Vibration in An Unbalance Multi-Stage Rotor - A Review

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 17

VIBRATION IN AN UNBALANCE MULTI-

STAGE ROTOR
- A REVIEW
SOUNDER R(17BME157)
UG STUDENTS - THIRD YEAR ‘A’ SECTION
DEPARTMENT OF MECHANICAL ENGINEERING
KUMARAGURU COLLEGE OF TECHNOLOGY
AIM

The main objective of this is to predict lateral vibration phenomena


arising in the multi-stage rotor (seven stages) with unbalance,
including damping and gyroscopic effects by using finite element
method(FEM) model.
DESCRIPTION

• The rotor dynamic analysis includes mathematical and experimental determination of


the first and second critical speeds of the rotor and the assessment of the effects induced
by the different unbalance combinations.
• In this section, the procedures and tools, both theoretical and experimental are used to
conduct this research.
• The geometrical and inertial characteristics of the rotor and its impellers are defined
along the elastic and damping properties of the rotor and its supports.
• The numerical model is developed through the FEM.
NUMERICAL MODEL

• The
  general equation that governs the behavior of the rotor
[𝑀]{̈} + [𝐶]{̇} + [𝐾]{x} = {𝐹(𝑡)}
[M] is the mass matrix,
[C] is the damping matrix,
[K] is the stiffness matrix,
{F} is the external force vector,
{𝑥̈},{𝑥̇} and {𝑥} are the acceleration, speed, and position vectors.
NUMERICAL MODEL
•• The
  properties of the supports in the rotor model, the global matrices are described as follows:
[] = [] + []
[] = []+[] + []
[] = [] + []
[𝑀𝑐 𝑒] and [𝐾𝑐 𝑒] are elements of mass and stiffness,
[Mr] is the mass matrix of the disk,
[Cr] is the support damping matrix,
[Kr] is the support stiffness matrix,
[Cd] is the gyroscopic effects matrix of the disk,
[CC] is the gyroscopic effects matrix of the shaft.
NUMERICAL MODEL

• The stiffness matrix of the supports is

• The damping matrix of the supports is


NUMERICAL MODEL

• If the rotor motion is represented through state vectors, where 𝑦̇ = [𝑥̈,𝑥̇] and 𝑦 = [𝑥,,𝑥], the equation is

• 0 is the null matrix.


• Assuming that the first matrix is -S and the second −R, the following expression can be established:
−Sẏ + Ry = 0
NUMERICAL MODEL

•  Assuming a solution like 𝑦 = 𝐶𝑌𝑒𝜆𝑡, an eigen-problem is generated as follows:


[𝑅 − 𝜆𝑆]𝑌 = 0
[𝐴 − 𝜆𝑆]𝑌 = 0
[𝑆 − 𝑙/𝜆]𝑌 = 0
Knowing that 𝐴 = 𝑆, eigenvalues λ can be extracted from matrix A through Ansys
and Matlab.
ROTOR GEOMETRY

• The rotor dynamic system studied


comprises a rotor with seven rigid
disks supported by two bearings.
• Disks and shaft are made with
stainless steel with Young’s
modulus of 193 GPa, Poisson’s
ratio is 0.3 and density of 8080
kg/m3.
ROTOR FEM MODEL

• The modeling of the lateral vibration


phenomenon was conducted in Matlab
geared toward the solution of the
mathematical formulation.
• For the analysis, the rotor was divided into
28 elements to have a simplify discretization
but a good accuracy level in the results of the
mathematical model.
• The seven impellers are located at nodes 9,
11, 13, 15, 17, 19, and 21, respectively, while
the bearings are located at nodes 6 and 24.
ROTOR FEM MODEL

• The results obtained through Ansys will be used as an


additional theoretical way of checking the current
theoretical model implemented in Matlab.

• The bearings are modeled with Combin14 elements


that are uniaxial tension compression elements with up
to three degrees of freedom in each node and consider
the effect of stiffness and damping.

• The CONTA174 and Targe170 elements are used to


interconnect the different parts of the rotor.

• The coefficients of stiffness and damping in the lateral


directions are introduced as real constants, and the
gyroscopic effects on the rotor are activated by the
CORIOLIS command.
EXPERIMENTAL DESIGN
• Experimental installation is composed of six main systems (structural support, power source, data
acquisition system, security system, rotating system and control system) and each of them groups
together components with specific functions for their proper operation.
• The dynamic behavior of the rotor is analyzed through the evaluation of levels of lateral vibration
severity in the spectrum graphs, rotor performance analysis when passing through critical speeds
and orbit graphs.
• The lateral vibration of the rotor is registered with two accelerometers installed orthogonally in
the journal bearings, according to technical standard ISO 10816 .
• The signals of the accelerometers are taken to a computer and using the code developed in the NI
Signal Express software ; the fundamental wave-time, spectrum and orbit graphics are displayed.
EXPERIMENTAL DESIGN

• In order to determine the dynamic behavior of the rotor with unbalance,


test unbalance masses of 2 grams were placed in the impellers.
• After registering the corresponding graphs, the unbalance mass is
changed to another angular position.
• This procedure was performed for each impeller.
• The seven impellers use dovetail housing to facilitate the addition of
unbalance masses
DISCUSSION

• Experimental tests were conducted, and it was observed that the highest
instability of the rotor occurs at a frequency range between 46 Hz and 50 Hz;
mainly at a frequency of 48 Hz, where the first lateral critical speed is
observed.
Wave amplitude vs time at 48 Hz for the horizontal Wave amplitude vs time at 48 Hz for the
axis. vertical axis.
DISCUSSION

DIAGRAM OF THE VIBRATION AMPLITUDE DIAGRAM OF THE VIBRATION AMPLITUDE


VS THE ROTATING SPEED FOR THE VS THE ROTATING SPEED FOR THE VERTICAL
HORIZONTAL AXIS.
LEARNING
• The effect of the unbalance induced in the rotor by addition of mass causes the frequencies, where critical
speeds are normally presented when the rotor is properly balanced, to move to lower frequencies.
• The theoretical model of the multi-stage rotor solved in Matlab and considering the effects of unbalance,
presents a percentage variation of up to 8.7% with respect to the experimental value.
• This variation is attributed to the considerations taken when establishing the data collection frequencies
in the experimental measurement and the residual unbalance of the multi-stage rotor.
• This leads to a low computational cost, but keeping the error below 10%, even for cases of unbalance.
• The current model must be used with care in the analysis of critical speeds in turbo machinery.
• The importance of the theoretical model presented in this paper is that despite including the elastic and
damping properties of the rotor and supports, gyroscopic effects and unbalance of the impellers,
maintains a high level of simplicity.
• It is recommended only for cases of small unbalances and considering a probable region of critical speed
consisting of a bilateral band of ± 10% around the calculated critical speeds.
THANK YOU

You might also like