- #1
aeroguy77
- 12
- 0
1.0 Introduction
As the thread title suggests, this is indeed referring to a lab experiment. I am not posting here to gain easy solutions to the lab manual questions, rather I'm posting here to gain knowledge on a subject I know very little about. Let me explain the lab experiment in detail so that you can subsequently provide me detailed answers to my questions.
The apparatus (pictures provided) consists of a three mass offset frame with a "shaker" (supplied by a power amp and a wave generator) and accelerometers measuring the plate acceleration via a charge amplifier and oscillator. The plates are considered to be point masses while the stiffness is to be provided by the six supporting columns. The experimental stiffness of the columns is determined by placing a weight and pulley system at the end of the apparatus in order to measure deflections by increasing load.
The following assumptions are made:
- The mass plates are rigid
- All significant vibrating mass is in these plates
- The columns are rigidly welded to these plates and the base
The analysis is to be performed for "in-plane" motion only (i.e no twisting or "out-of-plane" motion is to be considered).
At the beginning of the lab, we set the excitation level to 10 rms and inital amplitude to 880 mV. For each mass (plate) we recorded the corresponding voltages recorded by the accelerometer. We then increased the frequency and then took down additional values. There are two accelerometers. One is always placed on mass1 (plate1) (it measures input acceleration) and the other moves between mass2 and mass3.
2.0 Theory
The deflection is to be defined by a global co-ordinate system. Thus in-plane motion is said to have 3 degrees of freedom (DOF).
[x]^T = [x1 x2 x3]
For N DOF, the free vibration of a system with no damping has the following equation of motion:
[M][X]'' + [K][X] = 0
where,
[X]^T = [X1 X2 . . . Xn]
[M] = Mass matrix, N x N
[K] = Stiffness matrix, N x N
A system with N DOF has N natural frequencies and mode shapes. If the system is in simple harmonic motion with frequency w and deflection (mode) shape [X] then,
[X(t)] = [X]cos(wt) and
{[K]-w^2[M]}[X]
solving for the corresponding natural frequencies (assuming [X] =/ 0):
DET{[K]-w^2[M]} = 0 to obtain
w1^2 w2^2 w3^2 ... wN^2
and
[X]1 [X]2 [X]3 ... [X]N
The mode vectors also have orthogonality properties for (i) different modes and (ii) the same modes:
(i) [X]r^T [M] [X]s = 0
[X]r^T [K] [X]s = 0
(ii) [X]r^T [M] [X]r = Mrr
[X]r^T [K] [X]r = Krr
where Mrr and Krr are know as the modal mass and stiffness of the rth mode. The equation of motion for this mode is:
Mrrqr'' + Krrqr = 0
with natural frequency:
wr^2 = Krr/Mrr
To find the mass matrix and stiffness matrix, we employed the kinetic and strain energy functions. The lab manual conveniently derived the corresponding theoretical matrices.
3.0 My Questions
(1) What is the "excitation" level, what are its units, and how does this effect the overall outcome of the experiment?
(2) What is a mode shape? What is a natural frequency? How are the two related?
(3) Why is w1 and [X]1 the lowest (fundamental) mode frequency and mode shape?
(4) How can the properties of orthogonality be used to determine the modes on complex structures?
(5) Why does an accelerometer give readings in voltages and not accelerations and/or deflections? How can we determine the corresponding mode shape and frequency from this information?
(6) Why does the voltages measured by the accelerometer decrease as the shaker frequency increases?
Thank you very much for your time, I know that as engineers you don't have a lot of it. Also, any insight on the topic will be greatly appreciated. Again, thank you for your time.
As the thread title suggests, this is indeed referring to a lab experiment. I am not posting here to gain easy solutions to the lab manual questions, rather I'm posting here to gain knowledge on a subject I know very little about. Let me explain the lab experiment in detail so that you can subsequently provide me detailed answers to my questions.
The apparatus (pictures provided) consists of a three mass offset frame with a "shaker" (supplied by a power amp and a wave generator) and accelerometers measuring the plate acceleration via a charge amplifier and oscillator. The plates are considered to be point masses while the stiffness is to be provided by the six supporting columns. The experimental stiffness of the columns is determined by placing a weight and pulley system at the end of the apparatus in order to measure deflections by increasing load.
The following assumptions are made:
- The mass plates are rigid
- All significant vibrating mass is in these plates
- The columns are rigidly welded to these plates and the base
The analysis is to be performed for "in-plane" motion only (i.e no twisting or "out-of-plane" motion is to be considered).
At the beginning of the lab, we set the excitation level to 10 rms and inital amplitude to 880 mV. For each mass (plate) we recorded the corresponding voltages recorded by the accelerometer. We then increased the frequency and then took down additional values. There are two accelerometers. One is always placed on mass1 (plate1) (it measures input acceleration) and the other moves between mass2 and mass3.
2.0 Theory
The deflection is to be defined by a global co-ordinate system. Thus in-plane motion is said to have 3 degrees of freedom (DOF).
[x]^T = [x1 x2 x3]
For N DOF, the free vibration of a system with no damping has the following equation of motion:
[M][X]'' + [K][X] = 0
where,
[X]^T = [X1 X2 . . . Xn]
[M] = Mass matrix, N x N
[K] = Stiffness matrix, N x N
A system with N DOF has N natural frequencies and mode shapes. If the system is in simple harmonic motion with frequency w and deflection (mode) shape [X] then,
[X(t)] = [X]cos(wt) and
{[K]-w^2[M]}[X]
solving for the corresponding natural frequencies (assuming [X] =/ 0):
DET{[K]-w^2[M]} = 0 to obtain
w1^2 w2^2 w3^2 ... wN^2
and
[X]1 [X]2 [X]3 ... [X]N
The mode vectors also have orthogonality properties for (i) different modes and (ii) the same modes:
(i) [X]r^T [M] [X]s = 0
[X]r^T [K] [X]s = 0
(ii) [X]r^T [M] [X]r = Mrr
[X]r^T [K] [X]r = Krr
where Mrr and Krr are know as the modal mass and stiffness of the rth mode. The equation of motion for this mode is:
Mrrqr'' + Krrqr = 0
with natural frequency:
wr^2 = Krr/Mrr
To find the mass matrix and stiffness matrix, we employed the kinetic and strain energy functions. The lab manual conveniently derived the corresponding theoretical matrices.
3.0 My Questions
(1) What is the "excitation" level, what are its units, and how does this effect the overall outcome of the experiment?
(2) What is a mode shape? What is a natural frequency? How are the two related?
(3) Why is w1 and [X]1 the lowest (fundamental) mode frequency and mode shape?
(4) How can the properties of orthogonality be used to determine the modes on complex structures?
(5) Why does an accelerometer give readings in voltages and not accelerations and/or deflections? How can we determine the corresponding mode shape and frequency from this information?
(6) Why does the voltages measured by the accelerometer decrease as the shaker frequency increases?
Thank you very much for your time, I know that as engineers you don't have a lot of it. Also, any insight on the topic will be greatly appreciated. Again, thank you for your time.
Attachments
Last edited: