Simplified FEM Model of Paper Machine Roll for Multibody Rolling Contact Analyses

Author(s):  
Veli-Matti Järvenpää ◽  
Erno K. Keskinen

Abstract In this paper a finite element model of a rotating paper machine roll for nip unit rolling contact analyses is discussed. This work presented here is based on the earlier work of the authors presented in [1] and [2]. The major motivations for developing a tailored FE-model including the large spin rotation are firstly to include the complex vibration phenomena as the shell vibrations of the roll structure in the analyses and secondly to reduce the computational costs of the numerical simulations due to the large number of degrees of freedom. The approach used is the use of the modal analysis i.e. to express the dynamics of the roll in terms of the lowest eigenmodes. The equations of motion are at first written in the rotating coordinates and then in addition to this the equations are expressed by using the modal coordinates. Numerical tests executed show that this modeling technique reduces computational costs significantly. Furthermore, use of the (semidefinite) eigenmode basis maintains the vibration characteristics of the roll structure. For verification purposes a test model was constructed and these simulation results were compared to the standard geometrically non-linear finite element analysis.

Author(s):  
Ashwini Gautam ◽  
Chris Fuller ◽  
James Carneal

This work presents an extensive analysis of the properties of distributed vibration absorbers (DVAs) and their effectiveness in controlling the sound radiation from the base structure. The DVA acts as a distributed mass absorber consisting of a thin metal sheet covering a layer of acoustic foam (porous media) that behaves like a distributed spring-mass-damper system. To assess the effectiveness of these DVAs in controlling the vibration of the base structures (plate) a detailed finite elements model has been developed for the DVA and base plate structure. The foam was modeled as a poroelastic media using 8 node hexahedral elements. The structural (plate) domain was modeled using 16 degree of freedom plate elements. Each of the finite element models have been validated by comparing the numerical results with the available analytical and experimental results. These component models were combined to model the DVA. Preliminary experiments conducted on the DVAs have shown an excellent agreement between the results obtained from the numerical model of the DVA and from the experiments. The component models and the DVA model were then combined into a larger FE model comprised of a base plate with the DVA treatment on its surface. The results from the simulation of this numerical model have shown that there has been a significant reduction in the vibration levels of the base plate due to DVA treatment on it. It has been shown from this work that the inclusion of the DVAs on the base plate reduces their vibration response and therefore the radiated noise. Moreover, the detailed development of the finite element model for the foam has provided us with the capability to analyze the physics behind the behavior of the distributed vibration absorbers (DVAs) and to develop more optimized designs for the same.


1991 ◽  
Vol 113 (1) ◽  
pp. 93-101 ◽  
Author(s):  
S. M. Kulkarni ◽  
C. A. Rubin ◽  
G. T. Hahn

The present paper, describes a transient translating elasto-plastic thermo-mechanical finite element model to study 2-D frictional rolling contact. Frictional two-dimensional contact is simulated by repeatedly translating a non-uniform thermo-mechanical distribution across the surface of an elasto-plastic half space. The half space is represented by a two dimensional finite element mesh with appropriate boundaries. Calculations are for an elastic-perfectly plastic material and the selected thermo-physical properties are assumed to be temperature independent. The paper presents temperature variations, stress and plastic strain distributions and deformations. Residual tensile stresses are observed. The magnitude and depth of these stresses depends on 1) the temperature gradients and 2) the magnitudes of the normal and tangential tractions.


Author(s):  
J. Rodriguez ◽  
M. Him

Abstract This paper presents a finite element mesh generation algorithm (PREPAT) designed to automatically discretize two-dimensional domains. The mesh generation algorithm is a mapping scheme which creates a uniform isoparametric FE model based on a pre-partitioned domain of the component. The proposed algorithm provides a faster and more accurate tool in the pre-processing phase of a Finite Element Analysis (FEA). A primary goal of the developed mesh generator is to create a finite element model requiring only essential input from the analyst. As a result, the generator code utilizes only a sketch, based on geometric primitives, and information relating to loading/boundary conditions. These conditions represents the constraints that are propagated throughout the model and the available finite elements are uniformly mapped in the resulting sub-domains. Relative advantages and limitations of the mesh generator are discussed. Examples are presented to illustrate the accuracy, efficiency and applicability of PREPAT.


Author(s):  
L. C. Hau ◽  
Eric H. K. Fung

The finite element method, in conjunction with the Golla-Hughes-McTavish (GHM) viscoelastic model, is employed to model a clamped-free beam partially treated with active constrained layer damping (ACLD) elements. The governing equations of motion are converted to a state-space form for control system design. Prior to this, since the resultant finite element model has too many degrees of freedom due to the addition of dissipative coordinates, a model reduction is performed to revert the system back to its original size. Finally, optimal output feedback gains are designed based on the reduced models. Numerical simulations are performed to study the effect of different element configurations, with various spacing and locations, on the vibration control performance of a “smart” flexible ACLD treated beam. Results are presented for the damping ratios of the first two modes of vibration. It is found that improvement on the second mode damping can be achieved by splitting a single ACLD element into two and placing them at appropriate positions of the beam.


1998 ◽  
Vol 33 (4) ◽  
pp. 263-274 ◽  
Author(s):  
D J Smith ◽  
C G C Poussard ◽  
M J Pavier

Measurements of residual stresses in 6 mm thick aluminium alloy 2024 plates containing 4 per cent cold worked fastener are made using the Sachs method. The measurements are made on discs extracted from the plates. The measured tangential residual stress distribution adjacent to the hole edge are found to be affected by the disc diameter. The measured residual stresses are also in good agreement with averaged through-thickness predictions of residual stresses from an axisymmetric finite element (FE) model of the cold working process. A finite element analysis is also conducted to simulate disc extraction and then the Sachs method. The measured FE residual stresses from the Sachs simulation are found to be in good agreement with the averaged through-thickness predicted residual stresses. The Sachs simulation was not able to reproduce the detailed near-surface residual stresses found from the finite element model of the cold working process.


2018 ◽  
Vol 880 ◽  
pp. 163-170
Author(s):  
Ștefan Cristian Castravete ◽  
Gabriel Cătălin Marinescu ◽  
Nicolae Dumitru ◽  
Oana Victoria Oţăt

The paper studies the behavior of a quarter-car suspension model under periodic road excitation: sinusoidal and bump (trapezoidal shape) for a constant vehicle speed. A theoretical and a finite element model were developed. The theoretical model has two degrees of freedom and a modal and sinusoidal excitation was performed to compare with finite model analysis. The finite element analysis consists of three parts: preload, modal analysis and deterministic external excitation. The study consists of the analysis of forces, displacements and accelerations that are transmitted to the vehicle regarding their variation in time and frequency.


2017 ◽  
Vol 21 (2) ◽  
pp. 422-438 ◽  
Author(s):  
G Martakos ◽  
JH Andreasen ◽  
C Berggreen ◽  
OT Thomsen

A novel crack arresting device is implemented in foam-cored composite sandwich beams and tested using the Sandwich Tear Test (STT) configuration. A finite element model of the setup is developed, and the predictions are correlated with observations and results from a recently conducted experimental fatigue test study. Based on a linear elastic fracture mechanics approach, the developed FE model is utilised to simulate crack propagation and arrest in foam-cored sandwich beam specimens subjected to fatigue loading conditions. The effect of the crack arresters on the fatigue life is analysed, and the predictive results are subsequently compared with the observations from the previously conducted fatigue tests. The FE model predicts the energy release rate and the mode mixity based on the derived crack surface displacements, utilising algorithms for the prediction of accelerated fatigue crack growth as well as the strain field evolution in the vicinity of the crack tip on the surface of the sandwich specimens. It is further shown that the developed finite element analysis methodology can be used to gain a deeper insight onto the physics and behavioural characteristics of the novel peel stopper concept, as well as a design tool that can be used for the implementation of crack arresting devises in engineering applications of sandwich components and structures.


Author(s):  
Tong Y. Yi ◽  
Parviz E. Nikravesh

Abstract This paper presents a method for identifying the free-free modes of a structure by utilizing the vibration data of the same structure with boundary conditions. In modal formulations for flexible body dynamics, modal data are primary known quantities that are obtained either experimentally or analytically. The vibration measurements may be obtained for a flexible body that is constrained differently than its boundary conditions in a multibody system. For a flexible body model in a multibody system, depending upon the formulation used, we may need a set of free-free modal data or a set of constrained modal data. If a finite element model of the flexible body is available, its vibration data can be obtained analytically under any desired boundary conditions. However, if a finite element model is not available, the vibration data may be determined experimentally. Since experimentally measured vibration data are obtained for a flexible body supported by some form of boundary conditions, we may need to determine its free-free vibration data. The aim of this study is to extract, based on experimentally obtained vibration data, the necessary free-free frequencies and the associated modes for flexible bodies to be used in multibody formulations. The available vibration data may be obtained for a structure supported either by springs or by fixed boundary conditions. Furthermore, the available modes may be either a complete set; i.e., as many modes as the number of degrees of freedom of the associated FE model is available, or it can be an incomplete set.


Author(s):  
V. Prakash ◽  
R. J. Montague

Abstract This paper presents the diagnostics of the effects of vibration on the precise placement of electronic components in a surface mount assembly process. Experimental Modal Analysis using present day software/hardware as well as a three dimensional finite element model are performed on the machine structure. Correlation between the experiment and finite element model are also performed and the strength of using the finite element model as a test model for contemplating any design alterations are presented.


Author(s):  
Prabin Pathak ◽  
Y. X. Zhang

A simple, accurate and efficient finite element model is developed in ANSYS for numerical modelling of the nonlinear structural behavior of FRP strengthened RC beams under static loading in this paper. Geometric nonlinearity and material non-linear properties of concrete and steel rebar are accounted for this model. Concrete and steel reinforcement are modelled using Solid 65 element and Link 180 element, and FRP and adhesive are modelled using Shell 181element and Solid 45 element. Concrete is modelled using Nitereka and Neal’s model for compression, and isotropic and linear elastic model before cracking with strength gradually reducing to zero after cracking for tension. For steel reinforcement, the elastic perfectly plastic material model is used. FRPs are assumed to be linearly elastic until rupture and epoxy is assumed to be linearly elastic. The new FE model is validated by comparing the computed results with those obtained from experimental studies.


Sign in / Sign up

Export Citation Format

Share Document