Jimma University Jimma Institute of Technology Department of Civil and Environmental Engineering

Jimma University
Jimma Institute of Technology

Department of Civil and Environmental Engineering
Course Tittle: Theory of Structures-II Course No: CENG-2093
Credit hour- 3 Year-II semester-II
Academic year: - 2019/20 Instructor: -
Course Objective
 This course provides an introduction to the analysis of indeterminate structural systems
common in Civil Engineering.
Learning Outcomes:
Up on successful completion of the course, students will be able to:
 Identify, formulate, and solve support reactions of trusses, beams, and frames.
 Apply the displacement method to analyze statically indeterminate beams and frames.
 Use approximate methods to evaluate the statically indeterminate structural responses.
 Employ the stiffness method to solve complex trusses, beams, and frames.
 Analyze indeterminate structures using structural analysis soft-wares.
Course Outline
Chapter 1: Analysis of indeterminate Chapter 2: Influence lines for Indeterminate
structures using Structures.
1.1 Displacement Method 2.1 Beams and Frames
a. Kinematic indeterminacy 2.2 Truss and Arches
b. Slope deflection Method Chapter 3: Introduction to Matrix Methods
c. Moment Distribution Method
3.1 Stiffness Method
d. Kani’s Method
3.2 Flexibility Method
1.2 Approximate Methods
3.3 Computer application of Matrix Analysis
References
1. Hibbler, R. C. Structural Analysis, 6th& 8th Edition, Prentice Hall, 2005.
2. Theory of Structures by Aslam Kassimali, 4th Edition.
3. Nigussie Tebedge, Methods of Structural Analysis, 1983, AAU
4. M.S. Williams, Structures: Theory and Analysis, Palgrave Macmillan., 1999
5. Leet, M., et al. Fundamentals of Structural Analysis, 2nd Edition, McGraw Hill, 2004.
I
THEORY OF STRUCTURES-II 2019/20
Chapter One
Analysis of indeterminate structures
1.1 Introduction
As discussed previously, the support reactions and internal forces of statically determinate
structures can be determined from the equations of equilibrium (including equations of condition,
if any). However, since indeterminate structures have more support reactions and/or members
than required for static stability, the equilibrium equations alone are not sufficient for determining
the reactions and internal forces of such structures, and must be supplemented by additional
relationships based on the geometry of deformation of structures.
These additional relationships, which are termed the compatibility conditions, ensure that the
continuity of the displacements is maintained throughout the structure and that the structure’s
various parts fit together. For example, at a rigid joint the deflections and rotations of all the
members meeting at the joint must be the same. Thus the analysis of an indeterminate structure
involves, in addition to the dimensions and arrangement of members of the structure, its cross-
sectional and material properties (such as cross sectional areas, moments of inertia, moduli of
elasticity, etc.), which in turn, depend on the internal forces of the structure.
Advantages and disadvantages of indeterminate structures
The advantages of statically indeterminate structures over determinate structures include the
following.
1. Smaller Stresses: - The maximum stresses in statically indeterminate structures are
generally lower than those in comparable determinate structures.
2. Greater Stiffnesses: - Statically indeterminate structures generally have higher stiffnesses
(i.e., smaller deformations), than those of comparable determinate structures.
3. Redundancies: - Statically indeterminate structures, if properly designed, have the
capacity for redistributing loads when certain structural portions become overstressed or
collapse in cases of overloads due to earthquakes, tornadoes, impact (e.g., gas explosions
or vehicle impacts), and other such events. Indeterminate structures have more members
and/or support reactions than required for static stability, so if a part (or member or support)
of such a structure fails, the entire structure will not necessarily collapse, and the loads will
be redistributed to the adjacent portions of the structure.
JU, JIT, DEPT OF CIVIL ENG Year-2 Page 1

The main disadvantages of statically indeterminate structures, over determinate structures, are the
following.
1. Stresses Due to Support Settlements: - Support settlements do not cause any stresses in
determinate structures; they may, however, induce significant stresses in indeterminate
structures, which should be taken into account when designing indeterminate structures.
2. Stresses Due to Temperature Changes and Fabrication Errors: - Like support
settlements, these effects do not cause stresses in determinate structures but may induce
significant stresses in indeterminate ones.
Methods of Analysis
In the case of indeterminate structures either the reactions or the internal forces cannot be
determined from equations of statics alone. In such structures, the number of reactions or the
number of internal forces exceeds the number of static equilibrium equations. In addition to
equilibrium equations, compatibility equations are used to evaluate the unknown reactions and
internal forces in statically indeterminate structure. In the analysis of indeterminate structure it is
necessary to satisfy the equilibrium equations (implying that the structure is in equilibrium)
compatibility equation (requirement if for assuring the continuity of the structure without any
breaks) and force displacement equations (the way in which displacement are related to forces).
We have two distinct method of analysis for statically indeterminate structure depending on how
the above equations are satisfied:
1. Force method
2. Displacement method
Force Method
This method was one of the first available for the analysis of statically indeterminate structures. In
the force method of analysis, primary unknown are forces. In this method compatibility equations
are written for displacement and rotations (which are calculated by force displacement equations).
Solving these equations, redundant forces are calculated. Once the redundant forces are calculated,
the remaining reactions are evaluated by equations of equilibrium.
Displacement method
In the displacement method of analysis, the primary unknowns are the displacement. In this
method, first force-displacement relations are computed and subsequently equations are written
satisfying the equilibrium conditions of the structure. After determining the unknown

displacements, the other forces are calculated satisfying the compatibility conditions and force
displacement relations.
Method Unknowns Equations Used for Solution Coefficients of the
Unknowns
Forces Compatibility and Force
Force Method Flexibility Coefficients
Displacement
Displacement Equilibrium and Force
Displacements Stiffness Coefficients
Method Displacement
In this chapter, the displacement method of analysis will be discussed.

1.2 Displacement Method
All structures must satisfy equilibrium, load-displacement, and compatibility of displacements
requirements in order to ensure their safety. The displacement method works by satisfying
equilibrium equations for the structure. To do this the unknown displacements are written in terms
of the loads by using the load-displacement relations, then these equations are solved for the
displacements. Once the displacements are obtained, the unknown loads are determined from the
compatibility equations using the load-displacement relations. Every displacement method
follows this general procedure.
In the discussion that follows we will show how to identify the unknown displacements in a
structure and we will develop some of the important load-displacement relations for beam and
frame members.
a) Kinematic indeterminacy
When a structure is loaded, specified points on it, called nodes, will undergo unknown
displacements. These displacements are referred to as the degrees of freedom for the structure,
and in the displacement method of analysis it is important to specify these degrees of freedom
since they become the unknowns when the method is applied. The number of these unknowns is
referred to as the degree in which the structure is kinematically indeterminate.
To determine the kinematic indeterminacy we can imagine the structure to consist of a series of
members connected to nodes, which are usually located at joints, supports, at the ends of a member,
or where the members have a sudden change in cross section.
To clarify these concepts we will consider some examples, beginning with the beam in Fig. 1.1a.
Here any load P applied to the beam will cause node A only to rotate (neglecting axial
deformation), while node B is completely restricted from moving. Hence the beam has only one

unknown degree of freedom, θA and is therefore kinematically indeterminate to the first degree.
The beam in Fig. 1.1b has nodes at A, B, and C, and so has four degrees of freedom, designated by
the rotational displacements, θA, θB , θC and the vertical displacement, Δ It is kinematically
indeterminate to the fourth degree.
Consider now the frame in Fig.1.1c. Again, if we neglect axial deformation of the members, an
arbitrary loading P applied to the frame can cause nodes B and C to rotate, and these nodes can be
displaced horizontally by an equal amount. The frame therefore has three degrees of freedom, θB,
θC , ΔB and thus it is kinematically indeterminate to the third degree.
1.3 Slope-Deflection Equations

The slope-deflection method is so named since it relates the unknown slopes and deflections to the
applied load on a structure. In order to develop the general form of the slope-deflection equations,
we will consider the typical span AB of a continuous beam as shown in Fig.1.2, which is subjected
to the arbitrary loading and has a constant EI. We wish to relate the beam’s internal end moments
MAB and MBA in terms of its three degrees of freedom, namely, its angular displacements θA and
θB, and linear displacement Δ which could be caused by a relative settlement between the supports.
Since we will be developing a formula, moments and angular displacements will be considered
positive when they act clockwise on the span, as shown in Fig. 1.2. Furthermore, the linear
displacement Δ is considered positive as shown, since this displacement causes the cord of the
span and the span’s cord angle ψ to rotate clockwise.
EI is constant
positive sign convention

1. Angular Displacement at A, θA: - Consider node A of the member shown in Fig.1.3a to rotate
θA while its far-end node B is held fixed. To determine the moment MAB needed to cause this
displacement, we will use the conjugate-beam method. For this case the conjugate beam is
shown in Fig. 1.3b. Notice that the end shear at A’ acts downward on the beam, since  A is
clockwise. The deflection of the “real beam” in Fig.1.3a is to be zero at A and B, and therefore
the corresponding sum of the moments at each end A’ and B’ of the conjugate beam must also
be zero. This yields
 1  M AB   L  1  M BA   2 L
M A' 0  2  EI  L  3   2  EI  L  3  0
       
 1  M BA   L  1  M AB   2 L
M B' 0  2  EI  L  3   2  EI  L  3   A L  0
       
From which we obtain the following load-displacement relationships
4 EI 2 EI
M AB   A .................................1 M BA   A..........................2
L L
2. Angular Displacement at B, θB In a similar manner, if end B of the beam rotates to its final
position θB while end A is held fixed, Fig.1.4, we can relate the applied moment MBA to the
angular displacement θB and the reaction moment MAB at the wall. The results are
4 EI
M BA   B ................3
L
2 EI
M AB   B ...............4
L
3. Relative Linear Displacement, Δ If the far node B of the member is displaced relative to A, so
that the cord of the member rotates clockwise (positive displacement) and yet both ends do not
rotate, then equal but opposite moment and shear reactions are developed in the member, Fig.5a.
as before, the moment M can be related to the displacement Δ using the conjugate-beam
method. In this case, the conjugate beam, Fig.5b, is free at both ends, since the real beam
(member) is fixed support. However, due to the displacement of the real beam at B, the moment

at the end B’ of the conjugate beam must have a magnitude of Δ as indicated.* Summing
moments about B’ we have
1 M  2   1 M  1 
M B' 0  2 EI ( L)  3 L     2 EI ( L)  3 L      0
     
6 EI
M AB  M BA  M  2  .......................5
L
By our sign convention, this induced moment is negative since for equilibrium it acts
counterclockwise on the member.
4. Fixed-End Moments
When a beam whose supports are completely fixed against rotation or translation is subjected to
transverse loads the beam is called a fixed-end beam. The bending moments at the suppotrs of
such a beam are called fixed-end moments.
M AB  ( FEM ) AB .....................6a
M BA  ( FEM ) BA ......................6b
If the end moments due to each displacement (Eqs. 1 through 5) and the loading (Eq. 6a and 6b)
are added together, the resultant moments at the ends can be written as
 I    
M AB  2 E    2 A   B  3     ( FEM ) AB
 L   L 
............................ 7
 I    
M BA  2 E    2 B   A  3     ( FEM ) BA
 L   L 
Since these two equations are similar, the result can be expressed as a single equation. Referring
to one end of the span as the near end (N) and the other end as the far end (F), and letting the
member stiffness be represented as k= I/L and the span’s cord rotation as    / L we can write
M N  2Ek (2 N   F  3 )  ( FEM ) N .........................................8

For Internal Span or End Span with Far End Fixed

E.g. 1.1 Draw the shear and moment diagrams for the beam shown in fig. 1.7a. Assume the support
at A and C are fixed, at B is roller. EI is constant.
Solution: -
Degrees of Freedom. From Fig. 1.7(b), we can see that only joint B of the beam is free to rotate.
Thus, the structure has only one degree of freedom, which is the unknown joint rotation,  B
Slope-Deflection Equations. Two spans must be considered in this problem. Since there is no
span having the far end pinned or roller supported, Eq. 8 applies to the solution. Using the formulas
for the FEMs tabulated for the triangular and rectangular loading given, we have
 wL2 (4 kN / m)(15m) 2
( FEM ) AB    30 kNm
30 30
wL2 (4 kN / m)(15m) 2
( FEM ) BA    45 kNm
20 20
 wL2 (4 kN / m)(12m) 2
( FEM ) BC    48 kNm
12 12
wL2 (4 kN / m)(12m) 2
( FEM )CB    48 kNm
12 12
In order to identify the unknowns, the elastic curve for the beam is shown in fig. 1.7b below. As
indicated, there are four unknown internal moments. Only the slope at B,  B , is unknown. Since A
and C are fixed supports,  A  C  0. Also, since the supports do not settle, nor are they displaced
up or down,  AB   BC  0.
For span AB, considering A to be the near end and B to be the far end, we have

I
M AB  2 E (2 A   B  3 AB )  ( FEM ) AB
L
 I  2 EI
M AB  2 E    2(0)   B  3(0)   30   B  30 .....................1
 15  15
Now, considering B to be the near end and A to be the far end, we have
I
M BA  2 E (2 B   A  3 AB )  ( FEM ) BA
L
 I  4 EI
M BA  2 E    2 B  0  3(0)  45   B  45 .....................2
 15  15
In a similar manner, for span BC
I
(2 B  C  3 BC )  ( FEM ) BC
M BC  2 E
L
 I  EI
M BC  2 E    2 B  0  3(0)  48   B  48 .....................3
 12  3
I
M CB  2 E (2C   B  3 BC )  ( FEM )CB
L
 I  EI
M CB  2 E    2(0)   B  3(0)  48   B  48 .....................4
 12  6
Equilibrium Equations. The above four equations contain five unknowns. The necessary fifth
equation comes from the condition of moment equilibrium at support B. the free-body diagram of
a segment of the beam at B is shown in Fig. 1.7c. Here MAB and MBC are assumed to act in the
positive direction to be consistent with the slope-deflection equations.
M B 0 M BA  M BC  0 ........................5
To solve, substitute Eqs. (2) and (3) into Eq. (5)
4 EI EI 5
 B  45   B  48  0   B 
15 3 EI
Resubstituting this value into Eqs. (1)- (4) yields
2EI 2EI  5 
M AB   B  30     30  29.33 kNm
15 15  EI 
4 EI 4 EI  5 
M BA   B  45     45  46.33 kNm
15 15  EI 
EI EI  5 
M BC   B  48     48  46.33 kNm
3 3  EI 
EI EI  5 
M CB   B  48     48  48.83 kNm
6 6  EI 

The negative value for MAB and MBC indicates that these moments acts counter-clockwise on the
beam.
Shear at End span: using the above results, shear force at each end spans are determined from
the equilibrium equations.
1 2 
M A 0  ByL (15m)  29.33  46.33  (4 kN / m)(15m)  (15m)   0
2 3 
ByL  21.13 kN 
F y 0
1
Ay  ByL  (4 kNm)(15m)  0
2
Ay  8.87 kN 
1
M B 0  C y (12m)  46.33  48.83  (4 kN / m)(12m)( )(12m)  0
2
C y  24.21 kN 
F y 0 ByR  C y  (4 kNm)(12m)  0
ByR  23.79 kN 
Reaction force at B.
By  ByL  ByR
By  21.13 kN  23.79 kN  44.92 kN 

Shear and Moment diagrams
E.g. 1.2 Determine the Member End Moments and The Support Reactions for the Beam Loaded
as Shown in figure 1.8a. The Supports at A and B Settles By 20mm and 10mm
Respectively. E  200Gpa and I  600 106 mm4 . Use the Slope-Deflection Method.
Solution: - Fixed-End Moments. Using the formulas for the FEMs table, we have
 wL2 (15)(3m) 2
( FEM ) AB    11.25 kNm
12 12
wL2 (15)(3m)2
( FEM ) BA    11.25 kNm
12 12
 Pa 2b (20* 42 *1)  (20* 22 *3)

( FEM ) BC    22.4 kNm
L2 52

Pa 2b (20*4*12 )  (20*2*32 )
( FEM )CB  2   17.6 kNm
L 52
7 wL2 7(30)(4m)2
( FEM )CD    28 kNm
120 120
Slope-Deflection Equations.
 0.01
 AB  AB 
LAB 3
 BC 0.01
 BC  
LBC 5
I
M AB  2 E (2 A   B  3 AB )  ( FEM ) AB
L
 600 *106    0.01  
M AB  2(200 *106 )    2(0)   B  3     11.25
 3   3 
M AB  (8 *104 ) B  788.75 .....................1
 600 *106    0.01  
M BA  2(200 *106 )    2 B  0  3     11.25
 3   3 
M BA  (16 *104 ) B  811.25 .....................2
 600 *106    0.01  
M BC  2(200 *106 )    2 B  C  3     22.4
 5    5 
M BC  (9.6 *104 ) B  (4.8 *104 C )  265.6 .....................3
 600 *106    0.01  
M CB  2(200 *106 )    2C   B  3     17.6
 5   5 
M CB  (4.8 *104 ) B  (9.6 *104 )C  305.6 .....................4
 600 *106 
M CD  3(200 *106 )   C  3(0)   28
 4 
M CD  (9 *10 )C  28 .....................5
4
Equilibrium Equations
JOINT B M B 0 M BA  M BC  0 ........................6
(16*104 ) B  811.25  (9.6*104 ) B  (4.8*10 4 C )  265.6  0
(25.6*104 ) B  (4.8*104 )C  1076.85
M C 0 M CB  M CD  0 ........................7
(4.8 *104 ) B  (9.6 *104 )C  305.6  (9 *104 )C  28  0
(4.8 *104 ) B  (18.6 *104 )C  277.6

By the result of Eq. (6) and Eq. (7) solve  B and C
(25.6 *104 ) B  (4.8 *10 4 )C  1076.85

(4.8 *104 ) B  (18.6 *104 )C  277.6
 B  0.004126 and C  0.000428
Member End moments
M AB  (8 *10 ) B  788.75  (8 *104 )(0.004126)  788.75  458.67 kNm
4
M BA  (16 *104 ) B  811.25  (16 *104 )(0.004126)  811.25  151.09 kNm

M BC  (9.6 *104 ) B  (4.8 *104 C )  265.6
M BC  (9.6 *104 )(0.004126)  (4.8 *104 )(0.000428)  265.6  151.04 kNm
M CB  (4.8 *104 ) B  (9.6 *104 )C  305.6
M CB  (4.8 *104 )(0.004126)  (9.6 *104 )(0.000428)  305.6  66.464 kNm
M CD  (9 *104 )C  28  (9 *104 )(0.000428)  28  66.52 kNm
M DC  0
Shear at End span
3m
M A  0  3( ByL )  458.67  151.09  15(3m)(
2
)0
 ByL  225.75kN 
F y  0  Ay  ByL  15(3m)  0  Ay  180.75 kN 
M B  0  5(C yL )  66.464  151.04  20(3)  20(1)  0

 C yL  0.9152 
F y  0  ByR  C yL  20  20  0
 C yL  40.9152 kN 
1 2
M  0  4( Dy )  66.52  (30)(4m)( * 4m)  0
C
2 3
 Dy  23.37 
1
F y  0  C yR  Dy  (30)(4m)  0
2
 C yR  36.63 kN 
By  ByL  ByR  225.75 kN  40.9152 kN  266.6652 kN

Therefore,
C y  C yL  C yR  0.9152 kN  36.63 kN  35.7148 kN

Pin-Supported End Span.

When this occurs, the moment at the roller or pin must be zero; and provided the angular
displacement θB at this support does not have to be determined, we can modify the general slope-
deflection equation so that it has to be applied only once to the span rather than twice. To do this
we will apply Eq. 8 to each end of the beam in fig bellow. This results in the following two
equations:
M N  2Ek (2 N   F  3 )  ( FEM ) N
0  2Ek (2 F   N  3 )  0
Here the ( FEM ) N is equal to zero since the far end is pinned in Fig.1.9b. Furthermore, the
( FEM ) N can be obtained, by modified fixed end moment. Multiplying the first equation by 2 and
subtracting the second equation from it eliminates the unknown  F and yields
M N  3Ek ( N  )  ( FEM ) N ........................................................9

only for End Span with Far End pinned or Roller Supported

E.g. 1.3 Determine the member end moments and reactions for the beam shown in Fig. 1.10a due
to applied load and due to the support settlements of 18 mm at B, and 36 mm at C. Use
the slope-deflection method. E= 200 Gpa and I= 15*106 mm4
Fig. 1.10a
Solution: -
Slope-Deflection Equations. Two spans must be considered in this problem. Eq. 8 applies to span
AB. we can use Eq. 9 for span BC since C is on a roller support.
Using the formulas for the FEMs table, we have
 Pb 2 a (90kN )(2m) 2 (4m)
( FEM ) AB    40 kNm
L2 (6m) 2
Pa 2b (90kN )(4m) 2 (2m)
( FEM ) BA    80 kNm
L2 (6m) 2
3PL 3(50kN )(6m)
( FEM ) BC    56.25kNm
16 16
Chord Rotations:
It can be seen from Fig. 1.10b that since support A does not settle but support B settles by 18 mm
and support C settles by 36mm. the relative settlement between the two ends of member AB is 18
mm= 0.018m and the relative settlement between the two ends of member BC is 18 mm= 0.018m.
0.018m
Rotation of the chord of member AB is  AB   0.003
6m
0.036m  0.018m
Rotation of the chord of member BC is  BC   0.003
6m
Slope-Deflection Equations
Applying Eq. 8 for span AB and realizing that  A  0 , we have

I
M AB  2 E (2 A   B  3 AB )  ( FEM ) AB
L
 15*106 
 2(200*106 )    2(0)   B  3(0.003)   40  10  B  9  40 .....................1
3
M AB
 6 
I
M BA  2 E (2 B   A  3 AB )  ( FEM ) BA
L
 15*106 
M BA  2(200*106 )    2 B  0  3(0.003)   80  (2*10 ) B  9  80 .....................2
3
 6 
Applying Eq. 9 with B as the near end and C as the far end, we have
M N  3Ek ( N  )  ( FEM ) N
I
M BC  3E   ( B  BC )  ( FEM ) BC
L
6  15*10 
6
M BC  3(200*10 )   ( B  0.003)  56.25  (1.5*10 ) B  4.5  56.25 .......................3
3
 6 
Equilibrium Equations.
The above three equations contain four unknowns. The necessary fourth equation comes from the
conditions of equilibrium at the support B. The free-body diagram is shown in Fig. 1.10c.We have
M B 0 M BA  M BC  0 ........................4
To solve, substitute Eqs. (2) and (3) into Eq. (4), which yields
(2*103 ) B  9  80  (1.5*103 ) B  4.5  56.25  0
3.5*103 B  10.25   B  0.00293 rad
Substituting  B into Eqs. (1)– (3), we get

M AB  103 B  9  40  103 ( 0.00293)  9  40  51.93 kNm
M BA  (2*103 ) B  9  80  (2*103 )( 0.00293)  9  80  65.14 kNm
M BC  (1.5*103 ) B  4.5  56.25  (1.5*103 )( 0.00293)  4.5  56.25  65.14 kNm
Structures with Cantilever Overhangs

Consider a continuous beam with a cantilever overhang, as shown in Fig. 1.11a. Since the
cantilever portion CD of the beam is statically determinate in the sense that the shear and moment
at its end C can be obtained by applying the equations of equilibrium Fig. 1.11b, it is not necessary
to include this portion in the analysis. Thus, for the purpose of analysis, the cantilever portion CD
can be removed from the structure, provided that the moment and the force exerted by the

cantilever on the remaining structure are included in the analysis. The indeterminate beam, which
needs to be analyzed, is shown in Fig. 1.11c
E.g. 1.4 Determine the moment at B and C for the beam shown in Fig. 1.12a by using the slope-
deflection method. EI is constant.
Solution: Only one span (BC) must be considered in this problem since the moment M BA due to
the overhang can be calculated from statics.
1 
M B  0 M BA  (50 kN / m)(3m)  (3m) 
2 
M BA  225 kNm
F y 0 ByL  (50 kN / m)(3m)
ByL  150 kN 
Statically indeterminate beam to be analyzed

Due to internal joint B we can use Eq. 8 for span BC. Thus, the structure to be analyzed has one
degree of freedom, which are the unknown joint rotation  B
Using the formulas for the FEMs table, we have
5wL2 5(75 kN / m)(4m) 2
( FEM ) BC     93.75 kNm
96 64
5wL2 5(75 kN / m)(4m) 2
( FEM )CB    93.75 kNm
96 64
Since C are fixed support, C  0. Also, since the supports do not settle, nor are they displaced
up or down,  BC  0.
Slope-Deflection Equations
I
M BC  2 E (2 B  C  3 BC )  ( FEM ) BC
L
I
M BC  2 E    2 B  0  3(0)   93.75  EI B  93.75 .....................1
4
I
M CB  2 E (2C   B  3 BC )  ( FEM )CB
L
I
M CB  2 E    2(0)   B  3(0)  48  0.5EI B  93.75 .....................2
4
Equilibrium Equations.
The above two equations contain three unknowns. The necessary third equation comes from the
conditions of equilibrium at the support B. The free-body diagram is shown in Fig. 1.12d.We have
M B 0 M BA  M BC  0 ........................3
To solve, substitute Eqs. (1) and MBA into Eq. (3), which yields
225  EI B  93.75  0
131.25
B 
EI
Substituting  B into Eqs. (1) and (2), we get
 131.25 
M BC  EI B  93.75  EI    93.75  225 kNm
 EI 
 131.25 
M CB  0.5EI B  93.75  0.5 EI    93.75  28.125kNm
 EI 

Analysis of Frames: No Side sway
A frame will not sidesway, or be displaced to the left or right, provided it is properly restrained.
Example are shown in fig. 1.13a. also, no sidesway will occur in an unrestrained frame provided
it is symmetric with respect to both loading and geometry, as shown in Fig. 1.13b. for both cases
the term ψ in the slope-deflection equations is equal to zero, since bending does not cause the joints
to have a linear displacement.
E.g. 1.5. Determine the moments at B and C for the frame shown in Fig. 1.14a by using the slope-
deflection method. EI is constant.

Solution:
Slope-Deflection Equations. Three spans must be considered in this problem: AC, BC, and CD.
Eq. 8 applies to span BC. We can use Eq. 9 for span AC and CD since A and D are on a pin and
roller supports, respectively.
Fixed-End Moments. Using the formulas for the FEMs table, we have
P 2 a 2b  20kN  2 12 *3
( FEM )CA   2  (b a  )   2  (3 *1  )  13.125 kNm
L  2  4  2
 wL2 (10 kN / m)(6m) 2
( FEM ) BC    30 kNm
12 12
wL2 (10 kN / m)(6m) 2
( FEM )CB    30 kNm
12 12
( FEM )CD  0
Note that  B  0 and  AC   BC   CD  0, since no side-sway will occur.
Applying Eq. 8, we have
I
M BC  2 E (2 B  C  3 BC )  ( FEM ) BC
L
 2I  2 EI
M BC  2 E    2(0)  C  3(0)   30  C  30 .....................1
 6  3
I
M CB  2 E (2C   B  3 BC )  ( FEM )CB
L
 2I  4 EI
M CB  2 E    2C  0  3(0)   30  C  30 .....................2
 6  3
Applying Eq. 9, we have
M N  3Ek ( N  )  ( FEM ) N
I
M CA  3E   (C  AC )  ( FEM )CA
L
I
M CA  3E   (C  0)  13.125  0.75EIC  13.125 .......................3
4
M N  3Ek ( N  )  ( FEM ) N
I
M CD  3E   (C  CD )  ( FEM )CD
L
I
M CD  3E   (C  0)  0  0.5EIC .............................................4
6
Equilibrium Equations. The preceding four equations contain five unknowns. The remaining one
equilibrium equation come from moment equilibrium at joints C, Fig. 1.14b.We have

M C 0 M CA  M CB  M CD  0 ........................5
To solve, substitute Eqs. (2), (3) and (4) into Eq. (5), which yields
4 EI 16.6938
0.75 EIC  13.125  C  30  0.5 EIC  0  C  
3 EI
Resubstituting this value into Eqs. (1)- (4) yields
2 EI 2 EI  16.6938 
M BC  C  30     30  41.129 kNm
3 3  EI 
4 EI 4 EI  16.6938 
M CB  C  30     30  7.7416 kNm
3 3  EI 
 16.6938 
M CA  0.75 EIC  13.125  0.75 EI     13.125  0.60465 kNm
 EI 
 16.6938 
M CD  0.5EIC  0.5 EI     8.3469
 EI 
Analysis of Frames: Side sway
A frame will side sway, or be displaced to the side, when it or the loading acting on it is non-
symmetric. To illustrate this effect, consider the frame shown in Fig. 1.15 bellow. Here the loading
P causes unequal moments MBC and MCB at the joints B and C, respectively. MBC tends to displace
joint B to the right, whereas MCB tends to displace Joint C to the left. Since MBC is larger than MCB,
the net result is a side-sway Δ of both joints B and C to the right, as shown in the figure. When
applying the slope-deflection equation to each column of this frame, we must therefore consider
the column rotation ψ (since ψ=Δ/L) as unknown in the equation. As a result an extra equilibrium
equation must be included for the solution. In the previous sections it was shown that unknown
angular displacements θ were related by joint moment equilibrium equations. In a similar manner,
when unknown joint linear displacements Δ (or span rotations ψ) occur, we must write force
equilibrium equations in order to obtain the complete solution. The unknowns in these equations
however, must only involve the internal moments acting at the ends of the columns, since the
slope-deflection equations involve these moments.

E.g. 1.6. Determine the member end moments for the frames shown in Fig. 1.16a by using the
slope-deflection method. EI is constant.
Solution:
Degrees of Freedom. The degrees of freedom are  B ,  C ,  D and  .

But, we will eliminate the rotation of the simple support at end D by using the modified slope-
deflection equations for member CD. Thus the analysis will involve only two unknown joint
rotations,  B and  C
Fixed-End Moments. Using the formulas for the FEMs table, we have
( FEM ) AB  ( FEM ) BA  0
 wL2 (10 kN / m)(10m) 2
( FEM ) BC    83.33 kNm
12 12
wL2 (10 kN / m)(10m) 2
( FEM )CB    83.33 kNm
12 12
( FEM )CD  0
Chord Rotations. From Fig. 15(b), we can see that
 
 AB   CD   BC  0
5 5
Relating  AB to  CD , we have    AB   CD  
5
Slope-Deflection Equations. Three spans must be considered in this problem: AB, BC, and CD.
Eq. 8 applies to span AB and BC. We can use Eq. 9 for span CD since D are on a pin support.
For span AB, considering A to be the near end and B to be the far end, we have
I
M AB  2 E (2 A   B  3 AB )  ( FEM ) AB
L
I
M AB  2 E    2(0)   B  3   0  EI (0.4 B  1.2 ) .....................1
5
Now, considering B to be the near end and A to be the far end, we have
I
M BA  2 E (2 B   A  3 AB )  ( FEM ) BA
L
I
M BA  2 E    2 B  0  3   0  EI (0.8 B  1.2 ) .....................2
5
For span BC, considering B to be the near end and C to be the far end, we have
I
M BC  2 E (2 B  C  3 BC )  ( FEM ) BC
L
 I 
M BC  2 E    2 B  C  3(0)  83.33  EI (0.4 B  0.2C )  83.33 .....................3
 10 
Now, considering C to be the near end and B to be the far end, we have

I
(2C   B  3 BC )  ( FEM )CB
M CB  2 E
L
 I 
M CB  2 E    2C   B  3(0)  83.33  EI (0.2 B  0.4C )  83.33 .....................4
 10 
For span CD, considering C to be the near end and D to be the far end, we have
M N  3Ek ( N  )  ( FEM ) N
I
M CD  3E   (C  CD )  ( FEM )CD
L
I
M CD  3E   (C  )  0  0.6 EI (C  ) .............................................5
5
Equilibrium Equations. The preceding five equations contain eight unknowns. By considering
the moment equilibrium of joints B and C, we obtain the equilibrium equations
M B 0 M BA  M BC  0 .................6 M C 0 M CB  M CD  0 ..........7
To establish the additional third equilibrium equation, we apply the force equilibrium equation
F x  0 to the free body of the entire frame Fig 1.16c, to obtain
F x  0 VA  VD  20 kN ..........8
The horizontal reactions or column shears V A and VD can be related to the internal moments by
considering the free-body diagram of each column separately, Fig. 1.16d.We have

M B 0 5VA  M AB  M BA  0 M C 0 5VD  M CD  0
( M AB  M BA )  M CD
VA  VD 
5 5
Thus, from Eq. (8),
VA  VD  20 kN
( M AB  M BA )  M CD
  20  M AB  M BA  M CD  100  0 ....................9
5 5
Substituting the slope-deflection Eqs. (2) and (3) into Eq. (6), we have
M BA  M BC  0
EI (0.8 B  1.2 )  EI (0.4 B  0.2C )  83.33  0
1.2 EI B  0.2 EIC  1.2 EI  83.33 ................................10
Substituting the slope-deflection Eqs. (4) and (5) into Eq. (7), we have
M CB  M CD  0
EI (0.2 B  0.4C )  83.33  0.6 EI (C  )  0
0.2 EI B  EIC  0.6 EI  83.33 .....................................11
Now, substituting the slope-deflection Eqs. (1), (2) and (5) into Eq. (9), we have
M AB  M BA  M CD  100  0
EI (0.4 B  1.2 )  EI (0.8 B  1.2 )  0.6 EI (C  )  100  0
1.2 EI B  0.6 EIC  3.4 EI  100 ............................................12
Solving Eqs. (10) through (12) simultaneously yields
EI B  151.691 EIC  71.465 EI  70.338

Member End Moments. By substituting the numerical values of EI B , EI C and EI into the
slope-deflection equations (Eqs. (1) through (5)), we obtain
M AB  EI (0.4 B  1.2 )  0.4(151.691)  1.2(70.338)  23.729 kNm

M BA  EI (0.8 B  1.2  0.8(151.691)  1.2(70.338)  36.947 kNm
M BC  EI (0.4 B  0.2C )  83.33  0.4(151.691)  0.2(71.465)  83.33  36.947 kNm
M CB  EI (0.2 B  0.4C )  83.33  0.2(151.691)  0.4( 71.465)  83.33  85.082 kNm
M CD  0.6 EI (C  )  0.6(71.465)  0.6(70.338)  85.082 kNm
Exercise
1.1 Determine the reaction by slope-deflection method
1.2 Determine the moments at A, B, and C for the loading and a settlement of 8 mm at support B.
E  200 Gpa and I  150(106 ) mm 4
1.3 Determine the moments at A, B, and C for the loading and a support settlement of 10mm at A
and 5mm at D. E  200 Gpa and I  400(106 ) mm4

1.4 Determine the moments acting at the supports A and D of the frame. EI is constant
1.5 Determine the member end moments and reactions for the frames shown in Fig. by using the
slope-deflection method. E= Constant
MOMENT DISTRIBUTION BY CROSS METHOD

Moment distribution is a method of successive approximations that may be carried out to any
desired degree of accuracy. Essentially, the method begins by assuming each joint of a structure is
fixed. Then, by unlocking and locking each joint in succession, the internal moments at the joints
are “distributed” and balanced until the joints have rotated to their final or nearly final positions.
It will be found that this process of calculation is both repetitive and easy to apply.
Sign Convention
In applying the moment-distribution method, we will adopt clockwise member end moments are
considered positive.

Since a clockwise moment at an end of a member must act in a counter-clockwise direction on the
adjacent joint, the foregoing sign convention implies that counter-clockwise moments on joints are
considered positive.
Fixed-End Moments (FEMs) - The moments at the “walls” or fixed joints of a loaded member
are called fixed-end moments. These moments can be determined from the table, depending upon
the type of loading on the member.
Member Stiffness Factor- Consider the beam in Fig. 1.17 below, which is pinned at one end and
fixed at the other. Application of the moment M causes the end A to rotate through an angle  A . In
slope-deflection method we related M to  A using the conjugate-beam method, that is,
M  (4 EI L) A . The term in parentheses is referred to as the stiffness factor at A and can be defined
as the amount of moment M required to rotate the end A of the beam,  A  1 rad .
4EI
K For Far End Fixed
L
Joint Stiffness Factor- If several members are fixed connected to a joint and each of their far
ends is fixed, then by the principle of superposition, the total stiffness factor at the joint is the sum
of the member stiffness factors at the joint, that is, KT  K
Distribution Factor (DF) - If a moment M is applied to a fixed connected joint, the connecting
members will each supply a portion of the resisting moment necessary to satisfy moment
equilibrium at the joint. That fraction of the total resisting moment supplied by the member is
called the distribution factor (DF). To obtain its value, imagine the joint is fixed connected to n
members. If an applied moment M causes the joint to rotate an amount  then each member i
rotates by this same amount. If the stiffness factor of the i th member is K i , then the moment
contributed by the member is M i  K i . since equilibrium requires
M  M1  M n  K1  K n    Ki then the distribution factor for the i th member is
Mi Ki
DFi  
M   Ki

Canceling the common term  it is seen that the distribution factor for a member is equal to the
stiffness factor of the member divided by the total stiffness factor for the joint; that is, in general,
K
DF 
K
Member Relative-Stiffness Factor - Quite often a continuous beam or a frame will be made from
the same material so its modulus of elasticity E will be the same for all the members. If this is the
case, the common factor 4E will cancel from the numerator and denominator when the distribution
factor for a joint is determined. Hence, it is easier just to determine the member’s relative-stiffness
factor
I
KR  For Far End Fixed
L
Carry-Over Factor - Consider again the beam in Fig. 16. It was shown that M AB   4EI / L  A
and M BA   2EI / L  A , solving for  A and equating these equations we get M BA  M AB / 2 . The
carry-over factor represents the fraction of M that is “carried over” from the pin to the wall.
Stiffness-Factor Modifications
In some cases it is possible to modify the stiffness factor of a particular beam span and thereby
simplify the process of moment distribution.
Member Pin Supported at Far End. Many indeterminate beams have their far end span
supported by an end pin (or roller) as in the case of joint B in Fig. 1.18a. Here the applied moment
M rotates the end A by an amount  . To determine  the shear in the conjugate beam at must be
determined, Fig. 1.18b.We have
1 M  2 
M B'  0; V ' A ( L)   L  L   0
2  EI   3 
ML 3EI
V 'A    or M  
3EI L

3EI
Thus, the stiffness factor for this beam is K  For Far End Pinned or Roller Supported
L
Also, note that the carry-over factor is zero, since the pin at B does not support a moment. If this
modification is considered, the moment distribution process is simplified since the end pin does
not have to be unlocked–locked successively when distributing the moments. Also, since the end
span is pinned, the fixed-end moments for the span are computed using the values in the modified
column of the table.
E.g. 1.7. Determine the member end moments for the continuous beam shown in Fig. 1.19a due to
the load and a settlement of 20 mm at support B. Use the moment-distribution method.
E  200 Gpa and I  150(106 ) mm 4
Solution: The distribution factors at each joint must be computed first.

The stiffness factors for the members are
4 EI
K For span AB and BC sin ce far ends are fixed
L
3EI
K For span CD sin ce, far end is roller
L
4 EI 4 EI 4 EI 4 E (2 I ) 3EI 3EI
K AB    0.5 EI K BC    EI K CD    0.5 EI
L 8 L 8 L 6
Therefore,
K BA 0.5EI
DFAB  0 since, it is fixed support DFBA    0.3333
K BA  K BC 0.5EI  EI
K BC EI KCB EI
DFBC    0.6667 DFCB    0.6667
K BA  K BC 0.5EI  EI K CB  K CD EI  0.5EI
KCD 0.5EI
DFCD    0.3333
KCB  KCD EI  0.5EI
The fixed-end moments

A qualitative deflected shape of the continuous beam with all joints clamped against rotation and
subjected to the specified support settlement is depicted in Fig. 1.19(b) using an exaggerated scale.

It can be seen from this figure that the relative settlements for the two members are
 AB   BC  0.02m
By using the fixed-end-moment expressions, we determine the fixed-end moments due to the
support settlements to be
6 EI  6(200*106 )(150*106 )(0.02)
( FEM ) AB  ( FEM ) BA   2    56.25 kN .m
L 82
6 EI  6(200*106 )(2*150*10 6 )(0.02)
( FEM ) BC  ( FEM )CB  2   112.5 kN .m
L 82
( FEM )CD  ( FEM ) DC  0
The fixed-end moments due to the external load are
 wL2 (10 kN / m)(8m) 2
( FEM ) AB    21.3333 kNm
30 30
wL2 (10 kN / m)(8m)2
( FEM ) BA    32 kNm
20 20
 wL2 (10 kN / m)(8m) 2
( FEM ) BC    53.3333 kNm
12 12
wL2 (10 kN / m)(8m) 2
( FEM )CB    53.3333 kNm
12 12
PL (20 kN )(6m)
( FEM )CD     40 kNm
3 3
Thus, the total fixed-end moments due to the combined effect of the external load and the support
settlements are
( FEM ) AB  56.25 kN .m  21.3333 kNm  77.5833 kNm
( FEM ) BA  56.25 kN .m  32 kNm  24.25 kNm
( FEM ) BC  112.5 kNm  53.3333 kNm  59.1667 kNm
( FEM )CB  112.5 kNm  53.3333 kNm  165.8333 kNm
( FEM )CD  0  40 kNm  40 kNm

Moment Distribution
The moment distribution is carried out on the moment distribution table.
Joint A B C
Member AB BA BC CB CD
DF 0 0.3333 0.6667 0.6667 0.3333
FEM -77.5833 -24.25 59.1667 165.8333 -40
Dist. -11.6377 -23.279 -83.893 -41.94
CO -5.81885 -41.9465 -11.6395
Dist. 13.9808 27.9657 7.7601 3.8794
CO 6.9904 3.88005 13.98285
Dist. -1.29322 -2.5868 -9.3224 -4.6605
CO -0.64661 -4.6612 -1.2934
Dist. 1.5538 3.1076 0.8623 0.4311
CO 0.7769 0.43115 1.5538
Dist. -0.1437 -0.2874 -1.0359 -0.5179
CO -0.07185 -0.51795 -0.1437
Dist. 0.1726 0.3453 0.0479 0.0958
M -76.35331 -21.62 21.62 82.712 -82.712
Moment Distribution for Frames: No Side-sway

Application of the moment-distribution method for frames having no side-sway follows the same
procedure as that given for beams. To minimize the chance for errors, it is suggested that the
analysis be arranged in a tabular form, as in the previous examples. Also, the distribution of
moments can be shortened if the stiffness factor of a span can be modified as indicated in the
previous section.

E.g. 1.8. Determine the member end moments for the frames shown in Figs. 1.20(a) by using the
moment-distribution method.

4 EI
K For span AB since far ends are fixed
L
3EI
K For span BC since, far end is pinned
L
4 EI 4 EI 2 EI 3EI 3EI
K AB    K BC  
L 6 3 L 8
Therefore,
DFAB  0 since, it is fixed support
2 EI 3EI
K BA 3 K BC 8
DFBA    0.64 DFBC    0.36
K BA  K BC 2 EI  3EI K BA  K BC 2 EI  3EI
3 8 3 8
The fixed-end moments
Pb 2 a (12* 22 * 4)  (15* 42 * 2)
( FEM ) AB      18.6667 kNm
L2 62
Pa 2b (12*42 *2)  (15*22 *4)
( FEM ) BA    17.3333 kNm
L2 62
 wL2 (24 kN / m)(8m)2
( FEM ) BC    192 kNm
8 8

Moment Distribution
The moment distribution is carried out on the moment distribution table.
Joint A B
Member AB BA BC
DF 0 0.64 0.36
FEM -18.6667 17.3333 -192
Dist. 111.7867 62.88
CO 55.89335
M 37.22665 129.12 -129.12
Moment Distribution for Frames: Side-sway

In this section, we apply the moment-distribution method to analyze frames whose joints may
undergo both rotations and translations. Such frames are commonly referred to as frames with
side-sway.
Consider, for example, the rectangular frame shown in Fig. 1.21(a). A qualitative deflected shape
of the frame for an arbitrary loading is also shown in the figure using an exaggerated scale. While
the fixed joints A and B of the frame are completely restrained against rotation as well as
translation, the joints C and D are free to rotate and translate. However, since the members of the
frame are assumed to be inextensible and the deformations are assumed to be small, the joints C
and D displace by the same amount, D, in the horizontal direction only, as shown in the figure.
The moment-distribution analysis of such a frame, with side-sway, is carried out in two parts. In
the first part, the side-sway of the frame is prevented by adding an imaginary roller to the structure,
as shown in Fig. 1.21(b). External loads are then applied to this frame, and member end moments

are computed by applying the moment-distribution process in the usual manner. With the member
end moments known, the restraining force (reaction) R that develops at the imaginary support is
evaluated by applying the equations of equilibrium.
In the second part of the analysis, the frame is subjected to the force R, which is applied in the
opposite direction, as shown in Fig. 1.21(c). The moments that develop at the member ends are
determined and superimposed on the moments computed in the first part (Fig. 1.21(b)) to obtain
the member end moments in the actual frame (Fig. 1.21(a)). If M , M O , and M R denote,
respectively, the member end moments in the actual frame, the frame with side-sway prevented,
and the frame subjected to R, then we can write
M  MO  M R
An important question that arises in the second part of the analysis is how to determine the member
end moments M R that develop when the frame undergoes side-sway under the action of R. Since
the moment-distribution method cannot be used directly to compute the moments due to the known
lateral load R, we employ an indirect approach in which the frame is subjected to an arbitrary
known joint translation  ' caused by an unknown load Q acting at the location and in the direction
of R, as shown in Fig. 1.22. From the known joint translation,  ' we determine the relative
translation between the ends of each member, and we calculate the member fixed-end moments.
The fixed-end moments thus obtained are distributed by the moment-distribution process to
determine the member end moments M Q caused by the yet-unknown load Q. Once the member
end moments M Q have been determined, the magnitude of Q can be evaluated by the application
of equilibrium equations.
With the load Q and the corresponding moments M Q known, the desired moments M R due to the
lateral load R can now be determined easily by multiplying M Q by the ratio R Q , that is,

R
M R    MQ
Q
By substituting into M  M O  M R we can express the member end moments in the actual frame
(Fig. 1.21(a)) as
R
M  MO    MQ
Q
E.g. 1.9. Determine the member end moments for the frames shown in Figs. 1.23(a) by using the
moment-distribution method.

4 EI
K For span AB and BC sin ce far ends are fixed
L
3EI
K For span CD sin ce, far end is roller
L
4 EI 4 EI 4 EI 4 EI 3EI 3EI
K AB    EI K BC    0.6667 EI K CD    0.6 EI
L 4 L 6 L 5
Therefore,
K BA EI
DFAB  0 since, it is fixed support DFBA    0.6
K BA  K BC EI  0.6667 EI
K BC 0.6667 EI KCB 0.6667 EI
DFBC    0.4 DFCB    0.5263
K BA  K BC EI  0.6667 EI K CB  K CD 0.6667 EI  0.6 EI
KCD 0.6 EI
DFCD    0.4737
KCB  KCD 0.6667 EI  0.6 EI
Part I: Side-sway Prevented In the first part of the analysis, the side-sway of the frame is
prevented by adding an imaginary roller at joint C, as shown in Fig. 1.23(b).

Fixed End Moments due to load

PL 20kN (4m) wL2 12kN / m(6m) 2
( FEM ) AB     10 kNm ( FEM ) BC     36 kNm
8 8 12 12
PL 20kN (4m) wL2 12kN / m(6m) 2
( FEM ) BA    10 kNm ( FEM )CB    36 kNm
8 8 12 12
( FEM )CD  0
The moment-distribution of these fixed-end moments is then performed, as shown on the moment-
distribution table
Joint A B C
DF 0 0.6 0.4 0.5263 0.4737
FEM -10 10 -36 36 0

Dist. 15.6 10.4 -18.9468 -17.0532
CO 7.8 -9.4734 5.2
Dist. 5.68404 3.78948 -2.73676 -2.46324
CO 2.84202 -1.094704 1.89474
Dist. 0.6568224 0.4378816 -0.9972 -0.8975
CO 0.3284112 -0.4986 0.2189408
Dist. 0.29916 0.19944 -0.1152 -0.1037
CO 0.14958 0.0576 0.09972
Dist. -0.03456 -0.02304 -0.0525 -0.0472
M 1.12 32.205 -32.205 20.565 -20.565

To evaluate the restraining force R that develops at the imaginary roller support, we first calculate
the shears at the lower ends of the columns AB and CD by considering the moment equilibrium of
the free bodies of the columns shown in Fig. below.
M B  0;  32.205  1.2  20(2)  4 Ax  0

Ax  1.64875 kN 
M B  0;  C y (6m)  12(6)(3)  20.565  32.205  0

C y  34.06 kN 
F
y  0; Dy  C y  0
Dy  34.06 kN 
M C  0;  20.565  Dy (3m)  Dx (4m)  0
Dx  30.68625 kN 
Then apply the equation of equilibrium in the horizontal direction to the entire frame:
F x  0; 20kN  1.64875kN  30.68625kN  R  0

R  11.851125 kN

Part II: Side-sway
From this figure, we can see that the relative translation  AB between the ends of member AB in
the direction perpendicular to the member can be expressed in terms of the joint translation  ' as
 AB  BB '   '
Similarly, the relative translations for members BC and CD are given by
4 5
 BC  B ' C '  '  CD  CC '   '
3 3
The fixed-end moments due to the relative translations are
5 
3EI   ' 
6 EI  AB 6 EI  ' 3EI CD  3   EI  '
( FEM ) AB  ( FEM ) BA   ( FEM )CD  
L2 42 L2 52 5
4 
6 EI   ' 
6 EI  BC  3    2 EI  '
( FEM ) BC  ( FEM )CB  
L2 62 9
If we arbitrarily assume that
EI  '
( FEM )CD   100 kNm
5
EI  '  500
6 EI  ' 6(500)
( FEM ) AB  ( FEM ) BA    187.5 kNm
42 16
2 EI  ' 2(500)
( FEM ) BC  ( FEM )CB    111.11 kNm
9 9
These fixed-end moments are distributed by the moment-distribution process Q determine the
member end moments MQ..

Joint A B C
DF 0 0.6 0.4 0.5263 0.4737
FEM 187.5 187.5 -111.11 -111.11 100

Dist. -45.834 -30.556 5.847193 5.262807
CO -22.917 2.9236 -15.278
Dist. -1.75416 -1.16944 8.041 7.237
CO -0.87708 4.0205 0.58472
Dist. -2.4123 -1.6082 -0.30774 -0.27681
CO -1.20615 -0.15387 -0.8041
Dist. 0.092322 0.061548 0.4232 0.3809
CO 0.14958 0.2116 0.030774
Dist. -0.12696 -0.08464 -0.0162 -0.0146
M 162.65 137.465 -137.465 -112.59 112.59
To evaluate the magnitude of Q that corresponds to these member end moments, we first calculate
shears at the lower ends of the columns by considering their moment equilibrium and then apply
the equation of equilibrium in the horizontal direction to the entire frame:
M B  0; 162.65  137.465  4 Ax  0
Ax  75.02875 kN 
M B  0;  C y (6m)  137.465  112.59  0

C y  41.68 kN 

F y  0; Dy  C y  0
Dy  41.68 kN 
M C  0; 112.59  Dy (3m)  Dx (4m)  0
Dx  59.4075 kN 
Fx  0; 75.02875kN  59.4075kN  Q  0

Q  134.43625 kN 
Actual Member End Moments

The actual member end moments, M
R
M  MO    MQ
Q
 11.85 
M AB  1.12    (162.65)  15.46 kNm
 134.44 
 11.85 
M BA  32.205    (137.465)  44.32 kNm
 134.44 
 11.85 
M BC  32.205    (137.465)  44.32 kNm
 134.44 
 11.85 
M CB  20.565    (112.59)  10 kNm
 134.44 
 11.85 
M CD  20.565    (112.59)  10 kNm
 134.44 

Exercise
1.6 Determine the reactions and draw the shear and bending moment diagrams for the beams
shown in Fig. by using the moment-distribution method.
1.7 Determine the member end moments for the beam shown in Fig. by using the moment-
distribution method. Use the support settlements of 20 mm at C and 30 mm at E.
1.8 Determine the member end moments for the frame
1.9 Determine the member end moments and reactions for the frame.

KANI METHOD OF MOMENT DISTRIBUTION

The moment distribution by the cross and kani methods are both iterative procedures to solve the
slope deflection equations. However, the cross method obtains the unknown end moments by
iterating the moment increments, while Kani’s method iterates the end moments themselves as the
unknowns. Kanis’s method consists of carrying out a single operation applied repeatedly at the
joints of a structure in an arbitrary sequence. Using kani’s method the results may also be obtained
to any desired accuracy by continuing the calculations a sufficient number of times. In addition to
its simplicity, the method has the advantage of acquiring a built in error elimination scheme.
Moreover, the method is more suitable for frames with high degree of redundancy, including
frames with side-sway, since only one set of computations is necessary. For such frames the
required computational effort by kani method in minimal when compared to other methods.
FRAMES WITHOUT SIDE-SWAY
Consider member A-B as integral part of a frame (Fig. 1.24) and that there are many such members
meeting at joint A so that B is the general designation, for the far ends. Let MAB and MBA be the
end moments due to the applied loads at joints A and B, respectively.
The slope deflection equation for member A-B at joint A which is considered part of a frame
without side-sway is written as
M AB  2 Ek AB (2 A   B )  ( FEM ) AB
This same equation may be written in the form

M AB  2M ' AB  M 'BA  ( FEM ) AB
Where, M ' AB  2 Ek AB A M 'BA  2 Ek AB B
Since M ' AB constitutes the contribution by  A to the total moment M AB it is referred to as the
rotation contribution of the end B to M AB

For any joint A where the number of n members are joined to it, the joint being in equilibrium of
end-moments gives
M
n
AB 0 or  (2M '
n
AB  M 'BA )   ( FEM ) AB  0
n
Defining M A the algebraic sum of the fixed-end moments at joint A as the restraint moment
M A   ( FEM ) AB
n

1
Then M '
n
AB   ( M A   M ' AB )
2 n
Since M AB for any member must be proportional to the relative stiffness of the member, the
moment in any member A-B is
k AB
M ' AB  M '
 k n AB
1 k 
   AB  ( M A   M ' AB )
2   k  n
Denoting the term  1 (k AB

2  k ) as the rotation factor R AB , may be written as
M ' AB  RAB (M A   M ' AB )

n
Rules for Calculating Rotation Contributions: - Case-1: Without side-way.

Definition: “Restrained moment at a joint is the algebraic sum of FE.M’s of different members
meeting at that joint.”

1. Sum of the restrained moment of a joint and all rotation contributions of the far ends of
members meeting at that joint is multiplied by respective rotation factors to get the required
near end rotation contribution. For the first cycle when far end contributions are not known,
they may be taken as zero (1st approximation).
2. By repeated application of this calculation procedure and proceeding from joint to joint in
an arbitrary sequence but in a specific direction, all rotation contributions are known. The
process is usually stopped when end moment values converge. This normally happens after
three or four cycles. But values after 2nd cycle may also be acceptable for academic.
E.g. 1.10. Analyze the following beam shown in fig. 1.25 by using kani method. EI is constant
Solution: - Stiffness factor

3EI 3EI 4 EI 4 EI
K AB    0.3EI K BC    0.4 EI
L 10 L 10
Rotation Factor (R)
1  0.3EI  1 0.3EI 
RAB      0.5 RBA      0.2143
2  0  3EI  2  0.3EI  0.4 EI 
1 0.4 EI 
RBC      0.2857
2  0.3EI  0.4 EI 
Fixed end moments
3PL 3(10kN )(10m)
( FEM ) BA    18.75kNm
16 16
wL2 5(10) 2
( FEM ) BC    41.6667kNm
12 12
wL2 5(10)2
( FEM )CB    41.6667kNm
12 12
Draw Boxes, enter the values of FEMs near respective ends of exterior boxes and rotation
contribution factors appropriately (on the interior side).

1st Cycle
Joint-A 0.5(0  0)  0
0.2143(22.9167  0  0)  4.911
Joint-B
0.2857(22.9167  0  0)  6.547
2nd cycle
Joint-A 0.5(0  4.911)  2.4555
0.2143(22.9167  2.4555  0)  5.4373
Joint-B
0.2857(22.9167  2.4555  0)  7.2488
3rd Cycle
Joint-A 0.5(0  5.4373)  2.71865
0.2143(22.9167  2.71865  0)  5.4937
Joint-B
0.2857(22.9167  2.71865  0)  7.324
4th Cycle
Joint-A 0.5(0  5.4937)  2.74865
0.2143(22.9167  2.74865  0)  5.4997
Joint-B
0.2857(22.9167  2.74865  0)  7.3326
Final End Moments
M N  2M 'N  M 'F  ( FEM ) N
M AB  2(2.74865)  5.4997  0  0
M BA  2(5.4997)  2.74685  18.75  27 kNm
M BC  2(7.3326)  0  41.6667  27 kNm
M CB  2(0)  7.3326  41.6667  49 kNm

E.g. 1.11. Analyze the following frame shown in fig. 1.26 by using kani method. EI is constant

4 EI 4 EI 4 EI 4 EI 4 EI 4 EI
K AB    0.6667 EI K BC    EI K CD    EI
L 6 L 4 L 4
Rotation Factor (R)
1  0.6667 EI  1 EI 
RBA      0.2 RBC      0.3
2  0.6667  EI  2  0.6667 EI  EI 
1  EI  1  EI 
RCB      0.25 RCD      0.25
2  EI  EI  2  EI  EI 
RDC  0.5
Fixed End Moments

wL2 15(6) 2 wL2 15(6) 2
( FEM ) AB    45 kNm ( FEM ) BA    45 kNm
12 12 12 12
( FEM ) BC  ( FEM )CB  ( FEM )CD  0
Moment Distribution
1st Cycle 2nd Cycle
0.2(45  0  0)  9 0.2(45  0  3.375)  9.675
Joint-B Joint-B
0.3(45  0  0)  13.5 0.3(45  0  3.375)  14.5125
0.25(0  13.5  0)  3.375 0.25(0  14.5125  16.875)  7.8469
Joint-C Joint-C
0.25(0  13.5  0)  3.375 0.25(0  14.5125  16.875)  7.8469
Joint-D 0.5(30  3.375)  16.875 Joint-D 0.5(30  7.8469)  18.9234

3rd Cycle
0.2(45  0  7.8469)  10.5694
Joint-B
0.3(45  0  7.8469)  15.8541
0.25(0  15.8541  18.9234)  8.6944
Joint-C
0.25(0  15.8541  18.9234)  8.6944
Joint-D 0.5(30  8.6944)  19.322
4th Cycle
0.2(45  0  8.6944)  10.7389
Joint-B
0.3(45  0  8.6944)  16.1083
0.25(0  16.1083  19.322)  8.8576
Joint-C
0.25(0  16.1083  19.322)  8.8576
Joint-D 0.5(30  8.8576)  19.4288
5th Cycle
0.2(45  0  8.8576)  10.7715
Joint-B
0.3(45  0  8.8576)  16.1573
0.25(0  16.1573  19.4288)  8.8965
Joint-C
0.25(0  16.1573  19.4288)  8.8965
Joint-D 0.5(30  8.8965)  19.44825

6th Cycle
0.2(45  0  8.8965)  10.7793
Joint-B
0.3(45  0  8.8965)  16.16895
0.25(0  16.16895  19.44825)  8.90295
Joint-C
0.25(0  16.16895  19.44285)  8.90295
Joint-D 0.5(30  8.90295)  19.4515
Final End Moments
M AB  45  2(0)  10.7793  55.7793 kNm
M BA  45  2(10.7793)  0  23.44 kNm
M BC  0  2(16.16895)  8.90295  23.44 kNm
M CB  0  2(8.90295)  16.16895  1.64 kNm
M CD  0  2(8.90295)  19.4515  1.65 kNm
M DC  2(19.4515)  8.90295  30
Case 2: with side sway (joint translations)

When a frame is either structurally un-symmetric or is symmetric with un-symmetrical loading,
joint translation or side-sway occurs. Fig. 1.27 shows member A-B of a frame with lateral
displacement. The rotations at joints A and B are  A and  B respectively, and  AB is the relative
displacement between A and B.
The slope deflection equation for the member A-B is
M AB  2EK AB (2 A   B  3  L)  ( FEM ) AB
M AB  2M ' AB  M 'BA  M '' AB  ( FEM ) AB
The algebraic sum of the end moments of all members meeting at joint A is zero.
M n
AB 0
 (2M '
n
AB  M 'BA  M '' AB )   ( FEM ) AB  0
n
Defining the algebraic sum of the fixed-end moments at joint A as the restraint moment,
M A   ( FEM ) AB
n
1
Then, M '
n
AB   M A   ( M 'BA  M '' AB )
2

In this case in addition to rotation contribution, linear displacement contributions (Sway

contributions) of columns of a particular storey are calculated after every cycle as follows:
For the first cycle.
Linear Displacement Contribution (LDC) of a column = Linear displacement factor (LDF) of a
particular column of a story multiplied by [storey moment + contributions at the ends of columns
of that story]
Linear displacement factor (LDF) for columns of a storey = 3

2
3 k
Linear displacement factor of a column = 
2 k
Where k=stiffness of the column being considered and Σk is the sum of stiffness of all columns
of that storey.
Storey moment = Storey shear x 1 of storey height.

3

Storey shear: It may be considered as reaction of column at horizontal beam / slab levels due to
lateral loads by considering the columns of each sotrey as simply supported beams in vertical
direction.
E.g. 1.12. Analyze the following frame shown in fig. 1.28 by using kani method. EI is constant

4 EI 4 EI 4 EI 4 EI
K AB  K BC  K DE  K EF    EI K CD  K BE    0.5EI
L 4 L 8
Rotation Factor (R)
1 K BA  1 EI 
RBA        0.2
2  K BA  K BC  K BE  2  EI  EI  0.5EI 
1 K BC  1 EI 
RBC        0.2
1 K BE  1 0.5 EI 
RBE        0.1
1 K EB  1 0.5EI 
REB        0.1
2  K EB  K ED  K EF  2  0.5 EI  EI  EI 
1 K ED  1 EI 
RED        0.2
2  K EB  K ED  K EF  2  0.5EI  EI  EI 
1 K EF  1 EI 
REF        0.2
2  K EB  K ED  K EF  2  0.5EI  EI  EI 
1 K CB  1 EI 
RCB        0.3333
2  K CB  K CD  2  EI  0.5EI 
1 K CD  1  0.5 EI 
RCD        0.1667
2  K CB  K CD  2  EI  0.5EI 

1 K DC  1  0.5EI 
RDC        0.1667
2  K DC  K DE  2  EI  0.5EI 
1 K DE  1  0.5EI 
RDE        0.3333
2  K DC  K DE  2  EI  0.5EI 
Linear Displacement Factor
3 K 3  EI 
L.D.F   L.D.F1      0.75 For storey  1
2 K 2  EI  EI 
3  EI 
L.D.F2      0.75 For storey  2
2  EI  EI 
Fixed End Moments
( FEM ) AB  ( FEM ) BA  ( FEM ) BC  ( FEM )CB  0
( FEM ) DE  ( FEM ) ED  ( FEM ) EF  ( FEM ) FE  0
wL2 (16)(82 )
( FEM )CD     85.3333 kNm
12 12
wL2 (16)(82 )
( FEM ) DC    85.3333 kNm
12 12
wL2 (12)(82 )
( FEM ) BE    64 kNm
12 12
wL2 (12)(82 )
( FEM ) EB    64 kNm
12 12
Storey Shear
For determining storey shear the columns can be treated as simply supported vertical beams.
Storey Shear  1  10 kN  5 kN  15 kN For lower or ground storey
Storey Shear  2  10 kN For upper storey
Storey Moment (S.M)
S.M. = Storey shear (h/3) where h is the height of that storey.
SM 1  15 kN (4m )  20 kNm
3
SM 2  10 kN (4m )  13.3333 kNm
3
Moment Distribution
Rotation contribution = RF (restrained moment + far end contributions + linear displacement
contribution of columns. of different. storeys meeting at that joint.)
Linear Displacement Contribution = LDF (Storey moment + Rotation contribution at the end of
columns of that storey)

1st Cycle
Rotation Contribution
M 'BA  0.2(64  0  0  0  0  0)  12.8
Joint-B M 'BC  0.2(64  0  0  0  0  0)  12.8
M 'BE  0.1(64  0  0  0  0  0)  6.4
M 'CB  0.3333(85.3333  12.8  0  0)  24.1753

Joint-C
M 'CD  0.1667(85.3333  12.8  0  0)  12.0913
M 'DC  0.1667(85.3333  12.0913  0  0)  16.2407

Joint-D
M 'DE  0.3333(85.3333  12.0913  0  0)  32.4716
M 'ED  0.2(64  32.4176  6.4  0  0)  7.5857
Joint-E M 'EF  0.2(64  32.4176  6.4  0  0)  7.5857
M 'EB  0.1(64  32.4176  6.4  0  0)  3.7928
Linear Displacement Contribution
LDC1  0.75(20  12.8  7.5857)  18.9107
LDC2  0.75(13.3333  12.8  24.1753  32.4716  7.5857)  7.6885
2nd Cycle
Rotation Contribution
M 'BA  0.2(64  0  24.1753  3.7928  7.6855  18.9107)  14.0427
Joint-B M 'BC  0.2(64  0  24.1753  3.7928  7.6855  18.9107)  14.0427
M 'BE  0.1(64  0  24.1753  3.7928  7.6855  18.9107)  7.0214
M 'CB  0.3333(85.3333  14.0427  16.2407  7.6855)  31.7385

Joint-C
M 'CD  0.1667(85.3333  4.0427  16.2407  7.6855)  15.8726
M 'DC  0.1667(85.3333  15.8726  7.5857  7.6855)  14.3253

Joint-D
M 'DE  0.3333(85.3333  15.8726  7.5857  7.6855)  28.642
M 'ED  0.2(64  0  7.0214  28.642  7.6855  18.9107)  3.1566
Joint-E M 'EF  0.2(64  0  7.0214  28.642  7.6855  18.9107)  3.1566
M 'EB  0.1(64  0  7.0214  28.642  7.6855  18.9107)  1.5783
Linear Displacement Contribution
LDC1  0.75(20  14.0427  3.1566)  23.1648
LDC2  0.75(13.3333  14.0427  31.7358  28.642  3.1566)  20.4849
N.B- Repeat an iteration up to 10 cycles to increase an accuracy.


Exercise
1.10. Analyze the following beam shown in fig by using kani method. EI is constant
1.11. Analyze the following frame shown in fig by using kani method. EI is constant
1.12 Determine the member end moments for the frame

2. Approximate Analysis of Statically Indeterminate Structures

When a model is used to represent any structure, the analysis of it must satisfy both the conditions
of equilibrium and compatibility of displacement at the joints.
Exact analysis of indeterminate structures involves computation of deflections and solution of
simultaneous equations, so it can be quite time consuming. Moreover, such an analysis depends
on the relative sizes (cross-sectional areas and/or moments of inertia) of the members of the
structure. Because of these difficulties associated with the exact analysis, the preliminary designs
of indeterminate structures are often based on the results of approximate analysis, in which the
internal forces are estimated by making certain assumptions about the deformations and/or the
distribution of forces between the members of structures, thereby avoiding the necessity of
computing deflections.
Approximate analysis proves to be quite convenient to use in the planning phase of projects, when
several alternative designs of the structure are usually evaluated for relative economy. The results
of approximate analysis can also be used to estimate the sizes of various structural members needed
to initiate the exact analysis. The preliminary designs of members are then revised iteratively,
using the results of successive exact analyses, to arrive at their final designs. Furthermore,
approximate analysis is sometimes used to roughly check the results of exact analysis, which due
to its complexity can be prone to errors.
The objectives of this chapter are to consider the approximate analysis of rectangular building
frames as well as to gain an understanding of the techniques used in the approximate analysis of
structures in general.
Vertical Loads on Building Frames
Building frames often consist of girders that are rigidly connected to columns so that the entire
structure is better able to resist the effects of lateral forces due to wind and earthquake.
An example of such a rigid framework, often called a building bent, is shown in Fig 1.29(a) below.

In practice, a structural engineer can use several techniques for performing an approximate
analysis of a building bent. Each is based upon knowing how the structure will deform under load.
One technique would be to consider only the members within a localized region of the structure.
This is possible provided the deflections of the members within the region cause little disturbance
to the members outside the region. Most often, however, the deflection curve of the entire structure
is considered. From this, the approximate location of points of inflection, that is, the points where
the member changes its curvature, can be specified. These points can be considered as pins since
there is zero moment within the member at the points of inflection. We will use this idea in this
section to analyze the forces on building frames due to vertical loads, and an approximate analysis
for frames subjected to lateral loads.
Assumptions for Approximate Analysis
Consider a typical girder located within a building bent and subjected to a uniform vertical load,
as shown in Fig. 1.29(b). The column supports at A and B will each exert three reactions on the
girder, and therefore the girder will be statically indeterminate to the third degree (6 reactions – 3
equations of equilibrium). To make the girder statically determinate, an approximate analysis will
therefore require three assumptions.
If the columns are extremely stiff, no rotation at A and B will occur, and the deflection curve for
the girder will look like that shown in Fig. 1.29(c). An exact analysis reveals that for this case
inflection points, or points of zero moment, occur at 0.21L from each support.

If, however, the column connections at A and B are very flexible, then like a simply supported
beam, zero moment will occur at the supports, Fig. 1.29(d).
In reality, however, the columns will provide some flexibility at the supports, and therefore we
will assume that zero moment occurs at the average point between the two extremes, i.e., at
0.21L  0
 0.1L from each support, Fig. 1.29(e)
2
Furthermore, an exact analysis of frames supporting vertical loads indicates that the axial forces
in the girder are negligible.
In summary then, each girder of length L may be modeled by a simply supported span of length
0.8L resting on two cantilevered ends, each having a length of 0.1L, Fig. 1.29(f).
The following three assumptions are incorporated in this model:

1. There is zero moment in the girder, 0.1L from the left support.
2. There is zero moment in the girder, 0.1L from the right support.
3. The girder does not support an axial force.

E.g. 1.13 Determine (approximately) the internal moments at joints A and B of the frame.
Solution: -
For an approximate analysis the frame is modeled as shown in Fig. below. Note that the
cantilevered spans supporting the center portion of the girder EF and HG have a length of
0.1L=0.1(6m) = 0.6m and for girder FG have a length of 0.1L=0.1(8m) =0.8m
Equilibrium equations
For girder EF
M E '0 4.8m( Fy ')  (3 kN / m)(4.8m)(2.4m)  0

Fy '  7.2 kN 
F '  0
y E y ' Fy '  (3 kN / m)(4.8m)
E y '  7.2 kN 
The cantilevered spans are then subjected to a reaction of
M E 0 M E  (3 kN / m)(0.6m)(0.3m)  (7.2 kN )(0.6m)  0

M E  4.86 kN .m
F y 0 E y  7.2 kN  (3kN / m)(0.6)
E y  8.1kN

Moment at joint A
From Column AE
M  0 M A  M E  4.86 kN .m
F  0 y Ay  E y  8.1 kN 
For girder FG
M F '0 6.4m(Gy ')  (3 kN / m)(6.4m)(3.2m)  0

Gy '  9.6 kN 
F '  0
y Gy ' Fy '  (3 kN / m)(6.4m)
Gy '  9.6 kN 
The cantilevered spans
M F 0
M F  (3 kN / m)(1.4m)(0.1m)  (7.2 kN )(0.6m)  (9.6kN )(0.8)  0
M F  3.78 kN .m
F y 0
Fy  7.2 kN  (3kN / m)(1.4m)  9.6kN
Fy  21kN 
Moment at joint B
From Column BF
M  0 M B  M F  3.78 kN .m
F  0 y By  Fy  21 kN 
Exercise
1.13 Determine (approximately) the internal moments at A and B caused by the vertical loading.
1.14 Draw the approximate shear and bending moment diagrams for the girders DEF of the frames
shown in Fig.
Lateral Loads on Building Frames: Portal Method

Portal frames are frequently used over the entrance of a bridge and as a main stiffening element in
building design in order to transfer horizontal forces applied at the top of the frame to the
foundation. On bridges, these frames resist the forces caused by wind, earthquake, and unbalanced
traffic loading on the bridge deck. Portals can be pin supported, fixed supported, or supported by
partial fixity.
Fixed Supported
Portals with two fixed supports, Fig. (a). are statically indeterminate to the third degree since there
are a total of six unknowns at the supports. If the vertical members have equal lengths and cross-
sectional areas, the frame will deflect as shown in Fig (b). For this case we will assume points of
inflection occur at the midpoints of all three members, and therefore hinges are placed at these
points.

The reactions and moment diagrams for each member can therefore be determined by
dismembering the frame at the hinges and applying the equations of equilibrium to each of the four
parts. The results are shown in Fig (c).
In summary, the portal method for analyzing fixed-supported building frames requires the
following assumptions:
1. A hinge is placed at the center of each girder, since this is assumed to be a point of zero
moment.
2. A hinge is placed at the center of each column, since this is assumed to be a point of zero
moment.

3. At a given floor level the shear at the interior column hinges is twice that at the exterior
column hinges, since the frame is considered to be a superposition of portals.
These assumptions provide an adequate reduction of the frame to one that is statically determinate
yet stable under loading.
E.g. 1.14. Determine (approximately) the reactions at the base of the columns of the frame and
draw the moment diagram for girder EF and column AC.
Solution: - Applying the first two assumptions of the portal method, we place hinges at the centers
of the girders and columns of the frame.
A section through the hinges at G, H and I, J yields the free-body diagrams shown in Fig. The
column shears are calculated as follows:
Fx  0 V  V  40 kN  V  20 kN 
Fx  0 V ' V '  100 kN  V '  50 kN 
Using these results, we can now proceed to analyze each part of the frame.

Segment GEK
 F  0 K  V  40 kN  K  20 kN 
x x x
 M  0 K (3m)  (40kN )(2m)  ( K )(2m)  0

G y x
K y  40 kN 
F y  0 Gy  K y
G y  40 kN 
Segment KFH
Fy 0 Hy  Ky
H y  40 kN 
Segment GCIL
F  0 x Lx  50kN  60kN  20kN  Lx  30kN 

M  0 C Ly (3m)  (20kN )(2m)  (50kN )(2m)  0
Ly  46.667 kN 
F y 0 I y  G y  Ly
I y  86.667 kN 
Segment HLDJ
F y 0 J y  H y  Ly
H y  86.667 kN 

Segment AI Segment BJ
M A 0 M B 0
M A  (50kN )(2m) M B  (50kN )(2m)
M A  100 kNm M B  100 kNm
F y 0 F y 0
Ay  86.667 kN  By  86.667 kN 
F x 0 F x 0
Ax  50 kN  Bx  50 kN 
The moment diagrams for girder EF
From segment EK, segment KF above
M E  K y (3m)  (40kN )(3m)  120 kNm CW
M F  K y (3m)  (40kN )(3m)  120 kNm CW
The moment diagrams for column AC

From segment CI above
M C  V '(2m)  (50kN )(2m)  100 kNm CCW
M A  100 kNm CCW
Exercise
1.15 Use the portal method of analysis and draw the moment diagram for girder DEF and column
BE.

1.16 Use the portal method of analysis and draw the moment diagram for column BEH.
Lateral Loads on Building Frames: Cantilever Method

The cantilever method is based on the same action as a long cantilevered beam subjected to a
transverse load. It may be recalled from mechanics of materials that such a loading causes a
bending stress in the beam that varies linearly from the beam’s neutral axis, Fig. a. In a similar
manner, the lateral loads on a frame tend to tip the frame over, or cause a rotation of the frame
about a “neutral axis” lying in a horizontal plane that passes through the columns between each
floor. To counteract this tipping, the axial forces (or stress) in the columns will be tensile on one
side of the neutral axis and compressive on the other side, Fig. b. Like the cantilevered beam, it
therefore seems reasonable to assume this axial stress has a linear variation from the centroid of
the column areas or neutral axis. The cantilever method is therefore appropriate if the frame is tall
and slender, or has columns with different cross-sectional areas.

In summary, using the cantilever method, the following assumptions apply to a fixed-supported
frame.
1. A hinge is placed at the center of each girder, since this is assumed to be a point of zero
moment.
2. A hinge is placed at the center of each column, since this is assumed to be a point of zero
moment.
3. The axial stress in a column is proportional to its distance from the centroid of the cross-
sectional areas of the columns at a given floor level. Since stress equals force per area, then
in the special case of the columns having equal cross-sectional areas, the force in a column
is also proportional to its distance from the centroid of the column areas.
E.g. 1.15. Draw the moment diagram for girder IJKL of the building frame. Use the cantilever
method. Each column has the cross-sectional area indicated.
Solution: - The centroid of the columns’ cross-sectional areas is determined as follows:
x x A
i i

0(24)  (4)(16)  (9)(16)  (15)(20)
 6.6842m
A i 24  16  16  20
Hinges are assumed to exist at the centers of the girders and columns of the frame.

First we will consider the section through hinges at M, N, O and P. The columns have different
cross-sectional areas, so we must consider the axial stress in each column to be proportional to its
distance from the neutral axis, located at x  6.6842m
We can relate the column stresses by proportional triangles, expressing the relations in terms of
the force in each column, since   F / A
N M  2.6842  N y  2.6842  M y
 N    M     N y  0.26771M y
2.6842 6.6842  6.6842  16  6.6842  24
O M  2.3158  Oy  2.3158  M y
  O    M     Oy  0.23097 M y
2.3158 6.6842  6.6842  16  6.6842  24
P M  8.3158  Py  8.3158  M y
 P    M     Py  1.03675M y
8.3158 6.6842  6.6842  20  6.6842  24
Summing moments about the neutral axis, we have
M  0 M y (6.6842)  N y (2.6842)  Oy (2.3158)  Py (8.3158)  20kN (2m)  0

M y (6.6842)  0.26771M y (2.6842)  0.23097 M y (2.3158)  1.03675M y (8.3158)  40
M y  2.416 kN
N y  0.26771M y  0.26771(2.416)  0.647 kN

Oy  0.23097 M y  0.23097(2.416)  0.558 kN
Py  1.03675M y  1.03675(2.416)  2.505 kN

We begin with the upper corner by segment MIU
F
y 0 Uy  My
U y  2.416 kN 
M M 0 U x (2m)  (2.416kN )(2m)  20kN (2m)
U x  17.584 kN 
F
x 0 M x  U x  20kN
M x  2.416kN 
Segment UJVN
F y 0 Vy  U y  N y
Vy  3.063 kN 
M N 0
Vx (2m)  (2.416kN )(2m)  3.063(2.5m)  17.584(2m)  0
Vx  11.34 kN 
F x 0 N x  Vx  17.584kN
N x  6.245kN 
Segment VKWO
F y 0 Wy  V y  O y
Wy  2.505 kN 
M O 0
Wx (2m)  (2.505kN )(3m)  3.063(2.5m)  11.34(2m)  0
Wx  3.754 kN 
F x 0 Ox  Wx  11.34kN
Ox  7.586kN 
Segment WLP
F x 0 Px  Wx  0
Px  7.586kN 

We can now proceed to analyze moment at each ends of girder IJKL of the frame.
From segment IU, M I  2.416 kN (2m)  4.832 kNm
From segment UJV
M Jleft  2.416 kN (2m)  4.832 kNm
M Jright  3.063kN (2.5m)  7.6575kNm
From Segment VKW

M Kleft  3.063kN (2.5m)  7.6575kNm
M Kright  2.505kN (3m)  7.515kNm
From Segment WL
M L  2.505kN (3m)  7.515kNm
Exercise
1.17 Determine (approximately) the reactions at A and, draw moment diagram for girder IJKL
and column BFJM. Use the cantilever method.

Chapter Two
Influence line for statically indeterminate structures
An influence line is a graph of a response function of a structure as a function of the position of a
downward unit load moving across the structure.
The basic procedure for constructing influence lines for indeterminate structures is the same as
that for determinate structures.
The procedure essentially involves computing the values of the response function of interest for
various positions of a unit load on the structure and plotting the response function values as
ordinates against the position of the unit load as abscissa to obtain the influence line. The influence
lines for indeterminate structures, however, are generally curved lines. (For indeterminate girders
with floor systems and trusses and for other indeterminate structures to which moving loads are
transmitted via framing systems, the influence lines usually consist of chords of curved lines.)
Thus the construction of influence lines for indeterminate structures requires computation of many
more ordinates than necessary in the case of determinate structures.
INFLUENCE LINES FOR BEAMS
Consider the continuous beam shown in Fig. 2.1(a). Suppose that we wish to draw the influence
line for the vertical reaction at the interior support B of the beam. The beam is subjected to a
downward-moving concentrated load of unit magnitude, the position of which is defined by the
coordinate x measured from the left end A of the beam, as shown in the figure.
Fig. 2.1

To develop the influence line for the reaction By , we need to determine the expression for By in
terms of the variable position x of the unit load. Noting that the beam is statically indeterminate to
the first degree, we select the reaction By to be the redundant. The roller support at B is then
removed from the actual indeterminate beam to obtain the statically determinate primary beam
shown in Fig. 2.1(b). Next, the primary beam is subjected, separately, to the unit load positioned
at an arbitrary point X at a distance x from the left end, and the redundant By , as shown in Fig.
2.1(b) and (c), respectively. The expression for By can now be determined by using the
compatibility condition that the deflection of the primary beam at B due to the combined effect of
the external unit load and the unknown redundant By must be equal to zero. Thus
 f Bx  f BB By  0
f Bx
From which By  ......................(1)
f BB
In which the flexibility coefficient f Bx denotes the deflection of the primary beam at B due to the
unit load at X (Fig.2.1 (b)), whereas the flexibility coefficient f BB denotes the deflection at B due
to the unit value of the redundant By (Fig. 2.1(c)).
We can use Eq. (1) for constructing the influence line for By by placing the unit load successively
at a number of positions X along the beam, evaluating f Bx for each position of the unit load, and
plotting the values of the ratio f Bx / f BB .
Procedure for Analysis

The procedure for constructing influence lines for statically indeterminate structures by the method
of consistent deformations can be summarized as follows:
1. Determine the degree of indeterminacy of the structure and select redundants.
2. Select a number of points along the length of the structure at which the numerical values
of the ordinates of the influence lines will be evaluated.
3. To construct the influence lines for the redundants, place a unit load successively at each
of the points selected in step 2; and for each position of the unit load, apply the method of
consistent deformations to compute the values of the redundants. Plot the values of the
redundants thus obtained as ordinates against the position of the unit load as abscissa, to
construct the influence lines for the redundants.

4. Once the influence lines for the redundants have been determined, the influence lines for
the other force and/or moment response functions of the structure can be obtained through
equilibrium considerations.
E.g. 2.1 Draw the influence lines for the reaction at support B and the bending moment at point C
of the beam shown in Fig.
(a) Real Beam

Solution: - The ordinates of the influence lines will be computed at 3-m intervals at points A
through E, as shown in Fig.
Influence lines for the reaction at support B
In order to determine ordinates of the influence line we will use the conjugate-beam method. The
reactions at B and E on the “real beam,” when subjected to the unit load at B, are shown in Fig.
(b)
 f Bx  f BB By  0
f BX
From which By 
f BB
(b) Primary beam loaded with unit load at B.

The corresponding conjugate beam is shown in Fig. (c). Notice that the fixed support at E of real
beam is represented by free end on conjugate beam at E’ and the free end at A of primary beam is
replaced by fixed support at A’.

The displacements of points on the real beam, Fig. (b), will now be computed.
For A’, by using Fig. (d)
1 9  2
M A '  0  M A   (9m)(3  (9m))  0
2  EI  3
364.5
A  M A ' 
EI
For B’, by using Fig. (e)
1  9  2 
M B '  0  M B     (9m)  0
2  EI  3 
243
B  M B ' 
EI
For C’, by using Fig. (f)
M C' 0
1  93 2  3 
 M C '   (6m)   (6m)    (6m)(3m)  0
2  EI  3  EI 
126
C  M C ' 
EI

For D’, by using Fig. (g)
M D' 0
196 2  6 
 M D '   (3m)   (3m)    (3m)(1.5m)  0
2  EI  3  EI 
36
D  M D ' 
EI
For E’ E  M E '  0
243
Since a vertical 1-kN load acting at B on the beam, the displacement at B,  B  should
EI
correspond to a numerical value of 1 for the influence-line ordinate at B. Thus, dividing the other
computed displacements by this factor, we obtain
X By
A 1.5
B 1
C 0.5158
D 0.1481
E 0
Influence lines for the bending moment at point C
A hinge is inserted at C in order to remove the capacity of the beam to resist moment at this point,
Fig. (i). the reactions at B and E on the “real beam” when subjected to the unit couple moments at
C are shown.
fCX
MC 
 CC

The corresponding conjugate beam and its reactions are shown in Fig. (j). Notice that the fixed
support at E of real beam is represented by free end on conjugate beam at E’, hinged connection
at C of real beam is replaced by roller support, roller support at B replaced by hinged connection
and the free end at A of primary beam is replaced by fixed support at A’.
The displacements of points on the real beam, Fig. (j), will now be computed.
40.5
For A’,  A  M A ' 
EI
For B’, from real beam there is no displacement at B  B  M B '  0
For C’, by using Fig. (k)
M C' 0
1  3 1  2  1 
 M C '   (6m)   (6m)    (6m)(3m)  0
2  EI  3  EI 
42
C  M C ' 
EI
For C’, by using Fig. (l)
M D' 0
1  3 2  2  2 
 M D '   (3m)   (3m)    (3m)(1.5m)  0
2  EI  3  EI 
12
D  M D ' 
EI

For E’ E  M E '  0
The angular displacement  CC at C of the “real beam” in Fig. (i) is defined by the reaction at C on
27
the conjugate beam. This factor, C y '  is divided into the above values to give the ordinates
EI
of the influence line, that is,
X By
A -1.5
B 0
C 1.556
D 0.444
E 0
Exercise
1. Draw the influence line for
- The vertical reaction at A and B

- shear at C
- bending moment at A and C
EI is constant. Plot numerical values every 2 m.
2. Draw the influence line and plot numerical values every 2 m for
- The vertical reaction at supports A, B and C
- Shear at G and E
- Bending moment at G and E
EI is constant.

3. Draw influence lines for the reaction at supports A and B, and bending moment at B.
QUALITATIVE INFLUENCE LINES BY MULLER-BRESLAU’S PRINCIPLE

In many practical applications, such as when designing continuous beams or building frames
subjected to uniformly distributed live loads, it is usually sufficient to draw only the qualitative
influence lines to decide where to place the live loads to maximize the response functions of
interest. As in the case of statically determinate structures. Muller-Breslau’s principle provides a
convenient means of establishing qualitative influence lines for indeterminate structures.
Muller-Breslau’s principle can be stated as follows:
The influence line for a force (or moment) response function is given by the deflected shape of the
released structure obtained by removing the restraint corresponding to the response function from
the original structure and by giving the released structure a unit displacement (or rotation) at the
location and in the direction of the response function, so that only the response function and the
unit load perform external work.
The procedure for constructing qualitative influence lines for indeterminate structures is the same
as that for determinate structures.
The procedure essentially involves:
(1) Removing from the given structure the restraint corresponding to the response function of
interest to obtain the released structure.
(2) Applying a small displacement (or rotation) to the released structure at the location and in
the positive direction of the response function.
(3) Drawing a deflected shape of the released structure consistent with its support and
continuity conditions. The influence lines for indeterminate structures are generally curved
lines.
Once a qualitative influence line for a structural response function has been constructed, it can be
used to decide where to place the live loads to maximize the value of the response function. The
value of a response function due to a uniformly distributed live load is maximum positive (or
negative) when the load is placed over those portions of the structure where the ordinates of the

response function influence line are positive (or negative). Because the influence-line ordinates
tend to diminish rapidly with distance from the point of application of the response function, live
loads placed more than three span lengths away from the location of the response function
generally have a negligible effect on the value of the response function. With the live-load pattern
known, an indeterminate analysis of the structure can be performed to determine the maximum
value of the response function.
E.g. 2.2 Draw qualitative influence lines for the vertical reactions at supports A and B, the bending
moment at point B, and the shear and bending moment at point C of the four-span
continuous beam shown in Fig. (a). Also, show the arrangements of a uniformly
distributed downward live load w to cause the maximum positive reactions at supports A
and B, the maximum negative bending moment at B, the maximum negative shear at C,
and the maximum positive bending moment at C.
Solution: -
Influence Line for Ay
To determine the qualitative influence line for the vertical reaction Ay at support A, we remove the
vertical restraint at A from the actual beam and give the released beam a small displacement in the
positive direction of Ay. The deflected shape of the released beam thus obtained (Fig. (b))
represents the general shape of the influence line (i.e., the qualitative influence line) for Ay. Note
that the deflected shape is consistent with the support conditions of the released beam; that is,
points B;D;E, and F of the released beam, which are attached to roller supports, do not displace.
To maximize the positive value of Ay, the live load w is placed over spans AB and DE of the beam,
where the ordinates of the influence line for Ay are positive, as shown in Fig. (b).

Influence Line for By

The qualitative influence line for By and the live-load arrangement for the maximum positive value
of By are determined in a similar manner and are shown in Fig. (c).
Influence Line for MB

To determine the qualitative influence line for the bending moment at B, we insert a hinge at B in
the actual beam and give the released beam a small rotation in the positive direction of MB by
rotating the portion to the left of B counterclockwise and the portion to the right of B clockwise,
as shown in Fig. (d). the deflected shape of the released beam thus obtained represents the
qualitative influence line for MB.
To cause the maximum negative bending moment at B, we place the live load w over spans AB;
BD, and EF of the beam, where the ordinates of the influence line for MB are negative, as shown
in Fig. (d).

Influence Line for VC
The qualitative influence line for VC is determined by cutting the actual beam at C and by giving
the released beam a small relative displacement in the positive direction of VC by moving end C
of the left portion of the beam downward and end C of the right portion upward, as shown in Fig.
(e).
To obtain the maximum negative shear at C, the live load is placed over span DE and the portion
BC of the span BD of the beam, where the ordinates of the influence line for a VC are negative.
Influence Line for MC

The qualitative influence line for the bending moment at C and the live-load arrangement for the
maximum positive value of MC are shown in Fig. (f).

Exercise
3. Draw qualitative influence lines for the vertical reactions at supports A and B, the bending
moment at point B, and the shear and bending moment at point C of the beams shown in
Fig. Also, show the arrangements of a uniformly distributed downward live load w to cause
the maximum positive reactions at supports A and B, the maximum Positive bending
moment at B, the maximum Positive shear at C, and the maximum negative bending
moment at C.

Chapter Three
Introduction to Matrix Methods
3.1 Stiffness Method
In the matrix stiffness method of analysis, the structure is considered to be an assemblage of
straight members connected at their ends to joints. A member is defined as a part of the structure
for which the member force displacement relations to be used in the analysis are valid. In other
words, given the displacements of the ends of a member, one should be able to determine the forces
and moments at its ends by using the force displacement relations. A joint is defined as a structural
part of infinitesimal size to which the member ends are connected. The members and joints of
structures are also referred to as elements and nodes, respectively.
Truss Analysis Using the Stiffness Method

Application of the stiffness method requires subdividing the structure into a series of discrete finite
elements and identifying their end points as nodes. For truss analysis, the finite elements are
represented by each of the members that compose the truss, and the nodes represent the joints. The
force-displacement properties of each element are determined and then related to one another using
the force equilibrium equations written at the nodes. These relationships, for the entire structure,
are then grouped together into what is called the structure stiffness matrix K. Once it is established,
the unknown displacements of the nodes can then be determined for any given loading on the
structure. When these displacements are known, the external and internal forces in the structure
can be calculated using the force displacement relations for each member.
Before developing a formal procedure for applying the stiffness method, it is first necessary to
establish some preliminary definitions and concepts.
Member and Node Identification: - One of the first steps when applying the stiffness method is
to identify the elements or members of the structure and their nodes. We will specify each member
by a number enclosed within a square, and use a number enclosed within a circle to identify the
nodes. Also, the “near” and “far” ends of the member must be identified. This will be done using
an arrow written along the member, with the head of the arrow directed toward the far end.
Examples of member, node, and “direction” identification for a truss are shown in Fig. 3.1a. These
assignments have all been done arbitrarily.

Global and Member Coordinates: - Since loads and displacements are vector quantities, it is
necessary to establish a coordinate system in order to specify their correct sense of direction. Here
we will use two different types of coordinate systems. A single global or structure coordinate
system, x, y, will be used to specify the sense of each of the external force and displacement
components at the nodes, Fig. 3.1a. A local or member coordinate system will be used for each
member to specify the sense of direction of its displacements and internal loadings. This system
will be identified using x ' and y ' axes with the origin at the “near” node and the x ' axis extending
toward the “far” node.
Kinematic Indeterminacy: - the unconstrained degrees of freedom for the truss represent the
primary unknowns of any displacement method, and therefore these must be identified. As a
general rule there are two degrees of freedom, or two possible displacements, for each joint (node).
For application, each degree of freedom will be specified on the truss using a code number, shown
at the joint or node, and referenced to its positive global coordinate direction using an associated
arrow. For example, the truss in Fig. 3.1a has eight degrees of freedom, which have been identified
by the “code numbers” 1 through 8 as shown. The truss is kinematically indeterminate to the fifth
degree because of these eight possible displacements: 1 through 5 represent unknown or
unconstrained degrees of freedom, and 6 through 8 represent constrained degrees of freedom. Due
to the constraints, the displacements here are zero. For later application, the lowest code numbers
will always be used to identify the unknown displacements (unconstrained degrees of freedom)
and the highest code numbers will be used to identify the known displacements (constrained
degrees of freedom).The reason for choosing this method of identification has to do with the
convenience of later partitioning the structure stiffness matrix, so that the unknown displacements
can be found in the most direct manner.
Once the truss is labeled and the code numbers are specified, the structure stiffness matrix K can
then be determined. To do this we must first establish a member stiffness matrix k’ for each member
of the truss.
Member Stiffness Matrix
We will establish the stiffness matrix for a single truss member using local x ' and y ' coordinates,
oriented as shown in Fig. 3.2. The terms in this matrix will represent the load-displacement
relations for the member.
A truss member can only be displaced along its axis (x’ axis) since the loads are applied along this
axis. Two independent displacements are therefore possible. When a positive displacement d N is
imposed on the near end of the member while the far end is held pinned, Fig. 3.2a, the forces
developed at the ends of the members are
AE AE
q 'N  dN q 'F   dN
L L
Note that q 'F is negative since for equilibrium it acts in the negative x ' direction. Likewise, a
positive displacement d F at the far end, keeping the near end pinned, Fig. 3.2b, results in member
forces of
AE AE
q ''N   dF q ''F  dF
L L
By superposition, Fig. 3.2c, the resultant forces caused by both displacements are
AE AE
qN  dN  d F ......................... 3.1
L L
AE AE
qF   dN  d F ...................... 3.2
L l

These load-displacement equations may be written in matrix form as
 qN  AE  1 1  d N 
 q   L  1 1   d 
 F   F 
Or, q  k ' d ...................... 3.3
AE  1 1
Where k ' 
L 1 1 
....................... 4.4
This matrix, k ' is called the member stiffness matrix, and it is of the same form for each member
of the truss. The four elements that comprise it are called member stiffness influence coefficients,
k 'ij . Physically, k 'ij represents the force at joint i when a unit displacement is imposed at joint j.
Displacement and Force Transformation Matrices

Since a truss is composed of many members (elements), we will now develop a method for
transforming the member forces q and displacements d defined in local coordinates to global
coordinates. For the sake of convention, we will consider the global coordinates positive x to the
right and positive y upward. The smallest angles between the positive x, y global axes and the
positive x ' local axis will be defined as  x and  y as shown in Fig. 3.3. The cosines of these angles
will be used in the matrix analysis that follows. These will be identified as x  cos  x , y  cos  y
xF  x N xF  x N
x  cos  x   ..... 3.5
L ( xF  x N ) 2  ( y F  y N ) 2
yF  y N yF  y N
 y  cos  y   ..... 3.6
L ( xF  x N ) 2  ( y F  y N ) 2
Displacement Transformation Matrix

In global coordinates each end of the member can have two degrees of freedom or independent
displacements; namely, joint N has DNx and DNy Figs. 3.4a and 3.4b, and joint F has DFx and DFy
Figs. 3.4c and 3.4d. We will now consider each of these displacements separately, in order to
determine its component displacement along the member. When the far end is held pinned and the
near end is given a global displacement DNx Fig. 3.4a, the corresponding displacement

(deformation) along the member is DNx cos  x . Likewise, a displacement DNy will cause the
member to be displaced DNy y along the x’, Fig. 3.4b. The effect of both global displacements
causes the member to be displaced

d N  DNx cos x  DNy cos y
In a similar manner, positive displacements DFx and DFy successively applied at the far end F,
while the near end is held pinned, Figs. 3.4c and 3.4d, will cause the member to be displaced
dF  DFx cos x  DFy cos y
Letting x  cos  x and y  cos  y represent the direction cosines for the member, we have
d N  DNx x  DNy  y
d F  DFx x  DFy  y
Which can be written in matrix form as
 DNx 
 
 d N  x y 0 0   DNy 
 d    0 0     D  ........................... 3.7
 F  x y Fx
 
 DFy 
Or d  TD ............................................ 3.8
 x y 0 0
Where T  
 y 
.......................................... 3.9
0 0 x
T is referred to as the displacement transformation matrix.
Force Transformation Matrix

Consider now application of the force qN to the near end of the member, the far end held pinned,
Fig. 3.5a. Here the global force components of qN at N are

QNx  qN cos x QNy  qN cos y
Likewise, if qF s applied to the bar, Fig. 3.5b, the global force components at F are
QFx  qF cos x QFy  qF cos y

Using the direction cosines x  cos  x and y  cos  y these equations become
QNx  qN x QNy  qN  y
QFx  qF x QFy  qF x
Which can be written in matrix form as

QNx   x 0
Q    0   qN 
 Ny    y
x   qF 
..................................... 3.10
 QFx   0
   
 QFy   0  y 
Or Q  T T q ............................... 3.11
 x 0
 0 
T 
T y
Where .................................. 3.12
0 x 
 
 0  y 
Member Global Stiffness Matrix

We will now combine the results of the preceding sections and determine the stiffness matrix for
a member which relates the member’s global force components Q to its global displacements D.
If we substitute Eq. 3.8 into Eq. 3.3, we can determine the member’s forces q in terms of the global
displacements D at its end points, namely,
q  k ' TD ...................................... 3.13
Substituting this equation into Eq. 3.11, yields the final result,

Q  T T k ' TD
Or Q  KD ............................. 3.14
Where k  T T k ' T ......................... 3.15
The matrix k is the member stiffness matrix in global coordinates. Since T T , T and k ' are known,
we have
 x 0
 0  AE  1 1 x y 0 0
k
y
0 x  L  1 1   0 0 x y 
 
 0  y 
Performing the matrix operations yields
Nx Ny Fx Fy
 x 2 x  y x 2  x  y 
 
AE  x  y y 2  x  y  y 2 
k .............................. 3.16
L   x 2 x  y x 2 x  y 
 
 x  y  y 2 x  y  y 2 
Truss Stiffness Matrix

Once all the member stiffness matrices are formed in global coordinates, it becomes necessary to
assemble them in the proper order so that the stiffness matrix K for the entire truss can be found.
This process of combining the member matrices depends on careful identification of the elements
in each member matrix.
Application of the Stiffness Method for Truss Analysis
Once the structure stiffness matrix is formed, the global force components Q acting on the truss
can then be related to its global displacements D using
Q  KD ............................... 3.17
This equation is referred to as the structure stiffness equation. Since we have always assigned the
lowest code numbers to identify the unconstrained degrees of freedom, this will allow us now to
partition this equation in the following form

Q k , Dk  known external loads and displacements; the loads here exist on the truss as part of the
problem, and the displacements are generally specified as zero due to support constraints
such as pins or rollers.
Qu , Du  Unknown loads and displacements; the loads here represent the unknown support
reactions, and the displacements are at joints where motion is unconstrained in a

particular direction.
K  Structure stiffness matrix, which is partitioned to be compatible with the partitioning of Q
and D.
Expanding Eq. 3.18 yields
Qk  K11 Du  K12 Dk ........................... 3.19

Qu  K21 Du  K22 Dk ........................... 3.20
E.g. 3.1. Determine the reactions and the force in each member of the trusses shown in Fig. 3.6 by
using the matrix stiffness method.
Solution: - Degrees of Freedom- From the analytical model of the truss shown in Fig. 3.6(b), we
observe that only joint 2 is free to translate. Thus the truss has six degrees of freedom, which are
the unknown displacement of joint 2 in the X and Y directions, respectively and four constrained
from displacement at joint 1 and 3. The origin of the global coordinate system can be located at
any point. The members are identified arbitrarily and arrows are written along the two members
to identify the near and far ends of each member. The direction cosines and the stiffness matrix for
each member can now be determined.

Structure Stiffness Matrix

Member 1 As shown in Fig. 18.11(b), joint 1 has been selected as the beginning joint and joint 2
as the end joint for member 1. By applying Eqs. (3.5 and 3.6), we determine
Coordinate of joint-1 (0, 2) and joint-2 ( 4, 0)
L  ( X 2  X 1 )2  (Y2  Y1 )2  (4  0)2  (0  2) 2  4.4721 m

X 2  X 1 4  0
X  cos  X    0.8944
L 4.4721
Y Y 02
Y  cos  y  2 1   0.4472
L 4.4721
The member stiffness matrix in global coordinates can now be evaluated by using Eq. (3.16)
Nx Ny Fx Fy 5 6 1 2
 x 2 x  y  x 2  x  y   0.18 0.089 0.18 0.089  5
 2  
AE  x  y y 2  x  y  y  0.089 0.0447 0.4 0.0447  6
k1   AE 
L  x 2 x  y x 2 x  y   0.18 0.089 0.18 0.089  1
 2   
 x  y  y 2 x  y  y   0.089 0.0447 0.089 0.0447  2
Member 2 As shown in Fig. 18.11(b), joint 3 has been selected as the beginning joint and joint 2
as the end joint for member 2.
Coordinate of joint-3 (0, 0) and joint-2 ( 4, 0)
L  ( X 2  X 3 ) 2  (Y2  Y3 ) 2  (4  0) 2  (0  0) 2  4 m
X 2  X 1 4  0
X  cos  X    1
L 4
Y Y 0  0
Y  cos  y  2 1  0
L 4

3 4 1 2
 0.25 0 0.25 0 3
 0 0 0 0  4
k2  AE 
 0.25 0 0.25 0 1
 
 0 0 0 0 2
Structure Stiffness Matrix- This matrix has an order of 6  6 since there are six designated
degrees of freedom for the truss, Fig. 3.6b. Corresponding elements of the above two matrices are
added algebraically to form the structure stiffness matrix. Perhaps the assembly process is easier
to see if the missing numerical columns and rows in k1 and k2 are expanded with zeros to form
two 6  6 matrices. Then
5 6 3 4 1 2
 0.18 0.089 0 0 0.18 0.089  5
 0.089 0.0447 0 0 0.089 0.0447  6

 0 0 0 0 0 0 3
k1  AE  
 0 0 0 0 0 0 4
 0.18 0.089 0 0 0.18 0.089  1
 
 0.089 0.0447 0 0 0.089 0.0447  2
5 6 3 4 1 2
0 0 0 0 0 0 5
0 0 0 0 0 0  6

0 0 0.25 0 0.25 0 3
k2  AE  
0 0 0 0 0 0 4
0 0 0.25 0 0.25 0 1
 
0 0 0 0 0 0 2
5 6 3 4 1 2
 0.18 0.089 0 0 0.18 0.089  5
 0.089 0.0447 0 0 0.089 0.0447  6

 0 0 0.25 0 0.25 0 3
K  k1  k2  AE  
 0 0 0 0 0 0 4
 0.18 0.089 0.25 0 0.43 0.089  1
 
 0.089 0.0447 0 0 0.089 0.0447  2

Displacements and Loads
By inspection it is seen that the known external displacements are D3  D4  D5  D6  0 and
also, the known external loads are Q1  0 and Q2  24 kN
0 5
0 6
 0 1
Dk    Qk   
0 3  24  2
 
0 4
Writing Eq. 3.17, Q  KD for the truss we have
From this equation it is seen that the matrix multiplication, like Eq. 3.19, yields
 0   0.43 0.089   D1 
 24   AE 0.089 0.0447   D 
    2
Here it is easy to solve by a direct expansion,

0  AE (0.43D1  0.089D2 )
24  AE (0.089D1  0.0447 D2 )
189.027 913.274
We get D1  and D2 
AE AE
Using these results, the support reactions are now obtained from
Q5   0.18 0.089   47.257 
Q   0.089 0.0447   
 6   AE   1  189.027    24 
Q3   0.25 0  AE  913.274   47.257 
     
Q4   0 0   0 
Force in each member of the trusses for  X and Y

The force in each member is found from Eq. below using the data

 DNX 
D 
AE
q   X Y X Y   NY 
 DFX 
L
 
 DFY 
Member 1: X  0.8944 and Y  0.447 L  4.4721 m
 0 
 0 
AE 1 
q1  0.8944 0.4472 0.8944 0.447 2  189.027   53.52 kN (T )
4.4721 AE
 
 913.274
Member 2: X  1 and Y  0 L4m
 0 
 0 
AE 1 
q2  1 0 1 0  189.027   47.26 kN (C )
4 AE
 
 913.274
3.2 Analyze the truss by using stiffness matrix method. Find the support reaction (A, B, C) and
deflection and also calculate forces in each member. Assume EI is constant.
Solution: - Degrees of Freedom

- Node D will have two unknown displacement components. The other displacements are
constrained.
- Then displacement at node D numbered first.

Member Stiffness matrix

Member-1
( xN , yN )  (3, 5) and ( xF , yF )  (0, 0)
L  ( X F  X N ) 2  (YF  YN ) 2  (0  3) 2  (0  (5)) 2  5.831 m

XF  XN 03
X  cos  X    0.5145
L 5.831
Y  YN 0  (5)
Y  cos  y  F   0.8575
L 5.831
Nx Ny Fx Fy 1 2 3 4
 x 2 x  y x 2  x  y   0.0454 0.0757 0.0454 0.0757  1
 2   0.0757 0.1261 0.0757 0.1261 2
AE  x  y y 2  x  y  y 
k1   AE  
L   x 2 x  y x 2 x  y   0.0454 0.0757 0.0454 0.0757  3
   
 x y  y 2 x  y  y 2   0.0757 0.1261 0.0757 0.1261  4
Member-2
( xN , yN )  (3, 5) and ( xF , yF )  (3, 0)
L  ( X F  X N ) 2  (YF  YN ) 2  (3  3) 2  (0  (5)) 2  5 m
XF  XN 33 YF  YN 0  (5)
X  cos  X   0 Y  cos  y   1
L 5 L 5

Nx Ny Fx Fy 1 2 5 6
 x 2 x  y  x 2  x  y  0 0 0 0 1
 2  
AE  x  y y 2 x  y  y  0 0.2 0 0.2  2
k2   AE 
L   x 2 x  y x 2 x  y  0 0 0 0 5
 2   
 x  y  y 2 x  y y  0 0.2 0 0.2  6
Member-3.
( xN , yN )  (3, 5) and ( xF , yF )  (7, 0)
L  ( X F  X N )2  (YF  YN ) 2  (7  3) 2  (0  (5)) 2  6.403 m

XF  XN 73 YF  YN 0  (5)
X  cos  X    0.6247 Y  cos  y    0.7809
L 6.403 L 6.403
Nx Ny Fx Fy 1 2 7 8
 x 2 x  y  x 2  x  y   0.06095 0.07619 0.06095 0.07619 1
 2   0.07619 0.09524 0.07619 0.09524 2
AE  x  y y 2  x  y  y 
k3   AE  
L   x 2 x  y x 2 x  y   0.06095 0.07619 0.06095 0.07619  7
   
 x  y  y 2 x  y  y 2   0.07619 0.09524 0.07619 0.09524  8
Structure Stiffness Matrix
This matrix has an order of 8  8 since there are six designated degrees of freedom for the truss.
K  k1  k2  k3
1 2 3 4 5 6 7 8
 0.10635 0.00049 0.0454 0.0757 0 0 0.06095 0.07619  1
 0.00049 0.42134 0.0757 0.1261 0 0.2 0.07619 0.09524  2

 0.0454 0.0757 0.0454 0.0757 0 0 0 0 3
 
0.0757 0.1261 0.0757 0.1261 0 0 0 0 4
K  AE 
 0 0 0 0 0 0 0 0 5
 
 0 0.2 0 0 0 0.2 0 0 6
 0.06095 0.07619 0 0 0 0 0.06095 0.07619  7
 
 0.07619 0.09524 0 0 0 0 0.07619 0.09524  8
Displacements and Loads
By inspection it is seen that the known external displacements are D3  D4  D5  D6  D7  D8  0

and also, the known external loads are Q1  20 kN and Q2  15 kN

0  3  Q3 
0  4 Q 
   4
0  5 D   20  1 Q 
Dk    DU   1  Qk    QU   5 
0  6  D2   15 2 Q6 
0  7 Q7 
   
0  8  Q8 
Q  KD
From this equation it is seen that the matrix multiplication yields

 20   0.10635 0.00049   D1 
 15  AE 0.00049 0.42134   D 
    2
Here it is easy to solve by a direct expansion,
20  AE (0.10635D1  0.00049 D2 )
15  AE (0.00049 D1  0.42134 D2 )
188.223 35.82
We get D1  and D2 
AE AE
Using these results, the support reactions are now obtained from
 Q3   0.0454 0.0757   11.257 
Q   0.0757 
0.1261   18.765 
 4   
Q5   0 0  1 188.223  0 
   AE    35.82    7.164 
 Q6  0  0.2  AE    
Q7   0.06095 0.07619   8.743 
     
 Q8   0.07619 0.09524   10.929 
Force in each member of the trusses
 DNX 
D 
AE
q   X Y X Y   NY 
 DFX 
L
 
 DFY 
Member 1: X  0.5145 and Y  0.8575 L  5.831 m
188.223
 
AE 1  35.82 
q1  0.5145 0.8575 0.5145 0.8575  0   21.876 kN (T )
5.831 AE
 
 0 
Member 2: X  0 and Y  1 L5m
188.223
 
AE 1  35.82 
q2   0 1 0 1  0   7.164 kN (T )
5 AE
 
 0 
Member 3: X  0.6247 and Y  0.7809 L  6.403 m

188.223
 
AE 1  35.82 
q3   0.6247 0.7809 0.6247 0.7809   0   13.995 kN (C )
6.403 AE
 
 0 
Beam Analysis Using the Stiffness Method
Before we show how the stiffness method applies to beams, we will first discuss some preliminary
concepts and definitions related to these members.
Member and Node Identification: - In order to apply the stiffness method to beams, we must
first determine how to subdivide the beam into its component finite elements. In general, each
element must be free from load and have a prismatic cross section. For this reason the nodes of
each element are located at a support or at points where members are connected together, where
an external force is applied, where the cross-sectional area suddenly changes, or where the vertical
or rotational displacement at a point is to be determined. For example, consider the beam in Fig.
3.7a. Using the same scheme as that for trusses, four nodes are specified numerically within a
circle, and the three elements are identified numerically within a square. Also, notice that the
“near” and “far” ends of each element are identified by the arrows written alongside each element.

Global and Member Coordinates: - The global coordinate system will be identified using x, y, z
axes that generally have their origin at a node and are positioned so that the nodes at other points
on the beam all have positive coordinates, Fig. 3.7a. The local or member x ', y ', z ' coordinates
have their origin at the “near” end of each element, and the positive x ' axis is directed towards the
“far” end. Figure 3.7b shows these coordinates for element 2. In both cases we have used a right-
handed coordinate system, so that if the fingers of the right hand are curled from the x( x ') axis
towards the y ( y ') axis, the thumb points in the positive direction of the z ( z ') axis, which is directed
out of the page. Notice that for each beam element the x and x ' axes will be collinear and the global
and member coordinates will all be parallel. Therefore, unlike the case for trusses, here we will
not need to develop transformation matrices between these coordinate systems.
Kinematic Indeterminacy: - Once the elements and nodes have been identified, and the global
coordinate system has been established, the degrees of freedom for the beam and its kinematic
determinacy can be determined. If we consider the effects of both bending and shear, then each
node on a beam can have two degrees of freedom, namely, a vertical displacement and a rotation.
As in the case of trusses, these linear and rotational displacements will be identified by code
numbers. The lowest code numbers will be used to identify the unknown displacements
(unconstrained degrees of freedom), and the highest numbers are used to identify the known
displacements (constrained degrees of freedom). Recall that the reason for choosing this method
of identification has to do with the convenience of later partitioning the structure stiffness matrix,
so that the unknown displacements can be found in the most direct manner.
For example, consider again the continuous beam in Fig. 3.7a. Here the beam is kinematically
indeterminate to the fourth degree. There are eight degrees of freedom, for which code numbers 1
through 4 represent the unknown displacements, and numbers 5 through 8 represent the known
displacements, which in this case are all zero.

Beam-Member Stiffness Matrix
In this section we will develop the stiffness matrix for a beam element or member having a constant
cross-sectional area and referenced from the local x ', y ', z ' coordinate system, Fig. 3.8.The origin
of the coordinates is placed at the “near” end N, and the positive x ' axis extends toward the “far”
end F. There are two reactions at each end of the element, consisting of shear forces qNy ' and qFy '
and bending moments q Nz ' and qFz ' . These loadings all act in the positive coordinate directions. In
particular, the moments q Nz ' and q Fz ' are positive counterclockwise, since by the right-hand rule the
moment vectors are then directed along the positive z ' axis, which is out of the page.
We will now impose each of these displacements separately and then determine the loadings acting
on the member caused by each displacement.
y ' Displacements: - When a positive displacement d Ny ' is imposed while other possible
displacements are prevented, the resulting shear forces and bending moments that are created are
shown in Fig. 3.9a. Likewise, when d Fy ' is imposed, the required shear forces and bending
moments are given in Fig. 3.9b.
z ' Rotations: - If a positive rotation d Nz ' is imposed while all other possible displacements are
prevented, the required shear forces and moments necessary for the deformation are shown in Fig.
3.10a. Likewise, when d Fz ' is imposed, the resultant loadings are shown in Fig. 3.10b.

By superposition, if the above results in Figs. 3.9 and 3.10 are added, the resulting four load-
displacement relations for the member can be expressed in matrix form as
Ny' Nz ' Fy ' Fz '
 qNy '   12 EI 6 EI 12 EI 6 EI 
   L3 
L2 L3 L2 
   
 qNz '   6 EI 4 EI 6 EI
 2
2 EI 
   L2 L L L 
   .......................... 3.21
 q    12 EI 6 EI
 2
12 EI 6 EI 
 2
 Fy '   L3 L L3 L 
   6 EI 2 EI 6 EI 4 EI 
    2 
 qFz '   L L 
2
L L
These equations can also be written in abbreviated form as
q  kd ........................... 3.22
The symmetric matrix k in Eq. 3.21 is referred to as the member stiffness matrix
Beam-Structure Stiffness Matrix
Once all the member stiffness matrices have been found, we must assemble them into the structure
stiffness matrix K. This process depends on first knowing the location of each element in the
member stiffness matrix.
Application of the Stiffness Method for Beam Analysis
After the structure stiffness matrix is determined, the loads at the nodes of the beam can be related
to the displacements using the structure stiffness equation
Q  KD ........................ 3.23
Here Q and D are column matrices that represent both the known and unknown loads and
displacements. Partitioning the stiffness matrix into the known and unknown elements of load and
displacement, we have
Intermediate Loadings: - For application, it is important that the elements of the beam be free of
loading along its length. This is necessary since the stiffness matrix for each element was
developed for loadings applied only at its ends. Oftentimes, however, beams will support a
distributed loading, and this condition will require modification in order to perform the matrix
analysis.

Member Forces: - The shear and moment at the ends of each beam element can be determined
using Eq. 3.22 and adding on any fixed-end reactions q0 the element is subjected to an intermediate
loading. We have
q  kd  q0 ........................... 3.24
E.g. 3.3. Determine the reactions and the member end forces if the support 2 moves upward
5mm for the beams shown in Fig. 3.11 by using the matrix stiffness method.
E= 200 Gpa and I= 15*106 mm4
Fig 3.11

Reference
1. Hibbler, R. C. Structural Analysis, 8th Edition, Prentice Hall, 2012.

2. Aslam Kassimali. Structural Analysis, 4th Edition, Indian institute of Technology, 2011.
3. Nigussie Tebedge, Methods of Structural Analysis, AAU ,1983..
4. Alan Williams. Structural Analysis in Theory and Practice, 1st Edition, Butterworth–
Heinemann, 2009.

Jimma University Jimma Institute of Technology Department of Civil and Environmental Engineering

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Jimma University Jimma Institute of Technology Department of Civil and Environmental Engineering

Uploaded by

Copyright:

Available Formats

Jimma University

Jimma Institute of Technology

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 1

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 2

In this chapter, the displacement method of analysis will be discussed.

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 3

1.3 Slope-Deflection Equations

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 4

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 5

M N  2Ek (2 N   F  3 )  ( FEM ) N .........................................8

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 6

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 7

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 8

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 9

 Pa 2b (20* 42 *1)  (20* 22 *3)

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 10

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 11

(25.6 *104 ) B  (4.8 *10 4 )C  1076.85

M BA  (16 *104 ) B  811.25  (16 *104 )(0.004126)  811.25  151.09 kNm

Shear at End span

M B  0  5(C yL )  66.464  151.04  20(3)  20(1)  0

By  ByL  ByR  225.75 kN  40.9152 kN  266.6652 kN

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 12

Pin-Supported End Span.

M N  3Ek ( N  )  ( FEM ) N ........................................................9

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 13

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 14

Substituting  B into Eqs. (1)– (3), we get

Structures with Cantilever Overhangs

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 15

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 16

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 17

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 18

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 19

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 20

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 21

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 22

F x  0 to the free body of the entire frame Fig 1.16c, to obtain

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 23

EI B  151.691 EIC  71.465 EI  70.338

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 24

M BC  EI (0.4 B  0.2C )  83.33  0.4(151.691)  0.2(71.465)  83.33  36.947 kNm

M CB  EI (0.2 B  0.4C )  83.33  0.2(151.691)  0.4( 71.465)  83.33  85.082 kNm

M CD  0.6 EI (C  )  0.6(71.465)  0.6(70.338)  85.082 kNm

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 25

MOMENT DISTRIBUTION BY CROSS METHOD

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 26

slope-deflection method we related M to  A using the conjugate-beam method, that is,

contributed by the member is M i  K i . since equilibrium requires

M  M1  M n  K1  K n    Ki then the distribution factor for the i th member is

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 27

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 28

Solution: The distribution factors at each joint must be computed first.

The fixed-end moments

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 29

JU, JIT, DEPT OF CIVIL ENG Year-2 Page 30

DF 0 0.3333 0.6667 0.6667 0.3333

FEM -77.5833 -24.25 59.1667 165.8333 -40

Dist. -11.6377 -23.279 -83.893 -41.94

CO -5.81885 -41.9465 -11.6395

Dist. 13.9808 27.9657 7.7601 3.8794

CO 6.9904 3.88005 13.98285

Dist. -1.29322 -2.5868 -9.3224 -4.6605

 Pa 2b (20* 42 1)  (20 22 *3)

(25.6 104 ) B  (4.8 10 4 )C  1076.85

M BA  (16 104 ) B  811.25  (16 104 )(0.004126)  811.25  151.09 kNm