Detailed Notes: MDOF Structural Dynamics – A Problem-Solving Framework

The key to solving MDOF problems isn't just memorizing formulas, but understanding the
underlying principles and the steps involved in analyzing a dynamic system.

I. Understanding the System: The Foundation

Before touching any formulas, you need to thoroughly understand the problem.

1. Degrees of Freedom (DOFs):

o Definition: The minimum number of independent coordinates required to define
the position of all parts of the system.
o Identification: For structural systems, this usually corresponds to the number of
stories/masses and the type of motion (translational, rotational, or both).
o Crucial for MDOF: This determines the size of your matrices (N×N for an N-
DOF system).
2. System Parameters:
o Mass Matrix (M): Represents the inertia of the system.
 Lumped Mass: Often a diagonal matrix where mii is the mass associated
with the i-th DOF. This is the simplest and most common assumption for
buildings.
 Consistent Mass: More complex, considering distributed mass. Less
common in introductory problems.
o Stiffness Matrix (K): Represents the elastic properties of the system.
 Assembling K: This is crucial. For buildings, it's typically formed by
considering the lateral stiffness of each story. For other systems, it
involves individual spring stiffnesses and their connectivity. Think about
how a displacement at one DOF affects forces at all DOFs.
 Symmetry: K is always symmetric (kij=kji).
o Damping Matrix (C): Represents energy dissipation.
 Types: Viscous damping is most common.
 Modeling: Often assumed to be proportional (Rayleigh damping,
C=αM+βK) or in specific cases, modal damping (damping ratio for each
mode). If not given, you might assume undamped for initial analysis, then
introduce damping.
o External Force Vector (F(t)): The dynamic loads acting on the system.
 Types: Harmonic (sinusoidal), impulse, step, earthquake ground motion,
etc.
 Spatial Distribution: How the forces are applied across the DOFs.

II. Equation of Motion: The Governing Equation

The fundamental equation governing the dynamic behavior of an MDOF system is:

Mu¨(t)+Cu˙(t)+Ku(t)=F(t)
Where:

 u(t) is the displacement vector.

 u˙(t) is the velocity vector.
 u¨(t) is the acceleration vector.

III. Free Vibration Analysis: Unveiling System Characteristics

This is the starting point for almost all MDOF problems. It helps you understand the inherent
dynamic properties of the structure.

1. Assumption: No external forces (F(t)=0) and no damping (for undamped free vibration).
Mu¨+Ku=0
2. Assumed Solution: Harmonic motion u(t)=ϕsin(ωt+α) or u(t)=ϕeiωt. Substituting this
leads to the eigenvalue problem: (K−ω2M)ϕ=0
3. Solving the Eigenvalue Problem:
o Characteristic Equation: det(K−ω2M)=0. Solving this polynomial in ω2 gives
you N eigenvalues.
o Eigenvalues (ωn2): Square of the natural circular frequencies for each mode.
o Natural Frequencies (fn): fn=ωn/(2π)
o Natural Periods (Tn): Tn=1/fn=2π/ωn
o Eigenvectors (ϕn): For each ωn, substitute back into (K−ωn2M)ϕn=0 and solve
for the mode shape vector ϕn. These vectors represent the relative displacements
of the DOFs in that particular mode. They are unique up to a scalar multiplier.
o Normalization: Mode shapes are often normalized (e.g., maximum component is
1, or mass normalized such that ϕnTMϕn=1). Mass normalization is very useful
for modal analysis.
4. Orthogonality Properties of Mode Shapes:
o ϕiTMϕj=0 for i=j (M-orthogonality)
o ϕiTKϕj=0 for i=j (K-orthogonality)
o ϕiTMϕi=Mi∗ (Generalized Mass for mode i)
o ϕiTKϕi=Ki∗ (Generalized Stiffness for mode i)
o ωi2=Ki∗/Mi∗

IV. Forced Vibration Analysis: Response to External Loads

This is where you use the system's inherent properties to predict its response to dynamic forces.

1. Modal Analysis (Superposition of Modes):

o Concept: The total response of an MDOF system can be expressed as a linear
combination of its mode shapes, each oscillating with its own natural frequency
and subjected to a generalized force.
o Transformation: Introduce modal coordinates u(t)=∑n=1Nϕnqn(t)=Φq(t), where
Φ is the modal matrix (columns are mode shapes) and q(t) is the vector of modal
coordinates.
 o Decoupling: Substituting this into the EOM and pre-multiplying by ΦT (and
using orthogonality) decouples the N coupled equations into N independent
SDOF equations in terms of qn(t): Mn∗q¨n(t)+Cn∗q˙n(t)+Kn∗qn(t)=Fn∗(t)
Where:
 Mn∗=ϕnTMϕn (Generalized Mass)
 Kn∗=ϕnTKϕn (Generalized Stiffness)
 Cn∗=ϕnTCϕn (Generalized Damping, if Rayleigh damping is assumed)
 Fn∗(t)=ϕnTF(t) (Generalized Force for mode n)
2. Solving SDOF Equations:
o Each of these SDOF equations is solvable using standard SDOF techniques
(Duhamel's integral for arbitrary loads, steady-state response for harmonic loads,
etc.).
o For a given forcing function, you solve for qn(t). This will depend on the type of
forcing function.
o Damping Ratio (ζn): For each mode, the damping ratio is ζn=Cn∗/(2ωnMn∗).
This is usually given or assumed for each mode.
3. Reconstructing the Response:
o Once you have all qn(t), sum them up to get the total displacement response:
u(t)=∑n=1Nϕnqn(t).
o Similarly for velocities and accelerations: u˙(t)=∑n=1Nϕnq˙n(t) and u¨(t)=∑n=1N
ϕnq¨n(t).

V. Specific Problem Types & Considerations

 Harmonic Excitation (F(t)=F0sin(Ωt)):

o Solve for steady-state response of each SDOF system.
o qn(t)=Qn0sin(Ωt−ψn), where Qn0 and ψn depend on Ω, ωn, ζn, and Fn∗.
o Pay attention to resonance when Ω≈ωn.
 Earthquake Excitation (F(t)=−Mu¨g(t)):
o The effective force vector is Feff(t)=−Mru¨g(t), where r is the influence vector
(vector of ones for translational ground motion).
o Generalized force: Fn∗(t)=−ϕnTMru¨g(t).
o Modal participation factor for mode n: Γn=ϕnTMϕnϕnTMr.
o The SDOF equation becomes: q¨n(t)+2ζnωnq˙n(t)+ωn2qn(t)=−Γnu¨g(t).
o The solution for qn(t) is often obtained by solving the SDOF system for a unit
impulse and using Duhamel's integral, or by using a response spectrum if the peak
response is needed.
o Modal Combination Rules: For peak response (e.g., using a response spectrum),
you can't simply sum the peak modal responses because they occur at different
times. Common rules:
 SRSS (Square Root of Sum of Squares): Response=∑(Responsen)2
(Most common for well-separated modes).
 CQC (Complete Quadratic Combination): More accurate, especially for
closely spaced modes, considers cross-correlation between modes.
  Initial Conditions: For transient response, consider initial displacements u(0) and
velocities u˙(0). These can be transformed into initial modal coordinates qn(0) and q˙n(0)
using modal expansion.

VI. Steps to Solve an MDOF Problem (Conceptual Flow)

1. Define the System:

o Identify DOFs.
o Formulate Mass Matrix (M).
o Formulate Stiffness Matrix (K).
o Identify Damping (if any, typically given as damping ratios).
o Identify External Force Vector (F(t)) or ground motion.
o Identify Initial Conditions (if any).
2. Free Vibration Analysis (Always the first step for dynamic analysis!):
o Set up the eigenvalue problem: (K−ω2M)ϕ=0.
o Calculate natural frequencies (ωn, fn, Tn).
o Calculate mode shapes (ϕn).
o Normalize mode shapes (e.g., mass normalization: ϕnTMϕn=1).
o Calculate generalized masses Mn∗ and generalized stiffnesses Kn∗.
3. Forced Vibration Analysis (if applicable):
o Modal Decomposition: Transform the MDOF problem into N uncoupled SDOF
problems using u(t)=Φq(t).
o Calculate Generalized Forces: Fn∗(t)=ϕnTF(t) (or for earthquake, Fn∗(t)=−Γn
Mn∗u¨g(t)).
o Determine Modal Damping Ratios: ζn (usually given or assumed).
o Solve Each SDOF Problem for qn(t): Use appropriate SDOF techniques
(Duhamel's integral, steady-state harmonic response, etc.).
o Reconstruct Response: Sum the modal responses to get the total displacement,
velocity, and acceleration at each DOF: u(t)=∑ϕnqn(t).
4. Calculate Desired Quantities:
o Story shears, story forces, base shear, drift, etc. (These are derived from the
displacements and accelerations).
o For shears, remember Vi=∑j=iNmju¨j (inertial forces) or using story stiffnesses:
Vi=ki(ui−ui−1).

Cheat Sheet: MDOF Structural Dynamics

I. Fundamental Equation of Motion: Mu¨(t)+Cu˙(t)+Ku(t)=F(t)

II. Free Vibration Analysis (Undamped):

1. Eigenvalue Problem: (K−ω2M)ϕ=0

2. Characteristic Equation: det(K−ω2M)=0⟹ωn2 (Eigenvalues)
3. Natural Frequencies/Periods:
o ωn=ωn2 (rad/s)
 o fn=ωn/(2π) (Hz)
o Tn=1/fn=2π/ωn (s)
4. Mode Shapes: For each ωn, solve (K−ωn2M)ϕn=0 for ϕn (Eigenvectors).
5. Normalization: Often mass normalized such that ϕnTMϕn=1.

III. Orthogonality of Mode Shapes:

 ϕiTMϕj=0 if i=j
 ϕiTKϕj=0 if i=j
 Generalized Mass: Mn∗=ϕnTMϕn
 Generalized Stiffness: Kn∗=ϕnTKϕn=ωn2Mn∗

IV. Modal Analysis (Forced Vibration):

1. Modal Transformation: u(t)=Φq(t)=∑n=1Nϕnqn(t)

o Φ=[ϕ1 ϕ2 … ϕN] (Modal Matrix)
2. Decoupled SDOF Equations: Mn∗q¨n(t)+Cn∗q˙n(t)+Kn∗qn(t)=Fn∗(t) or q¨n(t)+2ζn
ωnq˙n(t)+ωn2qn(t)=Fn∗(t)/Mn∗
3. Generalized Damping: Cn∗=ϕnTCϕn=2ζnωnMn∗ (if proportional damping)
4. Generalized Force: Fn∗(t)=ϕnTF(t)

V. Special Cases for Forcing:

 Harmonic Excitation (F(t)=F0sin(Ωt)):

o Fn∗(t)=(ϕnTF0)sin(Ωt)=Fn0∗sin(Ωt)
o Steady-state modal response: qn(t)=Dnsin(Ωt−ψn)
 Dn=Kn∗Fn0∗(1−(Ω/ωn)2)2+(2ζnΩ/ωn)21
 tanψn=1−(Ω/ωn)22ζnΩ/ωn
 Earthquake Excitation (F(t)=−Mru¨g(t)):
o Influence Vector: r=[1 1… 1]T (for translational ground motion)
o Modal Participation Factor: Γn=ϕnTMϕnϕnTMr=Mn∗ϕnTMr
o Decoupled Equation: q¨n(t)+2ζnωnq˙n(t)+ωn2qn(t)=−Γnu¨g(t)
o Modal Displacement: un(t)=ϕnqn(t)
o Peak Modal Response (from Response Spectrum): qnmax=ΓnSd(ωn,ζn) or q¨
nmax=ΓnSa(ωn,ζn).

VI. Reconstructing Total Response:

 u(t)=∑n=1Nϕnqn(t)
 u˙(t)=∑n=1Nϕnq˙n(t)
 u¨(t)=∑n=1Nϕnq¨n(t)

VII. Modal Combination Rules (for Peak Response):

 SRSS (Square Root of Sum of Squares): If modes are well-separated (Ti/Tj<0.9 or

1.1): Rmax=∑n=1N(Rnmax)2
  CQC (Complete Quadratic Combination): More accurate for closely spaced modes
(formula involves correlation coefficients). For exam purposes, SRSS is usually sufficient
unless specified.

VIII. Derived Quantities:

 Story Shear at Level i (for building frames):

o Vi=ki(ui−ui−1) (relative displacement)
o Vi=∑j=iNmju¨j (sum of inertial forces above story i)
 Story Force at Level i: Fi=miu¨i (if considering only inertial forces)
 Base Shear: Vb=∑j=1Nmju¨j (total inertial force

Given Information:

 Building Type: 3-story, 2-bay frame (from Figure a)

 Damping Ratio: ξ=2%=0.02 (of critical damping)
 Girder Dimensions: b=0.40 m (width), h=0.50 m (depth)
 Column Dimensions: Square, side H=0.40 m
 Modulus of Elasticity: E=25 GPa =25×109 N/m$^2$
 Mass per Unit Area: 1000 kg/m$^2$
 Tributary Areas for Pressure:
o Top story: A3=10m×1.5m=15 m2
o Other floors (Story 1 & 2): A1=A2=10m×3m=30 m2
 Given Matrices:
o Stiffness Matrix: [Ks] (3x3)
o Mass Matrix: [M] (3x3)
o Modal Matrix (Eigenvectors): [Φ] (3x3)
 Pressure Wave: Triangular pulse (from Figure b - Need P0 and td from this graph!)

Steps to Solve:

Step 1: Define the Forcing Function P(t)

The pressure wave is a triangular pulse acting uniformly on the façade. Let the peak pressure be
P0 (from Figure b, y-axis peak) and the duration of the pulse be td (from Figure b, x-axis end of
triangle).

The pressure function p(t) can be defined as: p(t)={P0(1−tdt)00≤t≤tdt>td

Now, calculate the force acting on each floor. Since the pressure is uniform, the force on each
story is the pressure multiplied by its tributary area:

 Force on Story 3 (top): F3(t)=p(t)×A3=p(t)×15 m2

 Force on Story 2: F2(t)=p(t)×A2=p(t)×30 m2
  Force on Story 1: F1(t)=p(t)×A1=p(t)×30 m2

Form the force vector P(t): P(t)=⎩⎨⎧F1(t)F2(t)F3(t)⎭⎬⎫=p(t)⎩⎨⎧303015⎭⎬⎫

Step 2: Verify/Obtain Modal Properties (Natural Frequencies and Modal Matrix)

You are given [M], [Ks], and [Φ].

1. Natural Frequencies (ωn): The natural frequencies are crucial. They should be
obtainable from the "Mode" table in your exam paper, or you would typically find them
by solving the eigenvalue problem: ([Ks]−ωn2[M]){ϕn}={0}.
o Crucial: Read ω1,ω2,ω3 directly from the "Mode" table. These are essential for
the next steps.
2. Modal Mass (Mn): For each mode n, the modal mass is calculated as: Mn=ϕnT[M]ϕn
(Where ϕn is the n-th column of [Φ])
3. Modal Stiffness (Kn): For each mode n, the modal stiffness is: Kn=ϕnT[Ks]ϕn (You
should verify that Kn=ωn2Mn)
4. Modal Damping (Cn): For each mode n, the modal damping is: Cn=2ξωnMn Since ξ is
given as 0.02 for all modes, calculate Cn for each ωn.

Step 3: Calculate Modal Forcing Functions (Pn(t))

For each mode n, the modal forcing function is: Pn(t)=ϕnTP(t)

Substitute the ϕn (columns of [Φ]) and the P(t) vector from Step 1. You will get three separate
modal forcing functions, one for each mode.

Step 4: Solve Uncoupled Modal Equations of Motion

For each mode n, the uncoupled equation of motion in modal coordinates (Yn(t)) is: MnY¨n(t)
+CnY˙n(t)+KnYn(t)=Pn(t)

Divide by Mn: Y¨n(t)+MnCnY˙n(t)+MnKnYn(t)=MnPn(t)

This simplifies to: Y¨n(t)+2ξωnY˙n(t)+ωn2Yn(t)=MnPn(t)

This is the equation for a damped single-degree-of-freedom (SDOF) system under excitation.
Since the forcing function Pn(t) (and thus MnPn(t)) is a triangular pulse, you will use the
response solution for a damped SDOF system subjected to a triangular pulse load.

Response for a Damped SDOF to a Triangular Pulse:

The solution for Yn(t) will depend on two phases:

 Phase 1: 0≤t≤td (during the pulse)

 Phase 2: t>td (after the pulse)
The general solution for a triangular pulse is complex, involving the Duhamel integral. However,
for common cases, there are tabulated or formula-based solutions.

Let ωdn=ωn1−ξ2 be the damped natural frequency.

The dynamic load factor (DLF) for a triangular pulse is often used. Let Yst,n(t)=Pn(t)/Kn
=Pn(t)/(ωn2Mn) be the static displacement.

The response Yn(t) for 0≤t≤td is: Yn(t)=Yst,max,n[1−tdt−e−ξωnt(cos(ωdnt)+1−ξ2ξsin(ωdnt))+td

ωdne−ξωntsin(ωdnt)] This formula is for the case where P0 is the peak value. You'll need to
apply it carefully, ensuring Yst,max,n correctly reflects the peak static response for the modal
force.

A simpler approach is to use numerical integration (if allowed and feasible) or to break down the
triangular pulse into Ramp functions and superimpose the solutions.

A common way to express the response to a triangular pulse: For 0≤t≤td: Yn(t)=KnP0[(1−tdt)
−e−ξωnt(cos(ωdnt)+1−ξ2ξsin(ωdnt))+tdωdn1e−ξωntsin(ωdnt)]

For t>td: The response is the free vibration response from the state (Yn(td),Y˙n(td)) at t=td. Yn
(t)=e−ξωn(t−td)[Yn(td)cos(ωdn(t−td))+ωdnY˙n(td)+ξωnYn(td)sin(ωdn(t−td))] (You'll need to
calculate Yn(td) and Y˙n(td) by evaluating the 0≤t≤td solution at t=td).

Step 5: Combine Modal Responses to Get Total Response u(t)

Once you have Y1(t), Y2(t), and Y3(t) for each mode, you can find the total displacement vector
u(t) by modal superposition:

u(t)=[Φ]{Y(t)}

⎩⎨⎧u1(t)u2(t)u3(t)⎭⎬⎫=ϕ11ϕ21ϕ31ϕ12ϕ22ϕ32ϕ13ϕ23ϕ33⎩⎨⎧Y1(t)Y2(t)Y3(t)⎭⎬⎫

This will give you the displacement of each floor (u1(t),u2(t),u3(t)) as a function of time.

Okay, let's focus solely on finding the mathematical equation for the provided graph, which
shows "pressure (kPa)" on the y-axis and "t (s)" on the x-axis. As identified before, it's a
piecewise linear function.

We'll define the function P(t) for pressure at time t.

Identifying the Key Points (Nodes):

From the graph, we can identify the coordinates of the vertices where the slope changes:
1. Point 1 (Start): (t0,P0)=(0,0)
2. Point 2 (Peak): (t1,P1)=(0.1,5)
3. Point 3 (Zero Crossing): (t2,P2)=(0.35,0)
4. Point 4 (End): (t3,P3)=(0.5,−1)

Now, we'll determine the equation for each linear segment using the slope-intercept form
(y=mx+b) or point-slope form (y−y1=m(x−x1)).

Segment 1: From t=0 to t=0.1 s

 Points: (0,0) and (0.1,5)

 Slope (m1): m1=t1−t0P1−P0=0.1−05−0=0.15=50
 Y-intercept (b1): Since the line passes through (0,0), the y-intercept is 0.
 Equation for Segment 1: P(t)=50t for 0≤t≤0.1

Segment 2: From t=0.1 s to t=0.35 s

 Points: (0.1,5) and (0.35,0)

 Slope (m2): m2=t2−t1P2−P1=0.35−0.10−5=0.25−5=−20
 Equation using point-slope form (using point (0.35, 0)): P(t)−0=−20(t−0.35) P(t)=−20t+(−20)
(−0.35) P(t)=−20t+7
 Equation for Segment 2: P(t)=−20t+7 for 0.1<t≤0.35

Segment 3: From t=0.35 s to t=0.5 s

 Points: (0.35,0) and (0.5,−1)

 Slope (m3): m3=t3−t2P3−P2=0.5−0.35−1−0=0.15−1=−3/201=−320 (approximately -6.67)
 Equation using point-slope form (using point (0.35, 0)): P(t)−0=−320(t−0.35) P(t)=−320t−320
(−0.35) P(t)=−320t+320×0.35 P(t)=−320t+37
 Equation for Segment 3: P(t)=−320t+37 for 0.35<t≤0.5

Complete Piecewise Equation for the Graph:

Combining these segments, the equation for the pressure P(t) in kPa as a function of time t in
seconds is:

P(t)=⎩⎨⎧50t −20t+7 −320t+370

for 0≤t≤0.1for 0.1<t≤0.35for 0.35<t≤0.5otherwise(assuming P(t) is zero outside this range)

Okay, let's focus solely on providing the equation for the graph, interpreted as a piecewise linear
function.

The graph displays pressure (kPa) on the y-axis and time (s) on the x-axis.

We need to identify the coordinates of the "break points" or "nodes" where the slope of the graph
changes.
 How to find p from graph with linear equation solution

 Point 1 (Start): (t0,P0)=(0,0)

 Point 2 (Peak): (t1,P1)=(0.1,5)
 Point 3 (Zero crossing): (t2,P2)=(0.35,0)
 Point 4 (End): (t3,P3)=(0.5,−1)

Now, we will determine the equation for each linear segment. A linear equation has the form
P(t)=mt+b, where m is the slope and b is the y-intercept.

Segment 1: From t=0 s to t=0.1 s

 Points: (0,0) and (0.1,5)

 Slope (m1): m1=t1−t0P1−P0=0.1−05−0=0.15=50
 Y-intercept (b1): Since it passes through (0,0), b1=0.
 Equation for Segment 1: P(t)=50t

Segment 2: From t=0.1 s to t=0.35 s

 Points: (0.1,5) and (0.35,0)

 Slope (m2): m2=t2−t1P2−P1=0.35−0.10−5=0.25−5=−20
 Using point-slope form (P−P1=m(t−t1)) with point (0.35, 0): P−0=−20(t−0.35) P(t)=−20t+
(−20×−0.35) P(t)=−20t+7
 Alternatively, using point (0.1, 5): P−5=−20(t−0.1) P−5=−20t+2 P(t)=−20t+7

Segment 3: From t=0.35 s to t=0.5 s

 Points: (0.35,0) and (0.5,−1)

 Slope (m3): m3=t3−t2P3−P2=0.5−0.35−1−0=0.15−1=−3/201=−320
 Using point-slope form (P−P2=m(t−t2)) with point (0.35, 0): P−0=−320(t−0.35) P(t)=−320t+
(−320×−0.35) P(t)=−320t+(−320×−207) P(t)=−320t+37

The complete piecewise equation for the pressure P(t) in kPa is:

P(t)=⎩⎨⎧50t−20t+7−320t+370for 0≤t≤0.1for 0.1<t≤0.35for 0.35<t≤0.5for t<0 or t>0.5 (assumin

g P=0 outside this range)
 Problem 1: Two-storey Building Natural Frequencies and Mode Shapes

(From: past papers.pdf, Page 1, Q3)

Problem Statement:

A two-storey building with very stiff floor slabs relative to the supporting columns.

 Top mass (m2) = 3 tonne

 Bottom mass (m1) = 5 tonne
 Both stories have a height (h) = 3 m.
 The width of the frame is 10 m.
 Columns have stiffness (EIc) = 4.5×103kNm2.
 Task: Calculate the natural frequencies and mode shapes.

Procedure to Solve:

1. Determine the Stiffness Matrix ([K]):

o For a two-story shear building, the lateral stiffness of each story is contributed by its columns.
Assuming fixed-fixed boundary conditions for columns at floor levels due to rigid beams, the
stiffness of a single column is kcolumn=h312EIc.
o Since there are two columns per story, the stiffness of the first story (k1) and the second story
(k2) are: k1=k2=2×h312EIc
o The stiffness matrix for a two-story shear building is: [K]=[(k1+k2)−k2−k2k2]
2. Determine the Mass Matrix ([M]):
o Convert masses from tonnes to kg (1tonne=1000kg).
o The mass matrix for a lumped mass system is diagonal: [M]=[m100m2]
3. Formulate the Eigenvalue Problem:
o The equation for free vibration is [M]u¨+[K]u=0.
o Assuming harmonic motion u=ϕsin(ωt), substitute into the equation to get the eigenvalue
problem: ([K]−ω2[M])ϕ=0.
o To find the natural frequencies (ωn), solve the characteristic equation: det([K]−ω2[M])=0. This
will result in a polynomial equation in ω2.
4. Solve for Natural Frequencies (ωn):
o Solve the polynomial equation for ω12 and ω22.
o Take the square root to get ω1 and ω2 in rad/sec.
5. Solve for Mode Shapes (ϕn):
o For each natural frequency ωn, substitute it back into the equation ([K]−ωn2[M])ϕn=0.
o Solve for the corresponding eigenvector ϕn={ϕ1n,ϕ2n}T. You can set one component of the
eigenvector to a reference value (e.g., ϕ1n=1) to find the relative shape of the mode.

Problem 2: Response of a Building to Explosion

(From: image_33a935.jpg, Q1)

 Problem Statement:

A four-story building subjected to an explosion. The air pressure wave is a triangular pulse from
P0 at t=0 to 0 at t=0.1 sec.

 Damping ratio (ξ) = 2% of critical.

 Girders: width (b) = 0.40 m, depth (h) = 0.50 m.
 Columns: square section, side dimension (H) = 0.40 m.
 Modulus of elasticity (E) = 25 GPa.
 Building mass per unit area = 1000kg/m2.
 Pressure is uniform and applied to the façade.
 Tributary area for top story = 10m×1.5m=15m2.
 Tributary area for other floors = 10m×3m=30m2 each.
 Given Matrices:
o Stiffness Matrix [Ks]=10476.869−99.69125.583−3.4747−99.691202.4−119.0223.108
25.583−119.02273.11−142.11−3.474723.108−142.11252.30kNm
o Mass Matrix [M]=60000060000060000060tonne
 Given Natural Frequencies (ωn) and Mode Shapes (Φn):
o ω1=8.09274rad/s, Φ1={0.074828,0.050226,0.023604,0.010254}T
o ω2=20.2500rad/s, Φ2={0.081804,−0.081692,−0.060934,−0.030584}T
o ω3=32.8937rad/s, Φ3={0.026884,0.064504,0.096223,0.069665}T
o ω4=37.9126rad/s, Φ4={0.021268,0.061745,0.086883,0.069665}T
 Task: Obtain the response of the structure in the short direction.

Procedure to Solve (using Modal Superposition Method):

1. Define the Applied Force Vector (F(t)):

o The pressure P(t) is a triangular pulse: P(t)=P0(1−t/td) for 0≤t≤td, and P(t)=0 for t>td, where td
=0.1sec. The peak pressure P0 is not given, so the response will be in terms of P0.
o The force applied to each floor is Fi(t)=P(t)×Areai.
o F(t)=P(t)×{30,30,30,15}T[m2] (assuming these are floor 1, 2, 3, 4 from bottom to top).
2. Calculate Modal Properties:
o Modal Mass (Mn): Mn=ϕnT[M]ϕn. Remember to convert mass from tonnes to kg for
calculations.
o Modal Stiffness (Kn): Kn=ϕnT[K]ϕn=ωn2Mn. Remember to convert stiffness from kNm to
N/m.
o Modal Damping (Cn): Cn=2ξωnMn, where ξ=0.02.
o Modal Load Factor (Ln(t)): Ln(t)=ϕnTF(t).
o Generalized Force (Pn(t)): Pn(t)=Ln(t).
3. Solve the Uncoupled Single-Degree-of-Freedom Equations for Generalized Coordinates
(Dn(t)):
o For each mode n, solve the uncoupled equation: MnD¨n(t)+CnD˙n(t)+KnDn(t)=Pn(t)
o This is a standard ordinary differential equation for a damped SDOF system subjected to a
triangular pulse load. The solution involves finding the homogeneous and particular solutions
based on the time function of Pn(t). For a triangular pulse, the response can be found using
 convolution integral or by solving piecewise. Assume zero initial conditions for Dn(0) and D˙n
(0) unless specified.
4. Combine Generalized Coordinates to Find Floor Displacements (u(t)):
o The absolute displacement vector at each floor is given by the superposition of modal
contributions: u(t)=∑n=1NϕnDn(t), where N is the number of modes (here, 4).

Problem 3: Single-bay, Single-storey Frame Analysis

(From: past papers.pdf, Page 1, Q1a-e)

Problem Statement:

A 3 m high, 8 m wide single-bay single-storey frame is rigidly jointed.

 Beam mass (m) = 5,000 kg.

 Columns have negligible mass.
 Column stiffness (EIc) = 4.5×103kNm2.

Part (a): Calculate the natural frequency in lateral vibration and its period.

Procedure for Part (a):

1. Determine the Stiffness (k) of the Frame:

o For a single-story frame with fixed-base columns and a rigid beam (lumped mass at the beam),
the lateral stiffness for two columns is k=h32×12EIc.
2. Calculate Natural Frequency (ωn):
o ωn=mk
3. Calculate Natural Period (Tn):
o Tn=ωn2π

Part (b): Find the force required to deflect the frame 25 mm laterally.

Procedure for Part (b):

1. Use Hooke's Law: F=k×Δ.

2. Use the stiffness (k) calculated in Part (a) and the given deflection Δ=25mm=0.025m.

Parts (c) and (d): Damping and Dynamic Properties

 Information from a test: A jack applied a load of 100 kN and then instantaneously released. On
the first return swing, a deflection of 19.44 mm was noted. The period of motion was measured
at 0.223 sec.
 Assumption: Stiffness of the columns cannot change (use k from Part a).
 Task: Find (a) the effective weight of the beam; (b) the damping ratio; (c) the coefficient of
damping; (d) the undamped frequency and period; and (e) the amplitude after 5 cycles.
 Procedure for Parts (c) and (d):

1. Interpret Test Data and Calculate Damping Ratio (ξ):

o The phrase "first return swing a deflection of 19.44 mm was noted" after release from a load of
100 kN. This is often interpreted as the initial displacement u0=100kN/k (using k from part a),
and the amplitude of the first peak in the opposite direction (first return swing) is u1/2
=19.44mm.
o The ratio of successive amplitudes in a damped free vibration is ui/ui+1=eδ, where δ is the
logarithmic decrement. The ratio of amplitude to first return swing is u0/u1/2=eδ/2.
o So, calculate δ=2×ln(u0/19.44mm).
o Then, calculate the damping ratio: ξ=(2π)2+δ2δ.
o Alternatively, if 25mm (from Part b) and 19.44mm are assumed to be consecutive peaks in a free
vibration test, then δ=ln(25/19.44). This interpretation is often used when the initial displacement
for the decay is not explicitly given in the damping section. I will assume this interpretation for
calculating ξ.
2. Calculate Undamped Period (Tn) and Undamped Frequency (ωn) (Q1d.d):
o Given damped period TD=0.223sec.
o Use the relationship: TD=Tn/1−ξ2 to find Tn.
o Then ωn=2π/Tn.
3. Calculate Effective Weight of the Beam (Q1d.a):
o Using the calculated ωn and the stiffness k from Part (a) (since stiffness is assumed constant):
m=k/ωn2.
o Effective weight = m×g (where g≈9.81m/s2).
4. Calculate Coefficient of Damping (c) (Q1d.c):
o c=2ξωnm (using the calculated ξ, ωn, and m).
5. Calculate Amplitude after 5 Cycles (Q1d.e):
o If u0 is the initial amplitude (e.g., you can use 25 mm if you assume that as the initial
displacement), the amplitude after 5 cycles (u5) for damped free vibration is: u5=u0e−5δ or u5
=u0e−5×2πξ/1−ξ2.

Problem 4: Three-story Frame Analysis

(From: past papers.pdf, Page 2, Q6)

Problem Statement:

A three-story frame with lumped masses subjected to lateral forces. Flexural rigidity is EI for
columns and beams.

 Heights of stories: h, h, h (from bottom to top).

 Width of bay: 2h.
 Masses: m at first floor, m at second floor, m/2 at third floor.
 Task (a): Identify the DOFs for representing the elastic properties and determine the stiffness
matrix. Neglect axial deformation of the members.
 Task (b): Identify the DOFs for representing the inertial properties and determine the mass
matrix. Assume members are massless and neglect their inertia.
 Task (c): Formulate the equations governing the motion of the frame in the DOFs in part (b).

Procedure to Solve:

Part (a): DOFs for Elastic Properties and Stiffness Matrix ([K])

1. Identify Degrees of Freedom (DOFs):

o For a shear building model with lumped masses, the primary DOFs are the lateral displacements
of each floor. Let u1,u2,u3 be the absolute lateral displacements of the first, second, and third
floors, respectively.
2. Determine Story Stiffnesses:
o Assuming a shear building idealization where floor slabs are rigid and columns provide lateral
stiffness:
o For a single column with both ends fixed against rotation (due to rigid beam connections and
fixed base), its lateral stiffness is kcolumn=h312EI.
o Since there are two columns per story, the stiffness of each story (k1,k2,k3) is: k1=k2=k3
=2×h312EI=h324EI.
3. Formulate the Stiffness Matrix ([K]):
o For a multi-story shear building, the stiffness matrix is typically tridiagonal, relating the forces at
each floor to the displacements: [K]=(k1+k2)−k20−k2(k2+k3)−k30−k3k3
o Substitute the expressions for k1,k2,k3 in terms of EI and h.

Part (b): DOFs for Inertial Properties and Mass Matrix ([M])

1. Identify DOFs for Inertial Properties:

o These are the same as the elastic DOFs: u1,u2,u3.
2. Determine the Mass Matrix ([M]):
o Given lumped masses: m1=m, m2=m, m3=m/2.
o The mass matrix for a lumped mass system is a diagonal matrix: [M]=m1000m2000m3=m00
0m000m/2

Part (c): Formulate Equations Governing the Motion

1. General Equation of Motion:

o For an undamped multi-degree-of-freedom system subjected to lateral forces P(t)={p1(t),p2(t),p3
(t)}T (as shown in the figure, applied at each floor): [M]u¨+[K]u=P(t)
o Substitute the derived mass matrix [M] and stiffness matrix [K] into this equation.

Problem 5: Diagonal Remedial Ties to Reduce Natural Period

(From: past papers.pdf, Page 1, Q2)

Problem Statement: Workers' movements on a platform (8×6 m high, m=200kN) are causing
large dynamic motions. An engineer investigated and found the natural period in sway to be 0.9
sec. Diagonal remedial ties (E=200kN/mm2) are installed.

 Task (a): Determine the minimum cross-sectional area of each tie to reduce the natural
period to 0.4 sec.
 Task (b): If the ties have a cross-sectional area of 100 mm$^2$, what is the new natural
period?
 Task (c): What is the damping ratio for the original platform if the amplitude ratio in
successive cycles is 0.8?

Procedure to Solve:

This problem involves understanding the relationship between stiffness, mass, natural period,
and damping.

Initial State Analysis:

1. Original System Mass: m=200kN (This is weight, convert to mass: M=g200kN, where
g≈9.81m/s2).
2. Original Natural Period (Tn,orig): 0.9sec.
3. Original Stiffness (korig): Use the formula Tn,orig=2πkorigM. Rearrange to find korig.

Part (a): Minimum Cross-sectional Area for New Period

1. Target Natural Period (Tn,new): 0.4sec.

2. Calculate Required New Stiffness (knew):
o Assume the mass of the platform remains the same (ties add negligible mass).
o Use Tn,new=2πknewM. Rearrange to find knew.
3. Calculate Stiffness Added by Ties (kties):
o kties=knew−korig.
4. Determine Stiffness of Each Tie:
o A diagonal tie acts as an axial member. The axial stiffness of a single tie is ktie
=LAE, where A is the cross-sectional area, E is the modulus of elasticity, and L is
the length of the tie.
o The problem implies two diagonal ties (one for tension, one for compression, or
acting together as a bracing system). Assuming they work in parallel for lateral
stiffness (e.g., in a typical X-bracing system), their combined lateral stiffness
contribution needs to be considered. For a single diagonal bracing member
contributing to lateral stiffness of a story, its effective lateral stiffness is kdiag
=LAEcos2θ, where θ is the angle the diagonal makes with the horizontal. You'll
need to infer the geometry (length L and angle θ) from the platform dimensions
(8×6 m high) and the tie configuration. The problem states "8×6 m high",
suggesting 8m is width and 6m is height. So, L=82+62=10m and cosθ=8/10=0.8.
o If there are two ties acting together, then kties=2×kdiag.
 o Solve for A from the calculated kties. Remember to be consistent with units (kN
and mm).

Part (b): New Natural Period with Given Tie Area

1. Calculate Stiffness of Given Ties: Use A=100mm2 in the formula for kdiag (or 2×kdiag
if two ties).
2. Calculate New Total Stiffness (knew): knew=korig+kties.
3. Calculate New Natural Period (Tn,new): Use Tn,new=2πknewM.

Part (c): Damping Ratio for Original Platform

1. Logarithmic Decrement: Given amplitude ratio in successive cycles = 0.8. This means
ui+1/ui=0.8, so ui/ui+1=1/0.8=1.25.
2. Calculate Logarithmic Decrement (δ): δ=ln(ui/ui+1)=ln(1.25).
3. Calculate Damping Ratio (ξ): ξ=(2π)2+δ2δ.

Problem 6: Construct Design Spectrum

(From: past papers.pdf, Page 1, Ex 7.1 - "EX 7.1 Construct a design spectrum for 5%
damping for a site characterized by the following response characteristics: Peak ground
acceleration=0.25 g ...")

Problem Statement: Construct a design spectrum for 5% damping for a site characterized by the
following response characteristics:

 Peak ground acceleration (amax) = 0.25 g

 Peak ground velocity (vmax) = 12 in/sec
 Peak ground displacement (dmax) = 10 in
 Task: Construct the design spectrum.

Procedure to Solve:

This problem requires constructing a typical smooth design spectrum based on key ground
motion parameters. The spectrum is generally divided into several regions based on period,
corresponding to constant acceleration, constant velocity, and constant displacement. For 5%
damping, standard amplification factors are often used.

1. Identify Ground Motion Parameters:

o amax=0.25g (convert g to appropriate units, e.g., in/sec$^2$)
o vmax=12in/sec
o dmax=10in
o Damping ratio ξ=5%=0.05.
2. Determine Characteristic Periods:
 o Period Ta (Transition from acceleration-controlled to velocity-controlled):Ta
=amax2πvmax (Ensure consistent units).
o Period Tb (Transition from velocity-controlled to displacement-
controlled):Tb=vmax2πdmax (Ensure consistent units).
3. Calculate Spectral Amplification Factors (for 5% damping):
o For Sa/amax: Typically between 2.0 and 2.5 (e.g., ≈2.5 for many codes, for
acceleration-controlled region).
o For Sv/vmax: Typically between 1.5 and 2.0 (e.g., ≈1.5 for velocity-controlled
region).
o For Sd/dmax: Typically between 1.0 and 1.2 (e.g., ≈1.0 for displacement-
controlled region).
o Note: These factors can vary based on specific building codes or typical design
spectrum shapes. If specific factors are implied by the course material, use those.
A common set of values might be around 2.5,1.7,1.0 for Sa/amax,Sv/vmax
,Sd/dmax respectively for 5% damping.
4. Define Regions of the Design Spectrum:
o Region 1: Short Periods (T<Ta) - Constant Spectral Acceleration (Sa):
 Sa=(Amplification Factor for Sa/amax)×amax
 This region starts from Sa≈amax at T=0 and rises to the constant spectral
acceleration. Some models have a linear increase up to a certain period T0
and then constant Sa. The problem does not specify T0.
o Region 2: Intermediate Periods (Ta≤T≤Tb) - Constant Spectral Velocity (Sv
):
 Sv=(Amplification Factor for Sv/vmax)×vmax
 Alternatively, using the relationship Sv=ωSd=T2πSd and Sa=ωSv=(T2π
)2Sd.
 Sa=T2πSv. In this region, Sv is constant, so Sa decreases with increasing
T.
o Region 3: Long Periods (T>Tb) - Constant Spectral Displacement (Sd):
 Sd=(Amplification Factor for Sd/dmax)×dmax
 In this region, Sd is constant, so Sv decreases with increasing T (Sv=T2π
Sd), and Sa decreases with T2 (Sa=(T2π)2Sd).
5. Plot the Design Spectrum:
o Plot Spectral Acceleration (Sa) vs. Period (T).
o Plot Spectral Velocity (Sv) vs. Period (T).
o Plot Spectral Displacement (Sd) vs. Period (T).
o Typically, Sa is the main plot, with Sv and Sd derivable from it.

Problem 7: Static Condensation

(From: past papers.pdf, Page 4 - "For the structure shown below determine the natural
shapes and frequencies after static condensation coordinates and..." and Page 6 - "Figure
below shows a uniform four - storied shear building. For this structure, determine the
natural frequencies and mode shapes after static condensation.")
Problem Statement (Generalizing from multiple mentions): Given a multi-degree-of-freedom
system (e.g., a multi-story shear building), determine its natural frequencies and mode shapes
using the static condensation method. This method involves reducing the number of degrees of
freedom by eliminating "slave" DOFs, often those without significant mass.

Procedure to Solve:

Static condensation is a technique to reduce the size of the eigenvalue problem for dynamic
analysis by partitioning the DOFs into "master" (primary) and "slave" (secondary) DOFs.

1. Partition the Stiffness ([K]) and Mass ([M]) Matrices:

o Identify "master" DOFs (um) and "slave" DOFs (us). Master DOFs are typically
associated with significant masses or where direct dynamic response is desired.
Slave DOFs are usually massless or where their inertia effects are negligible.
o Partition the matrices as follows:[K]=[KmmKsmKmsKss][M]=[MmmMsmMms
Mss]
o In many static condensation problems, slave DOFs are assumed massless, so Mss
might be a zero matrix.
2. Derive the Condensed Stiffness Matrix (Kc):
o From static equilibrium of the slave DOFs (assuming zero inertia forces for
slaves):Ksmum+Kssus=0⟹us=−Kss−1Ksmum
o Substitute this into the force equilibrium for master DOFs to get the condensed
stiffness matrix:Kc=Kmm−KmsKss−1Ksm
3. Derive the Condensed Mass Matrix (Mc):
o The kinetic energy of the system is 21u˙T[M]u˙.
o Substitute us in terms of um and derive the condensed mass matrix in terms of
master DOFs. If Mss is zero, it simplifies significantly.
o If only Mmm is non-zero (i.e., slave DOFs are massless), then the condensed
mass matrix remains Mmm. However, if Mms or Msm are non-zero (due to
coupling between master and slave masses), then the derivation is more involved.
For simpler cases, it's often Mc≈Mmm. (A more rigorous derivation considering
transformation matrix: Mc=(TTMT), where T=[I−Kss−1Ksm])
4. Solve the Condensed Eigenvalue Problem:
o Formulate the eigenvalue problem for the condensed system:(Kc−ω2Mc)ϕm=0
o Solve for the natural frequencies (ωn) and master mode shapes (ϕm). This is a
smaller eigenvalue problem than the original one.
5. Expand Mode Shapes:
o Once ϕm (mode shapes of master DOFs) are found, expand them to include the
slave DOFs using the relationship derived in step 2:ϕs=−Kss−1Ksmϕm
o The full mode shape for each mode n is ϕn=[ϕmnϕsn].

Problem 8: Response Spectrum Analysis

(From: past papers.pdf, Page 5, Q1 - "A structure modeled as shown below is assumed to
be subjected to a support motion produced by El Centro earthquake of 1940. Assuming
elastic response and the response spectrum given below find the relative displacement
between the mass and the support. Also compute the maximum force acting on the spring.
Neglect damping.")

Problem Statement: A structure modeled as a single-degree-of-freedom (SDOF) system is

subjected to support motion from the El Centro earthquake of 1940.

 Assume elastic response.

 Neglect damping.
 A response spectrum is given (graph showing Spectral
Displacement/Velocity/Acceleration vs. Period/Frequency).
 Task: Find the relative displacement between the mass and the support. Compute the
maximum force acting on the spring.

Procedure to Solve:

This problem uses a pre-computed response spectrum to find the maximum response of a
structure without explicitly performing a time-history analysis.

1. Determine System Properties:

o From the SDOF model (mass m and spring stiffness k), calculate the natural
frequency (ωn) and natural period (Tn):ωn=mkTn=ωn2π
o The problem states "neglect damping", but response spectra are usually provided
for specific damping ratios (e.g., 5%). If the spectrum itself implies zero damping,
that's fine. If not, it means the damping effects are ignored by assuming ξ=0 or
using the zero-damping curve from the spectrum.
2. Read Response from the Given Spectrum:
o Using the calculated natural period (Tn) of your SDOF system, locate this period
on the horizontal axis of the provided response spectrum.
o Read the corresponding value from the vertical axis.
o If the task asks for relative displacement, read Spectral Displacement (Sd) directly
from the spectrum for your Tn.
o If the task asks for force, you might need Spectral Acceleration (Sa). You can
either read Sa directly (if available) or calculate it from Sd: Sa=ωn2Sd=(Tn2π
)2Sd.
3. Calculate Maximum Relative Displacement:
o The maximum relative displacement between the mass and the support is directly
given by the Spectral Displacement (Sd) value read from the spectrum for the
system's natural period.umax=Sd
4. Compute Maximum Force Acting on the Spring:
o The maximum force in the spring is Fmax=k×umax.
o Alternatively, using spectral acceleration, Fmax=m×Sa. Both methods should
yield the same result.
Problem 9: Fourier Series Expression for Periodic Excitation

(From: past papers.pdf, Page 7 or 9, Q6 - "Find the Fourier series expression for the
periodic excitation made of half-sine waves.")

Problem Statement: Find the Fourier series expression for the periodic excitation made of half-
sine waves. (The figure would show a periodic function resembling rectified sine waves).

Procedure to Solve:

This is a mathematical problem to represent a periodic non-sinusoidal function as a sum of sines

and cosines. For a periodic function f(t) with period T=2L, its Fourier series is:f(t)=2a0+∑n=1∞
(ancos(Lnπt)+bnsin(Lnπt))

where the coefficients are:a0=L1∫−LLf(t)dtan=L1∫−LLf(t)cos(Lnπt)dtbn=L1∫−LLf(t)sin(Lnπt)dt

For a function composed of half-sine waves, you need to define f(t) over one period and then
evaluate these integrals.

1. Define the Function (f(t)) over One Period:

o A "half-sine wave" typically refers to either a half-wave rectified sine wave (only
positive lobes) or a full-wave rectified sine wave (all lobes positive).
o Assume a peak amplitude F0 and a period Twave.
o For example, if it's a full-wave rectified sine wave:f(t)=F0∣sin(Twave2πt)∣ The
period of this function is L=Twave/2.
o If it's a half-wave rectified sine wave:f(t)=F0sin(Twave2πt) for
0≤t≤Twave/2f(t)=0 for Twave/2<t<Twave The period is T=Twave, so
L=Twave/2.
2. Calculate Fourier Coefficients (a0,an,bn):
o a0 (DC component): Integrate f(t) over one period and divide by L.
o an (Cosine coefficients): Integrate f(t)cos(Lnπt) over one period and divide by L.
o bn (Sine coefficients): Integrate f(t)sin(Lnπt) over one period and divide by L.
o Symmetry considerations: If the function is even, all bn will be zero. If it's odd,
all an will be zero. If it has half-wave symmetry, specific coefficients might be
zero. Half-sine waves often have certain symmetries that simplify the
calculations.
3. Write the Fourier Series Expression:
o Substitute the calculated coefficients back into the Fourier series formula.

This will provide the mathematical expression for the periodic excitation, which can then be used
in dynamic analysis for forced vibration.
 Here are procedures for more problems from your "past papers.pdf" file, continuing from where
we left off. I'm selecting problems that involve different dynamic analysis concepts.

Problem 10: Forced Vibration Response to Constant Force

(From: past papers.pdf, Page 6 - "Determine the response as a function of time for the two-
story frame as shown below. when a constant force of 5000 lbs is applied.")

Problem Statement:

A two-story frame is subjected to a constant force of 5000 lbs.

 The image shows a two-story frame with lumped masses (m1, m2) and story stiffnesses (k1, k2).
Assume these properties are given or can be derived from the frame geometry (similar to
Problem 1 or 4 if the image was clear).
 Task: Determine the response as a function of time.

Procedure to Solve:

This problem requires determining the time-history response of a multi-degree-of-freedom

system to a suddenly applied constant force. Modal superposition method is generally used.

1. Determine System Properties:

o Mass Matrix ([M]): Based on the lumped masses m1,m2.
o Stiffness Matrix ([K]): Based on the story stiffnesses k1,k2.
o Natural Frequencies (ωn) and Mode Shapes (ϕn): Solve the eigenvalue problem ([K]
−ω2[M])ϕ=0 to obtain these. (Refer to Problem 1 or 4 for detailed steps).
2. Define Applied Force:
o A constant force F0=5000lbs is applied. The problem statement does not specify if it's applied to
a specific floor. Assume it's applied as a step function. If applied to a single floor (e.g., F2
(t)=5000lbs and F1(t)=0), then the force vector P(t)={0,5000}T for t≥0. If applied to multiple
floors, the vector will change accordingly.
3. Calculate Modal Properties (if using Modal Superposition):
o Modal Mass (Mn): Mn=ϕnT[M]ϕn.
o Modal Stiffness (Kn): Kn=ϕnT[K]ϕn=ωn2Mn.
o Modal Participation Factor (Ln): Ln=ϕnTP(t). For a constant force, Ln will also be constant.
o Generalized Force (Pn): Pn=Ln.
4. Solve Uncoupled SDOF Equations for Generalized Coordinates (Dn(t)):
o The uncoupled equation for each mode is: MnD¨n(t)+KnDn(t)=Pn. (Assuming no damping is
mentioned).
o For a constant force applied as a step function (and zero initial conditions for Dn(0),D˙n(0)): The
solution is typically of the form: Dn(t)=KnPn(1−cos(ωnt)). (This represents the dynamic
response oscillating around the static displacement).
5. Combine Generalized Coordinates to Find Floor Displacements (u(t)):
o The absolute displacement vector at each floor as a function of time is: u(t)=∑n=1NϕnDn(t)
o This expression gives the time-history of displacements for each floor.

Problem 11: Forced Vibration Response to Triangular Impulsive Force

(From: past papers.pdf, Page 6 - "A two storied building as shown below was subjected to
triangular implusive forces also shown below. Determine the maximum displacement of the
follors and shear forces in the colums. Given that ω1=11.8 rad/sec, ω2=32.9 rad/sec, , ϕ21=
0.08130, ϕ12=0.0567, ϕ22=0.0924.")

Problem Statement:

A two-story building subjected to triangular impulsive forces.

 Given: Natural frequencies (ω1,ω2) and some mode shape components (ϕ21,ϕ12,ϕ22). The full
mode shapes would be needed. Assuming ϕ11 is the reference for ϕ1.
o ω1=11.8rad/sec
o ω2=32.9rad/sec
o ϕ1={ϕ11,ϕ21}T (e.g., set ϕ11=1, then ϕ1={1,0.08130}T)
o ϕ2={ϕ12,ϕ22}T (e.g., set ϕ12=1, then ϕ2={1,0.0924}T). Note: The provided ϕ12=0.0567 and
ϕ22=0.0924 seem to be for the second mode where the first component might be normalized to 1,
or these are actual values with ϕ11 and ϕ21 being the other components of the first mode.
Clarification on the full mode shape vectors is needed. I will assume they are normalized with
 ϕ11=1 and ϕ12=1 or given in the format as in Problem 2 earlier, i.e., column vectors for modes.
Let's assume the provided ϕ21 and ϕ12,ϕ22 values are specific entries within the mode shape
matrix, not complete normalized vectors.
o Clarification on mode shapes: A typical convention for a 2-story building might be ϕ1={ϕ11,ϕ21
}T and ϕ2={ϕ12,ϕ22}T. If ϕ11 and ϕ12 are normalized to 1, then ϕ1={1,0.08130}T and ϕ2
={1,0.0924}T. The value ϕ12=0.0567 would be a typo or mean something else. Assuming ϕ1
={1,ϕ21}T and ϕ2={ϕ12′,ϕ22}T where ϕ12′ is the first component of the second mode. Given
the ambiguity, I'll proceed with the methodology assuming full mode shapes are known.
 The image shows triangular impulsive forces applied to the floors (e.g., F1(t) and F2(t)).
 Task: Determine the maximum displacement of the floors and shear forces in the columns.

Procedure to Solve:

This is a forced vibration response problem for multi-DOF system subjected to transient
(impulsive) loads using modal superposition.

1. Define Applied Force History (P(t)):

o For each floor, define the triangular impulsive force as a function of time. A triangular pulse
from time t=0 to t=td.
2. Determine System Properties (if not given completely):
o Mass Matrix ([M]) and Stiffness Matrix ([K]): If not given explicitly, these would need to be
derived from the building geometry (similar to Problem 1 or 4).
o Natural Frequencies (ωn) and Mode Shapes (ϕn): Given in the problem. Ensure you have the
full mode shape vectors. If only parts are given, you'd usually complete them by normalization
(e.g., setting the roof displacement to 1). Assume the given values are components of normalized
modes.
3. Calculate Modal Properties:
o Modal Mass (Mn): Mn=ϕnT[M]ϕn.
o Modal Stiffness (Kn): Kn=ϕnT[K]ϕn=ωn2Mn.
o Modal Participation Factor (Ln(t)): Ln(t)=ϕnTP(t).
o Generalized Force (Pn(t)): Pn(t)=Ln(t).
4. Solve Uncoupled SDOF Equations for Generalized Coordinates (Dn(t)):
o The uncoupled equation for each mode is: MnD¨n(t)+KnDn(t)=Pn(t). (Assuming no damping is
mentioned).
o For a triangular impulsive force, the solution Dn(t) needs to be obtained by solving the
differential equation piecewise (during the rise/fall of the force and after the force becomes zero)
or by using the Duhamel Integral (convolution integral).
 Solution will depend on whether t<td or t≥td.
5. Combine Generalized Coordinates to Find Floor Displacements (u(t)):
o u(t)=∑n=1NϕnDn(t). This gives the time-history of displacements.
6. Determine Maximum Displacements:
o After finding u(t), analyze the time history to find the maximum displacement for each floor.
This usually involves finding the peak value of the ui(t) function.
7. Calculate Story Shear Forces:
o Story shear forces (Vi(t)) can be calculated from the inter-story drifts and story stiffnesses: Vi
(t)=ki(ui(t)−ui−1(t)) (where u0(t)=0).
o Alternatively, sum the inertial forces above the story.
o Determine the maximum shear force for each column from the time history.

Problem 12: Steady-State Response to Harmonic Force with Damping

(From: past papers.pdf, Page 6, Q7 - "Determine the steady - state response of the two-
story shear buliding as given in Q5, when F2(t)=10,000 sin 20t. This building is damping
too as shown below. Assume for this building that the damping constants c1=307 lb.sec/in
and c2=444 lb.sec/in.")

Problem Statement:

Determine the steady-state response of a two-story shear building (as given in Q5, which means
masses m1,m2 and stiffnesses k1,k2).

 Applied Force: F2(t)=10,000sin(20t) (applied to the second floor).

 Damping Constants: c1=307lb.sec/in, c2=444lb.sec/in.
 Task: Determine the steady-state response.

Procedure to Solve:

This problem involves finding the steady-state vibration of a damped multi-DOF system
subjected to harmonic excitation.

1. Determine System Properties:

o Mass Matrix ([M]): From Q5 (typically diagonal, m1,m2).
o Stiffness Matrix ([K]): From Q5 (typically tridiagonal for shear building).
o Damping Matrix ([C]):
 For a shear building with viscous dampers, the damping matrix for a two-story system is:
[C]=[(c1+c2)−c2−c2c2] (similar structure to K-matrix).
o Applied Force Vector (P(t)): Since force is only on the second floor: P(t)={0,10000sin(20t)}T.
The excitation frequency is Ω=20rad/sec.
2. Formulate Equations of Motion:
o [M]u¨+[C]u˙+[K]u=P(t)
3. Solve for Steady-State Response (Direct Method in Frequency Domain):
o For harmonic excitation P(t)=P0sin(Ωt) (where P0={0,10000}T), the steady-state response is
also harmonic: u(t)=u0sin(Ωt−α) or u(t)=Re(UeiΩt), where U is the complex amplitude vector.
o The complex amplitude U can be found from: ([K]−Ω2[M]+iΩ[C])U=P0 (where P0 here is the
peak force vector, e.g., P0={0,10000}T).
o Solve for U={U1,U2}T, where Uj=∣Uj∣eiθj.
o Then, uj(t)=∣Uj∣sin(Ωt+θj).
4. Alternatively, using Modal Analysis (for systems with proportional damping):
o First, find natural frequencies and mode shapes (ωn,ϕn) (assuming undamped modes are
sufficient, or if damping is explicitly proportional).
o Calculate modal properties (Mn,Kn,Cn).
o Transform forces to modal coordinates: Pn(t)=ϕnTP(t).
o Solve the uncoupled SDOF equations for steady-state response of generalized coordinates Dn(t).
For each mode: MnD¨n(t)+CnD˙n(t)+KnDn(t)=Pn(t) The steady-state solution for Dn(t) for a
harmonic Pn(t)=(Pn)0sin(Ωt) is: Dn(t)=(Dn)0sin(Ωt−αn) where (Dn)0=Kn(1−(Ω/ωn)2)2+(2ξn
Ω/ωn)2(Pn)0 and tanαn=1−(Ω/ωn)22ξnΩ/ωn.
o Recombine generalized coordinates to find floor displacements u(t)=∑ϕnDn(t).

Problem 13: Forced Vibration Due to Ground Motion

(From: past papers.pdf, Page 8 - "The tower shown below is subjected to a horizontal
ground acceleration a(t) as shown below also. Determine the relative displacement at the
top of the tower at t=0.1 sec. Neglect damping.")

Problem Statement:

A tower (modeled as a SDOF system with mass m and stiffness k) is subjected to horizontal
ground acceleration a(t).

 The ground acceleration a(t) is shown as a triangular pulse (e.g., peak value at t=0, linearly
decreasing to 0 at t=td).
 Task: Determine the relative displacement at the top of the tower at t=0.1 sec. Neglect damping.

Procedure to Solve:

This problem involves finding the dynamic response of a SDOF system to base excitation
(ground motion).

1. Determine System Properties:

o Mass (m) and Stiffness (k) of the tower: From the SDOF model.
o Natural Frequency (ωn) and Period (Tn): ωn=k/m, Tn=2π/ωn.
o Damping is neglected (ξ=0).
2. Define Effective Earthquake Force:
o The equation of motion for relative displacement u(t) (displacement of mass relative to ground)
due to ground acceleration u¨g(t) is: mu¨+ku=−mu¨g(t)
o So, the effective force is Peff(t)=−mu¨g(t). The given ground acceleration a(t) is u¨g(t).
o The problem defines a(t) as a triangular pulse. For example, if it starts at amax and goes to 0 at td
: a(t)=amax(1−t/td) for 0≤t≤td a(t)=0 for t>td
3. Solve the SDOF Equation for Relative Displacement (u(t)):
o u¨+ωn2u=−u¨g(t)
o This is a forced vibration problem for an undamped SDOF system.

o The solution u(t) can be obtained using the Duhamel Integral (convolution integral) or by solving
piecewise:
 For 0≤t≤td: u(t)=−ωn1∫0tu¨g(τ)sin(ωn(t−τ))dτ
 For t>td: The system undergoes free vibration with initial conditions determined from u(td) and
u˙(td).
o Since the question asks for u(t=0.1sec), you will need to evaluate the integral or the piecewise
solution at that specific time