Multiaxial block programs: A new method for the efficient testing of the durability of elastomeric bearings (#57)
T. A. Thüringer1
1 TU Dresden, Institut für Automobiltechnik, Dresden, Saxony, Germany
Elastomeric bearings are used inter alia for load transfer, vibration isolation and damping in different areas in the motor vehicle. To ensure that these functional properties are maintained despite the high loads over the life of the vehicle, these components are tested with highly complex and time-consuming test cycles. Multi-axial high-dynamic loads in three translative or rotational directions currently correspond to the state of the art. These test cycles require high performance test technology. Due to the increasing number of vehicles and drive variants and ever shorter development times, it is becoming increasingly difficult to carry out this complex test for every elastomer bearing.
The aim is to reduce the durability assurance to the essential damage component in order to significantly shorten the complexity of the signals and the duration of the assurance. The required boundary conditions are the preservation of the damage pattern and a comparable change of the bearing characteristics to the initial Test.
In a first step, the current durability testing method was investigated. The behaviour of the bearings to damage was determined on the basis of optical characteristics, such as cracks or stop abrasion, and general characteristic changes, such as stiffness properties. Fatigue, settling and abrasion were defined as relevant damage mechanisms in the durability testing of elastomers.
In order to reproduce these damage mechanisms, different methods of the Structural Durability were analysed to simplify the load-time cycles and checked for their potential. In addition to classical omission approaches or uniaxial tests in the main load direction, a combination of different methods proves to be highly promising.
The method presented in this thesis reproduces only the relevant damage mechanisms with a multiaxial block program and therefore requires significantly less time and has lower test bench requirements than the given load cycles. The realization is done with different load blocks:
-Peak Blocks: With these blocks, high critical load amplitudes with individual sinusoidal oscillations are multiaxially simulated.
-Fatigue Blocks: Material fatigue is tested with a sine wave with mean load amplitude and a high number of cycles.
-Friction Blocks: The abrasion of the elastomer stops is reproduced by a special combination of sinusoidal oscillations. The impact is caused by an increased preload. Abrasion then occurs multiaxially in the other relevant spatial directions.
The challenge is to adjust these different blocks individually for each bearing according to a defined method in order to reproduce the given load situation more effectively. In order to make this possible, the individual damage mechanisms were investigated separately in extensive measurement campaigns.
The method is validated by real and virtual tests. FEM simulations enable the comparison of local strains under critical load scenarios. Real tests are used to compare static and dynamic characteristic changes as well as local cracks between the multi-axial block program and the original load-time cycles. The results show that multiaxial block programs can reduce the duration of an durability test to more than 20%.In addition, the demands on the testing technology are significantly lower and the reproducibility of the results considerably higher.
The present paper reports on the methodical approach in a cooperative research project to increase efficiency in durability testing and shows current results.
too few characters...
Keywords: durability testing, elastomer bearing, efficiency, automotive, simulation
New modelling method with the focus on high dynamic loads, hydraulic damping and multi-axial excitation for use in load data determination using multi-body simulation (#63)
S. Ernst1, K. Büttner1, G. Prokop1
1 Technical Universaty Dresden, Chair of Automobile Engineering, Dresden, Saxony, Germany
For the design of elastomeric bearings, it is necessary to understand the deformation states that occur under operating loads. The aim and challenge of the presented research work is to record these deformation processes in the form of physical-mathematical models and to make them applicable for development relevant simulations.
On the base of a comparison between the current standard characterization of elastomeric bearings and the excitation amplitudes and frequencies occurring during their operation, it could be determined that the operating range is not fully covered by the bearing characterization (see Figure 1 - Motivation). Excitation amplitudes and frequencies could be extracted from a multiaxial service load test recorded on a multi-axial high dynamic component test rig. For the characterization of the dynamic properties, the loss angle and the dynamic stiffness at low amplitudes (usually less than 1 mm) are used according to the current state of the art. With this method, not all excitation states occurring during operation can be recorded from the point of view of structural durability.
Therefore, within the scope of the investigations described here, selected complex aggregate bearings were fundamentally investigated statically and dynamically outside the standard characterization. Physical effects could be investigated, which can only be represented to a limited extent with existing bearing models. These are multi-axial static, dynamic and transient effects. In order to map these effects, a new modelling method was created which is optimised for the mapping of high dynamic loads, hydraulic damping and multi-axial excitations. The model focus is placed on the range of structural durability and the model is to be used primarily for the determination of operational load signals with the aid of multi-body simulation. A modular modelling approach is chosen, whereby the model complexity can be adapted according to the application. To determine the model parameters, a parameter identification procedure based on optimization algorithms is created. The special feature of this identification method is the use of complex multi-axial operating load signals to determine the dynamic parameters. The described process is shown schematically in Figure 2- Overview Process New Modelling Method.
On the basis of multi-axial test rig measurements, a clear improvement of the image quality of measured load signals compared to standard models could be demonstrated for the newly developed modelling approach. The simulation error and the signal damage are considered as evaluation criteria on the basis of a standard fatigue strength evaluation. Different types of elastomeric and hydromounts were investigated for the validation. From this it can be deduced that the methods work robustly and will be suitable for practical use after a few adjustments.
Keywords: Elastomeric Modelling, Multibody Simulation, Multi-Axial Excitations, Load Data Determination, High Dynamic Loads
Development and validation of models for durability estimation for LSI exhaust manifolds under high temperature loads with consideration of weld seams (#75)
M. Wendt1, W. Rehm2
1 Daimler AG, Hamburg, Germany
At the Daimler plant in Hamburg, exhaust manifolds are built from high-strength sheet metal parts that are manufactured in drawing and inner high-pressure forming processes and assembled in a welding process.
Exhaust manifolds are exposed to combined thermal and mechanical loads with variable amplitudes: flue gases with temperatures of up to 1000°C flow through the manifold. Depending on the driving profile, the manifolds are impacted by alternating hot and cold temperatures. The main stress is the thermo-mechanical fatigue (TMF): thermal expansion and contraction during heating and cooling causes changing mechanical loads.
To assess these impacts before manufacturing and testing of prototypes, calculation models have been developed in recent years for essential manifold materials.
The core of the models is a non-linear kinematic hardening model. It was elaborated and adapted on the basis of a number of low cycle fatigue (LCF) tests and creep tests in a wide range of temperatures. Thus, the stiffness during heating and cooling can be considered correctly in a finite element calculation. To be able to estimate the life time of the components, the result of the FE calculation is assessed using a fatigue model that was also developed on the basis of the LCF tests. Thus, the number of load cycles until incipient cracks occur can be predicted.
In order to verify the predictions in the practice, a validation test was carried out: a curved pipe was mounted on a hot gas test rig and the number of load cycles up to a crack were determined. A problem was the detection of cracks. As a result, it could be shown that the predictions are correct within the expected tolerance range.
Furthermore, it was investigated how welds can be evaluated. On manifolds constructed from sheet metal, cracks usually occur at welds or in the heat-affected zone. Therefore, the assessment of the welds is essential. LCF tests using welded samples were made. On the basis of these tests, the hardening model could be adjusted for welding seams and subsequently, the life model could be checked.
On the basis of these investigations, an approach to the calculation of exhaust manifolds is created. Variable temperature profiles can be considered. The model can be used as a basis for calculations with variable amplitude loading, as well.
Keywords: exhaust manifold, welding, thermal loads, hardening model, Low cyle fatigue
Life dependent material parameters applied to lifetime calculation under multiaxial variable-amplitude loading (#83)
A. Karolczuk2, J. Papuga1, K. Kluger2
1 FME, Czech Technical University in Prague, Dept. of Mechanics, Biomechanics and Mechatronics, Prague 6, Czech Republic
Modelling of fatigue damage evolution based on simulation of the dislocation movement, strain localization, etc is still at the early stage even under uniaxial cyclic loading . On the other hand, the phenomenological models are proposed, which enable to predict fatigue life based on the experimentally determined fatigue properties. However, the applicability of such models is limited to conditions under which the fatigue properties were obtained. Among the multiaxial fatigue damage models, the concept of the critical plane is most often elaborated. The concept is based on the observation that fatigue damage in metals occurs on certain planes. Stress and strain vectors acting related to particular plane orientations act as the primary driving forces for the crack initiation. A wide range of linear and non-linear functions of stress and strain vector components on the critical plane are proposed to calculate the equivalent scalar quantity [2,3]. These phenomenological models mostly apply material parameters derived from the fatigue limits of at least two uniaxial loading, i.e. push-pull and torsion. In the recent research [4–6], it is shown that the applicability of the well-known fatigue strength models could be extended to the finite life if applied material parameters are life dependent. Instead of fatigue limits the fatigue characteristics (S-N curves) must be applied in such approaches. The published research [4–6] concerned only constant amplitude loading so far. Under variable amplitude loading, the value of damage parameter for each extracted cycle varies - different values of material parameters must be applied for each cycle. In such way, the damage induced by each extracted cycle is correctly estimated.
The first aim of the present research is to present the methodology of correct damage calculation for each extracted cycle through application of a few well-known multiaxial fatigue criteria while using life dependent material parameters. The second aim is to verify the proposed methodology, if applied to selected fatigue criteria. Based on the experimental fatigue tests performed on S355 steel under multiaxial variable-amplitude loading, the capability to correctly estimate the fatigue life by the new approach has been shown.
The publication was financed from a Grant by National Science Centre, Poland (Decision No. 2017/25/B/ST8/00684).
 Polák J, Man J. Experimental evidence and physical models of fatigue crack initiation. Int J Fatigue 2016;91:294–303. doi:10.1016/j.ijfatigue.2016.02.021.
 Karolczuk A, Macha E. A review of critical plane orientations in multiaxial fatigue failure criteria of metallic materials. Int J Fract 2005;134:267–304. doi:10.1007/s10704-005-1088-2.
 Papuga J. A survey on evaluating the fatigue limit under multiaxial loading. Int J Fatigue 2011;33:153–65. doi:10.1016/j.ijfatigue.2010.08.001.
 Karolczuk A. Analysis of revised fatigue life calculation algorithm under proportional and non-proportional loading with constant amplitude. Int J Fatigue 2016;88:111–20. doi:10.1016/j.ijfatigue.2016.03.027.
 Karolczuk A, Kluger K, Łagoda T. A correction in the algorithm of fatigue life calculation based on the critical plane approach. Int J Fatigue 2016;83:174–83. doi:10.1016/j.ijfatigue.2015.10.011.
 Karolczuk A, Skibicki D, Pejkowski Ł. Evaluation of the Fatemi-Socie damage parameter for the fatigue life calculation with application of the Chaboche plasticity model. Fatigue Fract Eng Mater Struct 2019;42:197–208. doi:10.1111/ffe.12895.
Keywords: fatigue life calculation, multiaxial fatigue, fatigue criteria, variable amplitude loading