Return to search

Predicting drag losses & oil flow characteristics in a wet clutch system using numerical analysis

This report presents the Master Thesis done by two students at the Mechanical Engineering program, Design Engineering at Luleå University of Technology. The thesis was performed at Luleå University of Technology in collaboration with BorgWarner PDS located in Landskrona. The objective for the thesis was to study the potential for predicting and optimizing drag losses and oil flow characteristics in a wet clutch system using numerical simulations in LS-DYNA. BorgWarner PDS manufactures and distributes the Haldex coupling, which is an electro-hydraulic coupling used in an active four-wheel drive system. BorgWarner PDS strives to make the Haldex coupling more efficient by minimizing energy losses in order to lower fuel consumption for the vehicle and a smaller impact on the environment. As testing on prototypes are costly and time consuming it is therefore desirable to use numerical simulations instead. To determine the potential for numerical simulations to predict and optimize both drag losses and oil flow characteristics a numerical model was set up using LS-DYNA. A test rig was designed in order to achieve a controlled testing environment in order to validate the numerical models, both qualitatively and quantitatively. The numerical model uses the built in Fluid Structure Interaction method in LS-DYNA, which is a combination of a structural model and an incompressible fluid model. The test rig measures torque on the clutch housing induced by the dragging of oil as the clutch drum rotates. The tests were performed using different rotational speeds, varying from 250-1500 rpm, and different clutch housing geometries. The results were gathered from an average of ten measurements of the drag torque. A high- speed camera was also used to capture the oil flow characteristics. The numerical simulations were done using the same parameters used for the tests i.e. different rotational speeds and clutch housing geometries. In addition, each simulation was performed using different mesh sizes; 0.5 mm, 1 mm and 2 mm in order to analyze mesh dependence. The numerical model was set to measure the drag torque on the clutch drum and was then compared to the experimental results in order to validate the numerical model quantitatively. The oil flow characteristics from the numerical simulation were compared to the high-speed movies to validate the numerical model qualitatively. The numerical model shows promising results when it comes to predicting drag torque and oil flow characteristics. The results also show that using a numerical model for optimization purposes are possible as the experimentally measured drag torque at different clutch house geometries corresponds to the numerically simulated drag torque. The conclusions from this Master Thesis is that numerical simulations do have a potential for predicting drag losses and oil flow characteristics in a wet clutch system. The numerical model created in this project can also be used to obtain valuable information about geometry changes when it comes to optimizing the wet clutch system. However, the numerical model is still at an early stage and can be seen as a second generation model where further work needs to be done in order to include more of the complex geometries of a wet clutch system have. Although LS-DYNA is state-of-the-art when it comes to these kinds of complex simulations it still needs to be developed further to handle such complex simulations.

Identiferoai:union.ndltd.org:UPSALLA1/oai:DiVA.org:ltu-60374
Date January 2016
CreatorsLiljegren, Marcus, Ljeskovica, Edin
PublisherLuleå tekniska universitet, Institutionen för teknikvetenskap och matematik, Luleå tekniska universitet, Institutionen för teknikvetenskap och matematik
Source SetsDiVA Archive at Upsalla University
LanguageEnglish
Detected LanguageEnglish
TypeStudent thesis, info:eu-repo/semantics/bachelorThesis, text
Formatapplication/pdf
Rightsinfo:eu-repo/semantics/openAccess

Page generated in 0.0024 seconds