SAINT MARY’S ANGELS COLLEGE OF PAMPANGA

Olongapo-Gapan Road, Sta. Ana, 2022, Pampanga

INSTRUCTIONAL MATERIAL
IN EARTHQUAKE ENGINEERING
CE PC 413 A

1st SEMESTER
A.Y. 2024-2025

REFERENCES:

1. PHIVOLCS. (2004). Earthquake portfolio. Manila, Philippines.

2. Association of Structural Engineers of the Philippines. (2003). Earthquake design manual, Volume 1
— Code Provisions for Lateral Forces. Manila.
3. Association of Structural Engineers of the Philippines. (2003). Earthquake design manual, Volume 2
— Earthquake Resistant Design of Structures. Manila.
4. Association of Structural Engineers of the Philippines. (2015). The national structural code of the
Philippines (2015). Manila.
5. Rotterdam, B. (2001). Dynamic analysis and earthquake resistant design. Mc Graw-Hill.
6. Hart, G. & Kai Fai Wong, K. (2000). Structural dynamics for structural engineers. Wiley and Sons.
Unit 4: Multiple Degree of Freedom Systems
OBJECTIVE

• To introduce the concept of structural dynamics

• To model structures as Multiple Degree of Freedom Systems.

INTRODUCTION

Structures cannot always be modelled as SDOF systems. In fact, structures are continuous systems and
possess infinite degrees of freedom. Multi-storey buildings are the most suitable example. A thorough
knowledge and understanding of the concepts discussed for SDOF system are of prime importance. It is
because, a structure modelled with MDOF system is transformed to consist of a number of SDOF
independent systems and then each one is solved as an SDOF system. The response spectrum of SDOF
system is extended to solve MDOF system as well.

The MDOF system may be divided into two groups according to their deformation characteristics. In one
group the floor moves only in the horizontal direction and there is no rotation of a horizontal section at
the level of floors. Such buildings are referred to as shear buildings. In the other group of structures, the
floors move in both rotational and horizontal directions and are referred to as moment-shear buildings.

For the present, we shall consider one of the most instructive and practical type of structure, which
involves many degrees of freedom, the multi-storey shear building. The following assumptions are made
about the structure:

(a) The total mass of the structure is concentrated at the levels of the floors, although it is distributed
throughout the building. This assumption is justified in case of multi-storey buildings where most of
the building mass is indeed at the floor levels. This assumption transforms the problem from a
structure with infinite degrees of freedom (due to distributed mass), to a structure that has only as
many degrees as it has lumped masses at the floor levels. For example, the structure shown below
has two degrees of freedom.

Two-degrees-of-freedom system under horizontal forces

(b) The girders on the floors are infinitely rigid as compared to the columns and the deformation of the
structure is independent of the axial forces present in the columns. This assumption introduces the
requirements that the joints between girders and columns are fixed against rotation and the girders
remain horizontal during motion.

Further, a shear building can be idealized as a single column [Fig. (b)] having concentrated masses at floor
levels, and the columns as massless springs.

The stiffness coefficient or spring constant 𝑘𝑗 is the force required to produce a unit displacement of the
two adjacent floor levels. For a uniform column with the two ends fixed against rotation, the spring
constant is 12𝐸𝐼/ℎ3 , and for a column with one end fixed and the other pinned it is 3𝐸𝐼/ℎ3, where 𝐸 is
the modulus of elasticity of the material, 𝐼 the moment of inertia, and h the height of the storey.

EQUATIONS OF MOTION

The equations of motion are developed for a simple MDOF system; a two-storey shear frame is selected
to permit easy visualization of elastic, damping, and inertial forces. The following equations of motion are
obtained for a two-storey shear building
(Eq. 4.1)

𝑓𝐼1 + 𝑓𝐷1 + 𝑓𝑆1 = 𝑝1 (𝑡)

𝑓𝐼2 + 𝑓𝐷2 + 𝑓𝑆2 = 𝑝2 (𝑡)
The inertial forces in the equations are
(Eq. 4.2)

𝑓𝐼1 = 𝑚1 𝑥̈ 1
𝑓𝐼2 = 𝑚2 𝑥̈ 2
or in matrix form
(Eq. 4.3)
𝑓 𝑚 0 𝑥̈ 1
{ 𝐼1 } = [ 1 ] { } = 𝐟𝐈 = 𝐌𝐱̈
𝑓𝐼2 0 𝑚2 𝑥̈ 2

where 𝐟𝐈 , 𝐱̈ , and 𝐌 are the inertial-force vector, acceleration vector, and mass matrix, respectively.

As shown in Fig. (b) above, the lumped masses are concentrated at floor levels, and the mass matrix is
therefore a diagonal matrix. The restoring forces and displacements are related as follows.
(Eq. 4.4)

𝑓𝑆1 = 𝑘1 𝑥1 − 𝑘2 (𝑥2 − 𝑥1 )
𝑓𝑆1 = 𝑘1 𝑥1 + 𝑘2 𝑥1 − 𝑘2 𝑥2
𝑓𝑆1 = (𝑘1 + 𝑘2 )𝑥1 − 𝑘2 𝑥2
and
(Eq. 4.5)

𝑓𝑆2 = 𝑘2 (𝑥2 − 𝑥1 ) = −𝑘2 𝑥1 + 𝑘2 𝑥2

or in matrix form
(Eq. 4.6)
𝑓𝑆1 𝑘 + 𝑘2 −𝑘2 𝑥1
{ }=[ 1 ] { } = 𝐟𝐒 = 𝐊𝐱
𝑓𝑆2 −𝑘2 𝑘2 𝑥2
where 𝐟𝐒 , 𝐱, and 𝐊 are the elastic-force vector, displacement vector, and stiffness matrix, respectively.

By introducing 𝑘11, 𝑘12, 𝑘21 , and 𝑘22

𝑘11 = 𝑘1 + 𝑘2 and 𝑘12 = −𝑘2

𝑘21 = −𝑘2 and 𝑘22 = 𝑘2

substituting into Eq (4.6), the new form of matrix can be obtained as

(Eq. 4.7)
𝑓 𝑘 𝑘12 𝑥1
{ 𝑆1 } = [ 11 ] { } = 𝐟𝐒 = 𝐊𝐱
𝑓𝑆2 𝑘21 𝑘22 𝑥2
If 𝑘𝑖𝑗 is the force applied to the 𝑖th storey when the 𝑗th storey is subjected to a unit displacement, while
all other stories remain undisplaced, then, this matrix form is true based on Maxwell-Betti reciprocal
theorem 𝑘𝑖𝑗 = 𝑘𝑗𝑖

Load and deflection of two-DOF system

If damping forces induced by viscous damping are assumed to be proportional to relative velocities then
(Eq. 4.8)
𝑓𝐷1 𝑐 +𝑐 −𝑐2 𝑥̇ 1 𝑐11 𝑐12 𝑥̇ 1
{ } = [ 1−𝑐 2 𝑐2 ] {𝑥̇ 2 } = [𝑐21 𝑐22 ] {𝑥̇ 2 } = 𝐟𝐃 = 𝐂𝐱̇
𝑓𝐷2 2

where 𝑓𝐷 , 𝑥̇ , and 𝐶 are the viscous damping-force vector, velocity vector, and viscous damping matrix,
respectively. The applied load vector is
(Eq. 4.9)

𝑝 (𝑡)
𝐩(𝐭) = { 1 }
𝑝2 (𝑡)
Using Eqns (4.3), (4.6), (4.8), and (4.9), the equations of motion for the two-degrees-of-freedom system
can be written as
(Eq. 4.10)

𝐟𝐥 + 𝐟𝐃 + 𝐟𝐒 = 𝐩(𝐭)
or,
(Eq. 4.11)

𝐌𝐱̈ + 𝐂𝐱̇ + 𝐊𝐱 = 𝐩(𝐭)

This expression is essentially in the same form as the equation for an SDOF system
If ground acceleration, 𝑥̈𝑔 , is applied to the structure, then
(Eq. 4.12)

𝐌𝐱̈ + 𝐂𝐱̇ + 𝐊𝐱 = −𝐦𝐈𝐱̈ 𝐠

where 𝐦 is the lumped mass matrix containing only diagonal terms, 𝐊 is a banded matrix, and 𝐈 is a unit
vector containing ones corresponding to DOFs in the direction of earthquake load, and zeroes elsewhere.

PERIODS AND MODE OF VIBRATION OF MDOF SYSTEMS

For an undamped multi-degrees-of-freedom (MDOF) system in free vibration, Eq. (4.12) reduces to
(Eq. 4.13)

𝐌𝐱̈ + 𝐊𝐱 = 𝟎
The solution of Eqn (4.13) is assumed to be
(Eq. 4.14)

𝑥 = 𝑎 sin(𝜔𝑡 + 𝜙)
where 𝑎 represents the vibrational shape (mode shape) of the system that does not change with respect
to time and 𝜙 is the phase angle. The second derivative of this equation can be obtained as
(Eq. 4.15)

𝑥̈ = −𝜔2 𝑎 sin(𝜔𝑡 + 𝜙) = −𝜔2 𝑥

Then, substitution of (4.14) and (4.15) into (4.13) yields:
(Eq. 4.16)

−𝜔2 𝐌𝑎 sin(𝜔𝑡 + 𝜙) + 𝐊𝑎 sin(𝜔𝑡 + 𝜙) = 0

(𝐊 − 𝝎𝟐 𝐌)𝑎 sin(𝜔𝑡 + 𝜙) = 0

Since the sine term changes with time, the following relation must be satisfied:
(Eq. 4.17)

(𝐊 − 𝝎𝟐 𝐌)𝑎 = 0

In order for a non-trivial solution of 𝑎 to be possible, the determinant of (𝐊 − 𝝎𝟐 𝐌) must be zero. Then,
(Eq. 4.18)
|𝐊 − 𝝎𝟐 𝐌| = 𝟎

An 𝑁-DOF system can thus vibrate in 𝑁 different modes, each having a different mode shape and each
occurring at a particular natural frequency. The lowest value of 𝜔 is called the first natural circular
frequency 𝝎𝟏 .

The 𝜔 are numbered sequentially so that the 𝑛th lowest value of 𝜔 is the 𝑛th natural circular frequency;
by substituting it into Eqn (4.17), the relative displacements 𝑥 of the system, which represent the shape
of vibration or the modal shape, can be determined. For a two-DOF system, Eq (4.17) would become
(Eq. 4.19)

(𝑘11 − 𝜔2 𝑀1 )𝑎1 + 𝑘12 𝑎2 = 0

𝑘21 𝑎1 + (𝑘22 − 𝜔2 𝑀2 )𝑎2 = 0
For 𝑎 to have a nontrivial solution, the determinant of Eqn (4.19) must be zero. Hence,
(Eq. 4.20)

𝑘 − 𝜔2 𝑀1 𝑘12
[ 11 ]=0
𝑘21 𝑘22 − 𝜔2 𝑀2
or,
(Eq. 4.21)
(𝑀1 𝜔2 − 𝑘11 )(𝑀2 𝜔2 − 𝑘22 ) − 𝑘12 𝑘21 = 0

the ratio of displacements, 𝑎2 /𝑎1 , is uniquely determined for each 𝜔1 and 𝜔2 as shown in figure below.
The modal shapes corresponding to 𝜔1 and 𝜔2 are called the first and second mode, respectively. As
evident from the condition specified by Eq. (4.19), only displacement ratios of 𝑎 can be obtained. In usual
practice, the maximum displacement corresponding to the top or the lowest storey is taken to be unity. If
a system has 𝑁 degrees of freedom, then the 𝑛th modal shape n is written as
(Eq. 4.22)
𝜙1𝑛 𝑎1𝑛
𝜙 1 𝑎2𝑛
𝜙𝑛 = [ 2𝑛 ] = [ ]
⋮ 𝑎𝑘𝑛 ⋮
𝜙𝑁𝑛 𝑎𝑁𝑛

Here, 𝑎 represents the reference component. The square matrix, consisting of 𝑛-modal-shape vectors, is
called the modal-shape matrix and is expressed as
(Eq. 4.23)
𝜙11 𝜙12 ⋯ 𝜙1𝑁
𝜙 𝜙22 ⋯ 𝜙2𝑁
𝜙 = [𝜙1 𝜙2 … 𝜙𝑁 ] = [ 21 ]
⋮ ⋮ ⋱ ⋮
𝜙𝑁1 𝜙𝑁2 ⋯ 𝜙𝑁𝑁
Modal-shape vectors possess an orthogonality relationship for elastic systems.

PROBLEM 1

A model of two-storey RCC frame is shown in figure below. Determine the natural frequency, assuming
the beam–column joints to be rigid, for the following data.

Dimensions of columns: 250 × 250 mm

Storey height: 3 m
Modulus of Elasticity, E = 2.5 × 1010 N/m2
Solution:

Moment of Inertia of the column:

0.25(0.25)3
𝐼= = 3.2552 × 10−4 𝑚4
12
Stiffness of each storey:

12𝐸𝐼 12(2.5 × 1010 𝑁/𝑚2 )(3.2552 × 10−4 𝑚4 )

𝑘 + 𝑘 = 2𝑘 = 2 ( 3 ) = 2 [ ] = 7234 × 103 𝑁/𝑚
ℎ (3 𝑚)3

Refer to figure below (convert the structure to spring mass model)

𝑚1 = 30 × 1000 = 30,000 𝑘𝑔
𝑚2 = 15 × 1000 = 15,000 𝑘𝑔
Stiffness matrix:
𝑘1 + 𝑘2 −𝑘2 7234 + 7234 −7234 14,468 −7234
[ ]=[ ]=[ ]
−𝑘2 𝑘2 −7234 7234 −7234 7234
Mass matrix:
30 0
[ ]
0 15
The frequency equation becomes;

14,468 − 30𝜔2 −7234

|𝐊 − 𝝎𝟐 𝐌| = | |=0
−7234 7234 − 15𝜔2
104,661,512 − 434,040𝜔2 + 450𝜔4 − 52,330,756 = 0
450𝜔4 − 434,040𝜔2 + 52330756 = 0
Using quadratic equation and setting the variable 𝑥 as 𝜔2

2
−(−434,040) ± √(−434,040)2 − 4(450)(52330756)
𝜔1,2 =
2(450)
𝜔12 = 823.2807
𝜔22 = 141.252
so

𝝎𝟏 = 𝟐𝟖. 𝟔𝟗𝟑 𝒓𝒂𝒅/𝒔

𝝎𝟐 = 𝟏𝟏. 𝟖𝟖𝟓 𝒓𝒂𝒅/𝒔
PROBLEM 2

Calculate the natural periods and mode shapes of the frame shown on problem no. 1.

Solution:

For natural periods:

2𝜋 2𝜋
𝑇1 = = = 0.219 𝑠𝑒𝑐
𝜔1 28.693
2𝜋 2𝜋
𝑇2 = = = 0.529 𝑠𝑒𝑐
𝜔2 11.885
Calculate the modes

For 𝝎𝟐𝟏 = 𝟖𝟐𝟑. 𝟐𝟖𝟎𝟕:

𝑘11 − 𝜔2 𝑀1 𝑘12 𝑎1 0
[ 2 ] {𝑎 } = { }
𝑘21 𝑘22 − 𝜔 𝑀2 2 0
14,468 − 30(823.2807) −7234 𝑎1 0
[ ] {𝑎 } = { }
−7234 7234 − 15(823.2807) 2 0
−10230 −7234 𝑎1 0
[ ]{ } = { }
−7234 −5115 𝑎2 0
Taking either equation, we calculate:

−10230𝑎1 − 7234𝑎2 = 0 → 𝑎1 = −0.707𝑎2

−7234𝑎1 − 5115𝑎2 → 𝑎1 = −0.707𝑎2
𝑎2 1
1𝑠𝑡 𝑀𝑜𝑑𝑒 = = = −𝟏. 𝟒𝟏𝟒
𝑎1 −0.707
Therefore the mode matrix can be written as:
𝟏
𝝓𝟏 = { }
−𝟏. 𝟒𝟏𝟒
For 𝝎𝟐𝟐 = 𝟏𝟒𝟏. 𝟐𝟓𝟐:

𝑘11 − 𝜔2 𝑀1 𝑘12 𝑎1 0
[ ]{ } = { }
𝑘21 𝑘22 − 𝜔2 𝑀2 𝑎2 0
14,468 − 30(141.252) −7234 𝑎1 0
| |{ } = { }
−7234 7234 − 15(141.252) 𝑎2 0
10230 −7234 𝑎1 0
| |{ } = { }
−7234 5115 𝑎2 0
Taking either equation, we calculate:

10230𝑎1 − 7234𝑎2 = 0 → 𝑎1 = 0.707𝑎2

−7234𝑎1 + 5115𝑎2 → 𝑎1 = 0.707𝑎2
𝑎2 1
2𝑛𝑑 𝑀𝑜𝑑𝑒 = = = 𝟏. 𝟒𝟏𝟒
𝑎1 0.707
Therefore the mode matrix can be written as:
𝟏
𝝓𝟏 = { }
𝟏. 𝟒𝟏𝟒
Then the mode shape matrix is given as
𝟏 𝟏
𝚽=[ ]
−𝟏. 𝟒𝟏𝟒 𝟏. 𝟒𝟏𝟒
Mode Shapes

ORTHOGONALITY PROPERTY OF THE NORMAL MODES

This property constitutes the basis of the most important method for solving dynamic problems, the
Modal Superposition Method of multi-degree-of-freedom systems. We begin by rewriting the equations
of motion in free vibration, in Eq. (4.13) as
(Eq. 4.24)

𝐊𝐚 = 𝝎𝟐 𝐌𝐚
For the two-degree-of-freedom system, we obtain from Eq. (4.19) as
(Eq. 4.25)

(𝑘1 + 𝑘2 )𝑎1 − 𝑘2 𝑎2 = 𝜔2 𝑚1 𝑎1
−𝑘2 𝑎1 + 𝑘2 𝑎2 = 𝜔2 𝑚2 𝑎2
These equations are exactly the same as Eq. (4.19) but written in this form they may be given a static
interpretation as the equilibrium equations for the system acted on by forces of magnitude 𝜔2 𝑚1 𝑎1 and
𝜔2 𝑚2 𝑎2 applied to masses 𝑚1 and 𝑚2 , respectively.

The modal shapes may then be considered as the static deflections resulting from the forces on the right-
hand side of Eq. (4.25) for any of the two modes. This interpretation, as a static problem, allows us to use
the results of the general static theory of linear structures. In particular, we may use of Betti’s theorem
which states: For a structure acted upon by two systems of loads and corresponding displacements, the
work done by the first system of loads moving through the displacements of the second system is equal
to the work done by this second system of loads undergoing the displacements produced by the first load
system. The two systems of loading and corresponding displacements which we shall consider are as
follows:
System I:

Forces 𝜔12 𝑚1 𝑎11 , 𝜔12 𝑚2 𝑎21

Displacements 𝑎11 , 𝑎21

System II:

Forces 𝜔22 𝑚1 𝑎12 , 𝜔22 𝑚2 𝑎22

Displacements 𝑎12 , 𝑎22

The application of Betti’s theorem for these two systems yields

(Eq. 4.26)

𝜔12 𝑚1 𝑎11 𝑎12 + 𝜔12 𝑚2 𝑎21 𝑎22 = 𝜔22 𝑚1 𝑎12 𝑎11 + 𝜔22 𝑚2 𝑎22 𝑎21
or
(Eq. 4.27)

(𝜔12 − 𝜔22 )(𝑚1 𝑎11 𝑎12 + 𝑚2 𝑎21 𝑎22 ) = 0

If the natural frequencies are different (𝜔1 ≠ 𝜔2 ), it follows that

(Eq. 4.28)

𝑚1 𝑎11 𝑎12 + 𝑚2 𝑎21 𝑎22 = 0

which is the so-called orthogonality relationship between modal shapes of a two degree-of-freedom
system. For an n-degree-of-freedom system in which the mass matrix is diagonal, the orthogonality
condition between any two modes i and j may be expressed as
(Eq. 4.29)
𝑛

∑ 𝑚𝑘 𝑎𝑘𝑖 𝑎𝑘𝑗 = 0, 𝑓𝑜𝑟 𝑖 ≠ 𝑗

𝑘=1

and in general for any n-degree-of-freedom system as

(Eq. 4.30)

{𝑎}𝑇𝑖 [𝑀]{𝑎}𝑗 = 0 𝑓𝑜𝑟 𝑖 ≠ 𝑗

in which {𝑎𝑖 } and {𝑎𝑗 } are any two modal vectors and [𝑀] is the mass matrix of the system.

As mentioned before, the amplitudes of vibration in a normal mode are only relative values which may
be scaled or normalized to some extent as a matter of choice. The following is an especially convenient
normalization for a general system:
(Eq. 4.31)
𝑎𝑖𝑗
𝜙𝑖𝑗 =
√{𝑎}𝑇𝑖 [𝑀]{𝑎}𝑗

which, for a system having a diagonal mass matrix, may be written as

(Eq. 4.32)
𝑎𝑖𝑗
𝜙𝑖𝑗 =
√∑𝑛𝑘=1 𝑚𝑘 𝑎𝑘𝑗
2
in which 𝜙𝑖𝑗 is the normalized 𝑖 component of the 𝑗 modal vector. For normalized eigenvectors, the
orthogonality condition is given by
(Eq. 4.33)

{𝜙}𝑇𝑖 [𝑀]{𝜙}𝑗 = 0, 𝑓𝑜𝑟 𝑖 ≠ 𝑗

{𝜙}𝑇𝑖 [𝑀]{𝜙}𝑗 = 1, 𝑓𝑜𝑟 𝑖 = 𝑗

Another orthogonality condition is obtained by writing Eq. (4.17) for the normalized j mode as
(Eq. 4.34)

[𝐾]{𝜙}𝑗 = 𝜔𝑗2 [𝑀]{𝜙}𝑗

Then pre-multiplying Eq. (4.34) by {𝜙}𝑇𝑖 i we obtain, in view of Eq. (4.33), the following orthogonality
condition between eigenvectors:
(Eq. 4.35)

{𝜙}𝑇𝑖 [𝐾]{𝜙}𝑗 = 0, 𝑓𝑜𝑟 𝑖 ≠ 𝑗

{𝜙}𝑇𝑖 [𝐾]{𝜙}𝑗 = 𝜔𝑗2 , 𝑓𝑜𝑟 𝑖 = 𝑗

PROBLEM 3

For the two-story shear building of PROBLEM 1, determine (a) the normalized modal shapes of vibration,
and (b) verify the orthogonality condition between the modes.

2
2
√ ∑ 𝑚𝑘 𝑎𝑘1 = √30(1)2 + 15(1.414)2 = 2√15 = 7.746
𝑘=1

2
2
√ ∑ 𝑚𝑘 𝑎𝑘2 = √30(1)2 + 15(−1.414)2 = 2√15 = 7.746
𝑘=1

Consequently, the normalized modes are

1 1
𝜙11 = = 0.1291, 𝜙21 = = 0.1291
2√15 2√15
1.414 −1.414
𝜙12 = = 0.1825, 𝜙22 = = −0.1825
2√15 2√15
The modal matrix is
0.1291 0.1291
𝚽=[ ]
0.1825 −0.1825
Check the orthogonality condition

{𝜙}𝑇𝑖 [𝑀]{𝜙}𝑗 = [0.1291 0.1825 ] [30 0 ] [0.1291 0.1291 ] = [1 0]

0.1291 −0.1825 0 15 0.1825 −0.1825 0 1
Therefore, the orthogonality condition satisfied.