Advanced

FE-analys av luftinlopp samt karaktärisering av EPDM-gummi

Johannesson, Linda LU (2015) In TFHF-5201 FHL820 20151
Solid Mechanics
Abstract
The aim of this study was to examine if simplified linear analysis can be used to capture the complex and non-elastic behaviour that filled rubber exhibits. Linear analysis includes both linear elastic material models and a linear solving technique. The studied part in this thesis was a hose, connecting the air-inlet and turbo in a Scania truck engine. The studied material was a filled EPDM-rubber. This study included both testing of the material, curve fitting of material data to hyperelastic and linear elastic material models and FE-simulations in Abaqus and Catia GAS.

Hyperelastic material models were used resulting in good correspondence with reality. However, this simplification requires good knowledge of the application and that... (More)
The aim of this study was to examine if simplified linear analysis can be used to capture the complex and non-elastic behaviour that filled rubber exhibits. Linear analysis includes both linear elastic material models and a linear solving technique. The studied part in this thesis was a hose, connecting the air-inlet and turbo in a Scania truck engine. The studied material was a filled EPDM-rubber. This study included both testing of the material, curve fitting of material data to hyperelastic and linear elastic material models and FE-simulations in Abaqus and Catia GAS.

Hyperelastic material models were used resulting in good correspondence with reality. However, this simplification requires good knowledge of the application and that the hyperelastic model is fitted to material data at expected strain amplitudes, strain rate, temperature and eventually, loading history. The difference in the collapse load between simulations and testing
of the studied part was just 2.5 %, which indicates that the chosen material model captures reality very well in this application.

Further simplifications were made evaluating a linear elastic model. The result differed slightly from the hyperelastic model, but still gave a surprisingly good approximation with just 6 % difference in stiffness, when the model was fitted to material data with strain amplitude of 10 %. However, it is important to note that the difference between linear elastic and hyperelastic models is expected to be significantly greater in other applications, where the strain amplitudes at the load bearing regions are higher.

A non-linear solving technique was not expected to be necessary in this application due to the relatively small strain amplitudes of 10 % in the load bearing regions. The difference in collapse load between simulations, using a linear and a non-linear solving technique, was as great as 72 %. A difference of this magnitude is not acceptable, and it is thought to be due to the deformations becoming large and that the direction of the load varies significantly during the deformation process. A non-linear solving technique, where the stiffness matrix and the geometry is updated in every load increment, is thus necessary in order to achieve a reasonable result. In other applications, where the geometry is less complex and the deformations are smaller, a linear solving technique is expected to be adequate. (Less)
Popular Abstract (Swedish)
Pålitliga simuleringar av komponenters hållfasthet kräver att materialets egenskaper studeras ordentligt. Gummi med fyllmedel uppvisar olika egenskaper beroende på tillämpning. I denna studie undersöks
möjligheten att använda förenklingar vid simulering av materialets egenskaper.
Please use this url to cite or link to this publication:
author
Johannesson, Linda LU
supervisor
organization
alternative title
Modelling of filled EPDM rubber turbo hose subjected to underpressure
course
FHL820 20151
year
type
H2 - Master's Degree (Two Years)
subject
keywords
icke-linjär, materialstyvhet, styvhet, lasthistorik, temperatur, töjningshastighet, töjningsamplitud, töjning, linjär-elastisk, linjärelastisk, hyperelastisk, kurvanpassning, EPDM-gummi, luftinlopp, turbokrök, gummibälg, lastbil, gummi med fyllmedel, linjär analys, deformation, non-linear, stiffness, loading history, temperature, strain rate, strain amplitude, strain, hose, turbo, Catia GAS, Catia, Abaqus, FEM, FE-analysis, FE-simulations, linear elastic, hyperelastic, curve fitting, truck, Scania, EPDM, filled rubber, linear analysis
publication/series
TFHF-5201
report number
5201
language
Swedish
id
7762568
date added to LUP
2015-12-02 15:34:09
date last changed
2015-12-02 15:34:09
@misc{7762568,
  abstract     = {The aim of this study was to examine if simplified linear analysis can be used to capture the complex and non-elastic behaviour that filled rubber exhibits. Linear analysis includes both linear elastic material models and a linear solving technique. The studied part in this thesis was a hose, connecting the air-inlet and turbo in a Scania truck engine. The studied material was a filled EPDM-rubber. This study included both testing of the material, curve fitting of material data to hyperelastic and linear elastic material models and FE-simulations in Abaqus and Catia GAS.

Hyperelastic material models were used resulting in good correspondence with reality. However, this simplification requires good knowledge of the application and that the hyperelastic model is fitted to material data at expected strain amplitudes, strain rate, temperature and eventually, loading history. The difference in the collapse load between simulations and testing
of the studied part was just 2.5 %, which indicates that the chosen material model captures reality very well in this application.

Further simplifications were made evaluating a linear elastic model. The result differed slightly from the hyperelastic model, but still gave a surprisingly good approximation with just 6 % difference in stiffness, when the model was fitted to material data with strain amplitude of 10 %. However, it is important to note that the difference between linear elastic and hyperelastic models is expected to be significantly greater in other applications, where the strain amplitudes at the load bearing regions are higher.

A non-linear solving technique was not expected to be necessary in this application due to the relatively small strain amplitudes of 10 % in the load bearing regions. The difference in collapse load between simulations, using a linear and a non-linear solving technique, was as great as 72 %. A difference of this magnitude is not acceptable, and it is thought to be due to the deformations becoming large and that the direction of the load varies significantly during the deformation process. A non-linear solving technique, where the stiffness matrix and the geometry is updated in every load increment, is thus necessary in order to achieve a reasonable result. In other applications, where the geometry is less complex and the deformations are smaller, a linear solving technique is expected to be adequate.},
  author       = {Johannesson, Linda},
  keyword      = {icke-linjär,materialstyvhet,styvhet,lasthistorik,temperatur,töjningshastighet,töjningsamplitud,töjning,linjär-elastisk,linjärelastisk,hyperelastisk,kurvanpassning,EPDM-gummi,luftinlopp,turbokrök,gummibälg,lastbil,gummi med fyllmedel,linjär analys,deformation,non-linear,stiffness,loading history,temperature,strain rate,strain amplitude,strain,hose,turbo,Catia GAS,Catia,Abaqus,FEM,FE-analysis,FE-simulations,linear elastic,hyperelastic,curve fitting,truck,Scania,EPDM,filled rubber,linear analysis},
  language     = {swe},
  note         = {Student Paper},
  series       = {TFHF-5201},
  title        = {FE-analys av luftinlopp samt karaktärisering av EPDM-gummi},
  year         = {2015},
}