Quantification and Implementation of Structural Damping in MBS Flexible Bodies

Abstract

In multi body simulation (MBS) using flexible bodies, damping is an important model input to be able to predict the response of a structure with sufficient precision. The purpose of this
thesis has been to estimate the structural damping of an experimental setup and implement the estimated damping into MBS-models of the setup.

In this thesis the modal properties in terms of natural frequencies, mode shapes and modal damping of a chassis frame with a number of attached components are identified through
physical testing. The experimental setup is modeled in the MBS software MSC Adams using both flexible and rigid bodies coupled together, a total of three different models are created.
Two with different modeling of the bolted joints in the setup and one where components are modeled as separate flexible bodies and coupled with joints. The modal properties of the
models are compared to the identified modal properties of the setup.

The simplification in modeling of the bolted joints results in small differences in modal properties of the model compared to more detailed modeling of the bolted joints. Modeling of
components as separate flexible bodies results in lesser consistency compared to identified modal parameters, making it more difficult to mimic the damping of the experimental setup.

For each of the models a damping function is suggested and implemented,based on the identified modal damping of the physical system. Simulation is then performed, comparing the
different models ability to predict the pseudo damage relative to the physical system.

The suggested damping functions shows improved ability to predict pseudo damage for Components attached to the chassis frame. The prediction of pseudo damage on the frame side
members and cross members is less sensitive to the choice of damping in the MBS-model. Parts of the modeling procedure as well as the implementation of identified damping shows room for improvement and further studies are recommended.

Popular Abstract

In heavy vehicle development, simula-
tion models are widely used to evaluate
designs and concepts in an early stage
in the product development process,
eliminating the need of physical proto-
types. In this thesis the focus has been to
improve simulation models by measuring
the damping of a physical system, and im-
plementing the damping as a model input.

Damping can be explained as the removal of
mechanical energy from a system in motion and
is the reason that the motion will eventually
stop unless energy is supplied to the system
at an equal or larger rate than it is dissipated.
Many dierent sources of damping exists, in this
work the most important one is friction between
parts bolted together and within materials such
as rubber. Friction transforms kinetic energy
into heat mostly.

When creating a model of a dynamic me-
chanical system, such as a heavy vehicle chassis
in this case, knowledge of the damping is of
great importance for successful simulations and
prediction of for example the life span of the
vehicle. In this thesis a simulation model of a
part of a heavy vehicle was created, based on
an existing experimental setup. The damping
was measured in the experimental setup and from the measurement results damping was
implemented into the simulation model. The
simulation results were nally compared to the
physical measurements.

The results show that knowledge of and
correct modeling of the damping in the system
is crucial in order to obtain reliable results
from simulation models. Implementing damping
based on measurements can greatly improve the
simulations, however the implementation itself
proves not to be straight forward. The study
also shows that the detail level in the modeling
together with the modeling of the bolted joints
also aects the ability to successfully predict
the behaviour of the system. In conclusion the
study contributes to increasing the knowledge
of damping in the physical system as well as in
the model and how the model can be improved.
However, further improvements can be made
and the thesis suggests future studies in the field.

As the development of computers continues at
a fast pace, the amount of computational power
we have access to is growing. More computa-
tional power means that we can create better,
more detailed and more complex simulation
models. For the models to be useful they must
be veried and compared to real measurements.
The use of this thesis is to verify the currently used models, showing strengths and weaknesses.
The thesis also suggests how to improve the
current models with respect to damping. Our
models will only serve useful if we are aware of
their limitations.

The work in the thesis was conducted in
cooperation with heavy vehicle manufacturer
Scania, who provided the necessary parts,
equipment and knowledge. Figure 1 shows
the experimental setup, which was subject to
random vibration while the acceleration was
measured at dierent locations. From the
acceleration data information about energy
dissipation and the deformation of the dierent
parts was obtained.

The information from the physical experiment
was then used to make the simulation model
shown in Figure 2 mimic the motion of the physi-
cal system, using the same random vibration sig-
nal as input.

Improving the simulation models opens up
the possibility in the future to completely
remove the need for physical prototypes and
physical testing. As the simulation models are
easily modied more concepts can be tested and
optimized virtually, resulting in better products.
From the results it is shown that the damping
of the experimental system is successfully identi-
ed and implemented into the simulation model.
The implementation leaves room for improve-
ment. Finally it is concluded that the prediction
of the lifespan of the involved components is im-
proved through better knowledge and modeling
of the damping in the system.

• author: Flink, Anton
• supervisor: Kristina Nilsson; Anders Ahlström
• organization: Mechanics
• course: FME820 20152
• year: 2015
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• publication/series: Diploma work
• language: English
• id: 8034398
• date added to LUP: 2015-09-30 16:37:35
• date last changed: 2015-10-01 13:38:12