Dynamics PhD Exam - UC Berkeley 2018
Dynamics PhD Exam - UC Berkeley 2018
An undamped SDOF system with mass m and stiffness k is initially at rest and is then subjected to a
full-cycle sine pulse of ground motion, as shown in Figure 1. The natural period of the system is Tn
= td /2. Determine the deformation, u(t), for the time interval 0 < t < td .
𝑢𝑢̈ 𝑔𝑔
𝑢𝑢̈ 𝑔𝑔0
−𝑢𝑢̈ 𝑔𝑔0
Figure 1
The beam in Figure 2 has a uniform mass per unit length of m(x) = m and a uniform bending
stiffness of EI(x) = EI. Using Rayleigh’s method, estimate the fundamental natural frequency of the
beam.
Recall:
EI(x), m(x)
L 2L L
Figure 2
Name: ________________________________
Figure 3a shows a small mass m and a large mass 5m that can only translate in the vertical direction.
(a) Determine the natural frequencies and mode shapes.
(b) The system experiences a vertical ground motion which can be described by the response
spectra in Figure 3b. Using modal superposition, estimate the maximum force in the spring
between the two masses during the earthquake. Assume that all modes of the system have
5% damping.
m
5m
2k 2k 𝑢𝑢̈ g
Figure 3a
Figure 3b
UNIVERSITY OF CALIFORNIA, BERKELEY Dept. of Civil and Environmental Engineering
Spring Semester 2018 Structural Engineering, Mechanics and Materials
Name: ________________________________
QUESTION 1 [30%]:
An air-conditioning unit weighing 1200 lb is bolted at the middle of two parallel simply supported
steel beams (E = 30,000 ksi) as shown below. The clear span of the beams is L = 8 ft. The second
moment of cross-sectional area of each beam is I = 10 in4. The motor in the unit runs at 300 rpm
and produces an unbalanced vertical force of 60 lb at this speed. Neglect the weight of the beams
and assume ζ = 1% viscous damping in the system. Determine the amplitudes of:
1) steady-state deflection, and 2) steady-state acceleration (in g’s) of the beams at their
midpoints which result from the unbalanced force.
Hints:
Deformation response factor:
g = 386 in/sec2
1
QUESTION 2 [70%]:
The umbrella structure shown below has 3 DOF with the listed eigen solution. It is made of
standard steel pipe and has the following properties: I = 28.1 in4, E = 29,000 ksi, weight = 18.97
lb/ft, m = 1.5 kips/g, and L = 10 ft. It is required to determine the peak response of this structure
to horizontal ground motion characterized by the design spectrum shown below (for 5%
damping) scaled to 0.20g peak ground acceleration and using the SRSS modal combination rule.
It is required to perform the following: 1) Check that the distributed weight of the umbrella
structure can be neglected (i.e. less than 10%) compared to the lumped weights and find the
corresponding mass matrix, 2) Compute the displacements u1, u2, and u3, and 3) Computed the
bending moments at the base of the column “b” and at location “a” of the beam.
2
Name: ________________________
University of California at Berkeley Structural Engineering, Mechanics & Materials
Civil and Environmental Engineering Fall Semester 2017
Problem 1
The simple structure shown below weighs 1,000 kips and has a period of 1.25 sec. It has no
viscous damping. It is subjected to the impulsive load shown in the figure. If the structure is to
be designed such that it has only one-third the strength necessary to keep it elastic, what is the
expected peak lateral displacement of the structure?
P 1200 kips
P(t)
1000 k
Time
Problem 2
Consider the elastic two degree-of-freedom shear building shown below. Stories and floors are
numbered starting at the bottom of the building (like an elevator). Floor masses and story
stiffnesses are indicated in the figure below. You should ONLY consider the first mode dynamic
response to horizontal base seismic excitations. The structure may be assumed to have zero
inherent viscous damping. It is desired to increase the effective viscous damping in the first
mode to 15% by adding linear fluid viscous dampers between each floor as shown below. The
viscous dampers may be assumed to provide NO stiffness to the structure, and can be assumed to
provide a transverse velocity dependent force between each level of Fd = Cv, where C is the
damping coefficient, and v is the relative lateral velocity between adjacent floors. Identical
dampers are to be placed in each story. The braces used to hold the dampers in place may be
assumed to be inextensible. You may make any other common simplifying assumptions in your
analyses, but please indicate all such assumptions.
Determine the value of C1 necessary to achieve the 15% effective viscous damping ratio in the
first mode (note C2 = C1/2).
UNIVERSITY OF CALIFORNIA, BERKELEY Dept. of Civil and Environmental Engineering
Fall Semester 2016 Structural Engineering, Mechanics and Materials
D(1)=? D(2)=?
V(1)=0.95
V [m/sec]
V(2)=0.60
A [g]
A(1)=?
A(2)=?
Tn [sec]
Fig. Q3: (a)(b)
Fig. 1: (a) Displacement, Displacement, (b) Pseudo-velocity,
Pseudo-velocity, (c) Pseudo-acceleration
(c) Pseudo-acceleration response spectra (ζ=0.02)
response spectra forfor
(ζ=0.02) El Centro ground
El Centro ground motion motion .
QUESTION 2 [80%: Part a(i)=25%; Part a(ii)=20%; Part a(iii)=5%; Part b(i)=25%; Part b(ii)=5%]:
a) Let’s take a look at a reactor building at the Fukushima Daichi nuclear power plant, shown
in Fig. 2a. A two-story frame shown in Fig. 2b is used to model the reactor building. Note
the distribution of masses (m vs. 8m) that reflect the weight of the spent fuel pool. The
building has the following properties: m = 5 kip−sec2/in, k = 100 kips/in. The primary
containment is considered to be rigid at this preliminary stage of the analysis. There is a
gap sized ∆ between the primary containment and the spent fuel pool. A pseudo-
acceleration design spectrum given in Fig. 2c is derived from the 2011 Great Tohoku
earthquake records. It is required to conduct the following:
i. Determine natural frequencies and mode shapes of the building.
1
ii. Using an appropriate modal combination (MC) rule, determine the smallest size of
the gap ∆req (rounded to nearest ¼ in.) to avoid pounding given the design spectrum.
iii. Briefly, justify your choice of the MC rule (no calculations needed)
b) A tsunami wave force idealized, as shown in Fig. 2d with two equivalent lateral forces,
P(t) at the spent fuel pool level and P(t)/2 at the roof level, where P(t) is modeled using
rectangular pulse function (duration td=1.0 sec., Fig. 2e). Conduct the following:
i. Determine the maximum lateral displacement of the reactor building at the spent
fuel pool level.[Hint: The dynamic response factor Rd=2.0 if td/Tn ≥ 0.5]
ii. Assuming the primary containment in unaffected by the tsunami wave, is the gap
size ∆req determined considering earthquake load in (a(ii)) sufficient to prevent
pounding due to tsunami wave load? State clearly Yes or No with justification.
Primary
Containment
Primary
Containment
2
Name: ________________________
University of California at Berkeley Structural Engineering, Mechanics & Materials
Civil and Environmental Engineering Spring Semester 2016
2. Consider the two-degree of freedom system shown below. The structure is assumed to have
no inherent viscous or hysteretic damping. All of the springs shown have the same stiffness,
but the value of stiffness is not specified. However, the first mode period of the structure is
given as 2 seconds. The structure is subjected to a horizontal earthquake excitation
represented by a simplified response spectrum Dn=3T, were the spectral displacement Dn is
measured in inches, and period is specified in seconds. Modal contributions to response can
be estimated using SRSS methods. It may be assumed that peak responses in each mode can
be represented by the specified spectrum. (60 points)
a. What is the stiffness required of the springs (all equal)?
b. What is the expected maximum force in the spring located between the two masses?
c. If linear viscous dampers are placed in parallel with all of the springs shown, and they
each exhibit the relation Fd = Cv, where v represents the relative velocity acting across
the damper, what is the value of C (expressed in k-sec/in) to give an effective viscous
damping ratio of 15% in the first mode of vibration?
Name: ________________________
Name: ________________________
University of California at Berkeley Structural Engineering, Mechanics & Materials
Civil and Environmental Engineering Fall Semester 2015
2. Consider the elastic two degree-of-freedom shear building shown below. Floor masses and
story stiffnesses are indicated in the figure below. You should ONLY consider the first mode
dynamic response to horizontal base seismic excitations. The structure may be assumed to
have zero inherent viscous damping. It is desired to increase the effective viscous damping
in the first mode to 10% by adding linear fluid viscous dampers between each floor as shown
below. The viscous dampers may be assumed to provide NO stiffness to the structure, and
can be assumed to provide a transverse velocity dependent force between each level of
Fd=Cv, where C is the damping coefficient, and v is the relative lateral velocity between
adjacent floors. Identical dampers are to be placed in each story. The braces used to hold the
dampers in place may be assumed to be inextensible. You may make any other common
simplifying assumptions in your analyses, but please indicate all such assumptions.
Determine the value of C necessary to achieve the 10% effective viscous damping ratio in the
first mode (note C is to be the same for both stories).
Name: ________________________
University of California at Berkeley Structural Engineering, Mechanics & Materials
Civil and Environmental Engineering Spring Semester 2015
Modal superposition involves breaking down the structural response into contributions from each mode of vibration. Given the system with various damping ratios and response spectra, calculate the response in each mode independently using modal mass, stiffness, and damping. The total response, including force in the spring, is then derived by superposing the modal contributions appropriately, often via an SRSS method for maximum values, adjusted for damping and spectral values .
Rayleigh's method involves assuming a trial mode shape for the beam's deflection, typically taking a simple function, such as a sine or cosine that satisfies the boundary conditions. By equating the maximum potential energy and kinetic energy under this assumed mode, you utilize the relation ω² = (potential energy / kinetic energy)max to estimate the natural frequency. Given the uniform properties, this simplifies the calculation as the mode shape assumption often directly lends itself to easily integrated expressions .
To find the necessary friction coefficient, begin with the energy dissipation concept in the cycle of oscillation, where reduction from 4 inches to 2 inches involves energy dissipation by friction. The relation Ff = µW provides the frictional force, where µ is the unknown coefficient. Equate the mechanical energy lost per cycle (based on amplitude reduction) to the work done by frictional force, solve for µ, leveraging the known mass, spring constant, and desired amplitude reduction .
The SRSS (Square Root of the Sum of the Squares) modal combination rule is used to estimate total seismic response by combining individual modal responses. For an umbrella structure subjected to seismic load characterized by a design spectrum, it reduces computational complexity by recognizing that modal responses are statistically independent. This is especially useful given a design spectrum scaled to multiple ground motion levels like in the examination setup for peak response estimation .
To determine the deformation u(t) of an undamped SDOF system subjected to a full-cycle sine pulse, you need to use the differential equation governing the dynamic response of the system. Given the system's natural period as Tn = td / 2, and knowing the initial conditions (initially at rest), you can apply the Duhamel integral using the specified pulse shape to find u(t) during the time interval 0 < t < td .
The amplitude of the steady-state deflection due to an unbalanced force from a rotating machine is obtained using the formula: deflection = (Force / effective stiffness), where the effective stiffness is modified by dynamic amplification due to resonance conditions. Specifically, you consider the force frequency and system's natural frequency to determine the dynamic load factor that affects steady-state deflection .
Checking that the distributed weight is less than 10% compared to lumped weights ensures that the simplification to lumped-mass model is valid. This approximation allows analytical simplicity in dynamic analysis, assuming concentrated masses don't omit significant distributed mass effects. This validation step is crucial to ensure that neglecting the distributed weight doesn't significantly affect dynamic characteristics and response predictions .
Underestimating stiffness in an SDOF system can compromise safety, particularly if sensitive to deformation or acceleration. Reduced stiffness can lead to larger deformations under loading, potentially exceeding design tolerances for equipment attachment, causing structural or mechanical failure. Additionally, underestimated stiffness results in lower natural frequencies, inducing larger resonant accelerations. Evaluating safety requires analyzing these outcomes relative to system-specific sensitivity .
Key assumptions include: the impact being significantly shorter than the structure's natural period (implying a quasi-static response analysis), the isolators acting inelastic-perfectly, and minimal energy dissipation beyond the isolators. These assumptions simplify dynamics to predominantly linear behavior with known dissipation characteristics, allowing for the use of conservation principles in estimating maximum displacements due to an impulsive load .
To compute the damping coefficient for achieving a specific damping ratio, use the relation: damping ratio = (C / (2 * √(M * K))), where C is the damping coefficient per floor, M is the mass, and K is the stiffness for the first mode. Given the desired damping ratio, rearrange to solve for C. For a two-degree-of-freedom structure, analyze in terms of its effective modal mass and stiffness for the first mode to determine the accurate coefficient to provide the desired damping .