IS RAL2005001 1707
Example Frame Application
A bus subjected to a static roof-crush analysis. In this model 599 frame elements and 357 nodes are used.
Concept of Substructure Analysis
Sometimes structures are too large to be analyzed as a single system or treated as a whole; that is, the final stiffness matrix and
equations for solution exceed the memory capacity of the computer. A procedure to overcome this problem is to separate the whole
structure into smaller units called substructures. For example, the space frame of an airplane, as shown below, may require thousands of
nodes and elements to completely model and describe the response of the whole structure. If we separate the aircraft into
substructures,

IS RAL2005001 1707
such as parts of the fuselage or body, wing sections, etc., as shown below, then we can solve the problem more readily and on
computers with limited memory.

IS RAL2005001 1707
Problems
14. Do problems 5.3, 5.8, 5.13, 5.28, 5.41, and 5.43 on pages 240 - 263 in your textbook A First Course in the Finite Element
Method by D. Logan.
15. Do problems 5.23, 5.25, 5.35, 5.39, and 5.55 on pages 240 - 263 in your textbook A First Course in the Finite Element
Method by D. Logan. You may use the SAP 2000 to do frame analysis.

Asa

Diunggah oleh

Informasi Dokumen

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Asa

Diunggah oleh

Hak Cipta:

Format Tersedia

IS RAL2005001 1707

Plane Frame and Grid Equations

Two-Dimensional Arbitrarily Oriented Beam Element

For a beam we will define the following as the transformation matrix:

12SC 12C2 6LC 12SC 12C2 6LC

6 LS 6LC 2L2 6LS 6LC 4L2

Recall the simple axial deformation, define in the spring element:

0 12C2 6LC2 0 12C2 6LC2

0 6LC2 4C2L2 0 6LC2 2C2L2

0 12C2 6LC2 0 12C2 6LC2

0 6LC2 2C2L2 0 6LC2 4C2L2

where the transformation matrix, including axial effects is:

Rigid Plane Frame Example

Consider the frame shown in the figure below.

12I 12( 200) 2 6I 6(200) 3

L2 = (120)2 = 0.167 in L = 120 =10.0 in

Therefore, for element 1:

k(1) = 250,000 10 0 800 10 0 400 lb

12I 12(100) 2 6I 6(100) 3

L2 = (120)2 = 0.0835 in L = 120 = 5.0 in

Therefore, for element 2:

k( 2) = 250,000 0 5 400 0 5 200 lb

12I 12( 200) 2 6I 6(200) 3

L2 = (120)2 = 0.167 in L= 120 =10.0 in

Therefore, for element 3:

k (3 ) = 250,000 10 0 800 10 0 400 lb

The boundary conditions for this problem are:

5,000 0 5 200 10 5 1200 3

Solving the above equations gives:

Element 1: The element force-displacement equations can be obtained using

f = kTd . Therefore, Td is:

0 0 0 0 0 1 2 = 0.00153 rad 0.00153 rad

0 6LC2 4C2L2 0 6LC2 2C2L2

0 12C2 6LC2 0 12C2 6LC2 4C2L2

0 6LC2 2C2L2 0 6LC2

1 0 0 0 0 d2y = 0.00148 in 0.00148 in

( 2) 200 0.00153 rad

Simplifying the above equations gives:

Element 3: The element force-displacement equations are:

Therefore, the local force-displacement equations are:

Simplifying the above equations gives:

Rigid Plane Frame Example

Consider the frame shown in the figure below.

m2 = m3 = wL2 = (1,000)402 = 133,33 lb ft = 1,600 k in 12 12

6I = 6(1,000) =11.78551 in3

Therefore, for element 1:

50.02 49.98 8.33

8.33 8.33 4000

Simplifying the above equation:

491 491 235,700

Element 2: The angle between x and x is 0

L = 480 = 62.5 lb / in L2 = (12 40)2 = 0.0521 in

Therefore, for element 2:

k ( 2) = 62.50 0 0.052 12.5 lb

The global beam equations reduce to:

0 9,198 2,945 491 d 2x

20 k = 2,945 2,951 290 d 2y

Element 1: The element force-displacement equations can be obtained using

f = kTd . Therefore, Td is:

0 0 0 0 0 1 0.0033 rad 0.0033 rad

C1 = AE = (100)30 106 = 5,893 k in C2 = EI 30 106 (1,000)

Therefore, the local force-displacement equations are:

0 2.730 694.8 0 2.730 694.8 0

10 694.8 117,900 0 694.8 117,000 0