The New simulation approaches session will take place on Thursday May 19th and will be chaired by Merten Stender of TU Hamburg and co-chaired by Thierry Chancelier of Hitachi Astemo.
Topics and speakers for the session include:
Brake dust simulations: A framework of CFD and DEM simulation methods
Kartik Upadhyay, Mercedes-Benz AG
The environment is an increasingly important concern today and no economy is unaffected.
This also applies to the friction braking system industry, because of the particles emitted during each braking process. As braking is a complex process, depending on speed, vehicle weight, level of deceleration, as well as environmental conditions such as temperature and humidity which influence the type and nature of the particles emitted, ranging from very fine to coarse. Depending on their nature, these emissions can end up in the environment in the draining water or in the air we breathe.
In the literature there are more and more studies about airborne particles; they all show that mainly emitted particles during braking have a distribution which varies with the braking conditions from nano to micro particles.
For these reasons, Mercedes Benz AG is looking for suitable methods to understand and analyse the braking emissions and countermeasures to recover the emitted particles. Simulations could support here to provide meaningful information regarding brake dust flow also it can enlarge the view including surrounding components and the full vehicle airflow in the wheel housing.
Companies specialized in numerical solutions are also challenged to identify which method would be the most suitable to study the trajectory evolution of these particles after their emission. In this context Altair Engineering proposes the use of the Discrete Element Method (DEM) used in the EDEM software. The DEM method is suitable for studying the behaviour of a very large number of particles interacting with each other and with their environment. The method can be easily coupled with other numerical solutions such as CFD. The numerical characteristics of the problem require a good strategy to solve the equations in an acceptable time.
In the presentation, a numerical model representing a vehicle wheel-housing and the braking system will be presented to visualize the evolution of the emission trajectories for certain braking scenario. The aim of these studies is to evaluate if the numerical models are solvable and if their results bring a better understanding of the problems encountered.
Eventually a simulation containing a countermeasure will be carried out to estimate the prediction of effectiveness of this measure.
Objective condensation of wheel-tire assemblies in finite element models for creep groan simulation
Tomas Bourdieu, EDAG Engineering GmbH
Within the spectrum of noises associated with brake operation, creep groan is positioned among current high-priority NVH-issues in the automotive industry. The phenomenon belongs to the family of self-excited vibrations and originates from a continuous alternation between stick and slip states of the disc-pads contact interaction. The generated vibrations travel to the chassis, exciting bulkier elements which amplify the noise reported by consumers. Although different types of experimental setups have been applied to reproduce and analyse the onset of creep groan, state of the art product development requires the integration of numerical tools to help improve versatility and reduce related costs.
On this matter, the influence of chassis components surrounding the brake system, e.g., the strut and the lower control arm, demand for broad system boundaries in the finite element (FE) environment and therefore increase the difficulty of achieving reliable results within practical time frames. Especially the complexity associated with the virtual representation of wheels and tires, which typically involves fine meshes and contact interaction between linear elastic and non-linear hyperelastic materials, either leads to the neglection or oversimplification of their real dynamical behaviour.
Consequently, this paper proposes an objective condensation methodology in order to produce systems of smaller scale while maintaining the dynamical characteristics being essential for the numerical emulation of the phenomenon. The technique relies on integrating arbitrary FE representations of the wheel-tire assembly, allowing for its implementation during the build-up process of creep groan models. Additionally, the approach enables the study of the tire's impact on the generated vibrations, currently missing in literature.
Depending on the frequency range of interest, the article proposes two different condensation methodologies. Firstly, for the emulation of the low-frequency creep groan signature (≈18 Hz), a replacement of the wheel-tire assembly with spring-damper elements is presented, involving the extraction of elastic properties through the deformation of the original model via an auxiliary simulation. Secondly, for the emulation of the high-frequency creep groan counterpart (≈ 80 Hz), an alternate FE substructure generation method is proposed which can retain relevant eigenmotions and their corresponding eigenfrequencies. The evaluation of both condensation methodologies relies on the calculation of key performance indicators from the results provided by steady-state dynamic analyses, yielding an objective metric for both, the impact on performance, as well as deviations from original acceleration signals.
Further simulation studies on hot judder issue of the brake disc with variable plate thickness
Hongtao Yan, Yantai Winhere Auto-Part Manufacturing Co., Ltd
Research and /or Engineering Questions/Objective: According to the statistics of the brake judder complaints from the market, it is commonly recognized and validated that DTV (disc thickness variation) is one of the main causes of brake judder. However, in specific applications, it was observed that some ventilated brake discs which met the overall DTV requirements still had hot judder issues. After inspection, it was discovered that those discs all had large plate thickness variation for both friction cheeks. A request was then raised to study the relationship between the variable friction plate thickness and the observed hot brake judder issue.
Methodology: In previous study (EB2020-FBR-015), Finite Element Analysis (FEA) method was employed to identify the effects of the variable friction plate thickness on the hot judder issue observed. In that study, only the disc, the pads, and the pistons were included in the analysis but the caliper was not considered. However, as one of the most important parts, the caliper will change the geometric boundary conditions, the loading and the contact at the interfaces between the pads and the disc. Therefore, in this study, a more detailed 3D model including caliper was established to simulate the same scenario as in the previous study. Same as in the previous study, comparison studies between discs with variable and uniform plate thickness were conducted.
Results: In the comparison studies, the contact pressure distribution and the overall temperature distribution on the friction surface were first examined. Meanwhile, the temperature history at some special locations, and temperature distributions in circumferential direction, radial direction and thickness direction were compared. Then the dynamic lateral runout, and the coning of the inboard and outboard friction surfaces and DTV were characterised. Finally, the braking torques were analysed. Same as in the previous study, the results from the 3D model with the caliper also showed a big difference between the variable plate thickness disc and the uniform plate thickness disc. However, the 3D model with the caliper and variable plate thickness disc exhibited an apparent brake torque variation, which was not identified in the previous study. It is believed that this brake torque variation is most likely the cause of the hot judder issue observed.
Limitations of this study: In this study, only FEA simulations were employed and the connection between the variable plate thickness and the hot brake judder will be further validated by the dynamometer tests.
What does the paper offer that is new in the field in comparison to other works: In this paper, it revealed the connection between the disc plate thickness variation and the hot judder issue by FEA simulations for the first time as we are aware of. The knowledge acquired will help us to fully understand and deal with the brake judder issues.
Conclusion: The studies in this paper show that the variable plate thickness does induce significant variations of the lateral run-out and the coning of the disc under an emergency stopping condition. However, because the overall disc thickness is uniform, the dynamic DTV is very small and can be neglected. Based on the results obtained from the 3D model with the caliper, the brake torque variation is obvious when the plate thickness variation is large. Therefore, it could be concluded that the variable plate thickness could induce the brake torque vibration through the lateral run-out and coning variation, which are most likely the main causes of the hot judder issue observed. In future study, dynamometer tests will be conducted, in which specially designed brake discs with variable friction plate thickness will be employed. If it is verified, corresponding actions to minimise the variable plate thickness of the brake disc will be taken in manufacturing process.
Vibro-acoustic analysis of a parking brake system with plastic gears
Zaid Boussattine, Hitachi Astemo
Automated Parking Brake (APB) are common systems on recent vehicles. They must fulfil NVH targets defined by car manufacturers. In this context, the dynamic behaviour of such a system should be well known to optimize its design and avoid noisy operation. For an APB system based on a geared transmission, the main source of excitation is generated by the meshing process. It is usually assumed that static transmission error (STE) and gear mesh stiffness fluctuations are responsible for noise radiated by the housing. They generate dynamic mesh forces which are transmitted to the housing through wheel bodies, shafts, and bearings. Housing vibratory state is directly related to the noise radiated from the gearbox (whining noise).
Although the prediction of whining noise is rather well mastered for steel gears, the case of plastic gears brings some news challenges. The important flexibility of wheel bodies and the difficulty to control the micro-geometry makes harder the estimation of the resulting whining noise.
This work presents an efficient method to compute the whining noise of a real industrial system with both plastic and metal gears. A complete measurements campaign has been carried out to attest the efficiency of the computation process in order to predict the dynamic behaviour of the system.
First, the excitation is computed in a deterministic way. Then, a complete Finite Element Model (FEM) is built and tuned to represent the modal behaviour of the system. The dynamic response is then computed using an efficient computational scheme (Spectral Iterative Method, developed by the École Centrale de Lyon). The procedure is based on a modal approach developed in the frequency domain, particularly efficient to analyse systems having many degrees of freedom and subjected to parametrical excitation. The computation output result has shown a good agreement with accelerometer measurements. Finally, a vibro-acoustic model is used to simulate acoustic radiation emitted by the APB gearbox. The computed acoustic response has been compared to microphone measurements.
Operational gear measurement and calculation model are analysed in frequency domain during idle phase. This allows us to correlate, and an algorithm is proposed to deal with the acoustic power and ERP. the relationship between those two calculations shows us a good behaviour and level from low and medium frequency ranges. Next step will be to introduce the DC motor in our model and will allow us to complete the modal behaviour from our system (mechanical and electrical).
Aluminium brake discs: casting quality assurance by computer simulation
Samuel Awe, AUTOMOTIVE COMPONENTS FLOBY AB
As a result of the increasing demand to minimize the weight of automotive vehicles as well as reducing particle emissions, Al-matrix composite (Al-MC) is gaining interest as a potential material for manufacturing automobile brake discs because of its good corrosion property, high thermal conductivity and higher wear resistance when compared to the traditional grey cast-iron brake disc material. However, to mass-produce Al-MC brake discs, an efficient and robust manufacturing process is required. The squeeze casting technique is considered as one of the economical casting processes by which Al-MC material can be shaped into readily usable components. This process is attractive because squeezed cast products exhibit better mechanical properties due to the presence of fewer common defects such as porosity and shrinkage cavities, and the segregation of the reinforcing material is eliminated.
To efficiently use squeeze casting industrially, its processing parameters such as squeeze pressure, pouring temperature, mould die temperature and melt flow speed must be optimized to ensure sound castings. When a newly designed component is to be produced by squeeze casting, the casting parameters must be optimized for such a component. This optimization is conventionally performed through experimental trials which consume a lot of material resources and time. Nowadays, the application of computer simulation to model casting processes has enabled foundry engineers to shorten casting development time, maximize material usage, establish a robust process window by optimizing casting process parameters, reduce quality costs by preventing metal casting defects resulting from turbulence and other inclusions, predict and prevent defects in castings, and hence ensures quality castings and reduces product manufacturing costs.
This paper discusses how a casting simulation program could help to ensure a high-quality casting of Al-MC brake discs by investigating the influence of squeeze casting parameters on the possibility to minimize casting defects through parameters optimization. The 3D mechanical CAD software (Inventor LT) program was used to construct the 3D model of the brake disc and the forming tool, and the step files were imported into the NOVACAST software to simulate the casting and solidification processes. Further, two-parameter settings from the simulation results were implemented experimentally to produce squeeze cast Al-MC brake discs. The cast discs were examined and compared with the simulation results to check the adequacy of the simulation in predicting the quality of the brake disc.