Search Results

The work presented in this paper outlines the design and development of a compliant sprocket for balancer drives in an effort to reduce the noise levels related to chain-sprocket meshing. An experimental observation of a severe chain noise around a resonant engine speed with the Dual-Mass Flywheel (DMF) and standard build solid (fixed) balancer drive sprocket. Torsional oscillation at the crankshaft nose at full load is induced by uneven running of crankshaft with a dual-mass flywheel system. This results in an increase of the undesirable impact noise caused by the meshing between the chain-links and the engagement/disengagement regions of sprockets, and the clatter noise from the interaction between the vibrating chain and the guides. This paper evaluates and discusses the benefits that the compliant sprocket design provided. A multi-body dynamics system (MBS) model of the balancer chain drive has been developed, validated, and used to investigate the chain noise.

A numerical investigation of automobile sunroof buffeting on a prototype sport utility vehicle (SUV) is presented, including experimental validation. Buffeting is an unpleasant low frequency booming caused by flow-excited Helmholtz resonance of the interior cabin. Accurate prediction of this phenomenon requires accounting for the bi-directional coupling between the transient shear layer aerodynamics (vortex shedding) and the acoustic response of the cabin. Numerical simulations were performed using the PowerFLOW code, a CFD/CAA software package from Exa Corporation based on the Lattice Boltzmann Method (LBM). The well established LBM approach provides the time-dependent solution to the compressible Navier-Stokes equations, and directly captures both turbulent and acoustic pressure fluctuations over a wide range of scales given adequate computational grid resolution.

This work provides an effective engineering method of building a part-load engine simulation model from a wide-open throttle (WOT) engine model and available dynamometer data. It shows how to perform part-load engine simulation using optimizer for targeted manifold absolute air pressure (MAP) on a basic matrix of engine speed and MAP. Key combustion parameters were estimated to cover the entire part-load region based on affordable assumptions and limitations. Engine rubbing friction and pumping friction were combined to compare against the motoring torque. The emission data from GM dynamometer laboratory were used to compare against engine simulation results after attaching the RLT sensor to record emission data in the engine simulation model.

Development and integration of the cooling system for an automotive vehicle requires a balancing act between several performance and styling objectives. The cooling system needs to provide sufficient air for heat rejection with minimal impact on the aerodynamic drag, styling requirements and other criteria. An optimization of various design parameters is needed to develop a design to meet these objectives in a short amount of time. Increase in the accuracy of the numerical predictions and reduction in the turn-around time has made it possible for Computational Fluid Dynamics (CFD) to be used early in the design phase of the vehicle development. This study shows application of the CFD for robust design of the engine cooling system.

It is well known that lining reduction through wear affects contact pressure profile and noise generation. Due to high complexity in brake noise analysis, many factors were not included in previous analyses. In this paper, a new analysis process is performed by running brake “burnishing” cycles first, followed by noise analysis. In the paper, brake lining reduction due to wear is assumed to be proportional to the applied brake pressure with ABAQUS analysis. Brake pads go through four brake application-releasing cycles until the linings settle to a more stable pressure distribution. The resulting pressure profiles show lining cupping and high pressure spots shifting. The pressure distributions are compared to TekScan measurements. Brake noise analysis is then conducted with complex eigenvalue analysis steps; the resulting stability chart is better correlated to testing when the wear is comprehended.

In today’s highly competitive automotive environment people are always looking to develop processes that are fast, efficient, and effective. Moving testing from expensive prototype vehicles into the laboratory is an approach being implemented for many different vehicle subsystems. Specifically a process has recently been developed at General Motors that predicts exhaust back pressure performance for a running vehicle using laboratory testing and spreadsheet calculations. This paper describes the laboratory facility and procedure, the theory behind the calculations, and the correlation between vehicle test and laboratory based results. It also comments on the benefits of the process with respect to reduction in design iterations, quicker availability of results, and money savings.

Exhaust gas recirculation (EGR) is one of the key approaches applied to reduce emissions for an internal combustion engine. Recirculating a desired amount of EGR requires accurately estimating EGR mass flow. This can be calculated either from the gas flow equation of an orifice, or from the difference between charge air mass flow and fresh air mass flow. Both calculations need engine exhaust pressure as an input variable. This paper presents a method to estimate exhaust pressure for a variable geometry turbo charged diesel engine. The method is accurate and simple to fit production ECU application, therefore, saves cost of using a physical sensor.

This paper presents the development of a transmission closed loop pressure control system. The objective of this system is to improve transmission pressure control accuracy by employing closed-loop technology. The control system design includes both feed forward and feedback control. The feed forward control algorithm continuously learns solenoid P-I characteristics. The closed loop feedback control has a conventional PID control with multi-level gain selections for each control channel, as well as different operating points. To further improve the system performance, Robust Optimization is carried out to determine the optimal set of control parameters and controller hardware design factors. The optimized design is verified via an L18 experiment on spin dynamometer. The design is also tested on vehicle.

CFD method for the calculation of flow rate and air re-circulation at vehicle idle conditions is described. A small velocity is added to the ambient airflow in order to improve the numerical stability. The flow rate passing through the heat exchangers is insensitive to the ambient velocity, since the flow rate is largely determined by the fan operation. The air re-circulation, however, is quite sensitive to the ambient air velocity. The ambient velocity of U=-1m/s was found to be the more critical case, and is recommended for the air re-circulation analysis. The CFD analysis can also lead to design modifications improving the air re-circulation.

Superior fuel economy was achieved for a small-displacement spark-ignition direct-injection (SIDI) engine by optimizing the stratified combustion operation. The optimization was performed using computational analyses and subsequently testing the most promising configurations experimentally. The fuel economy savings are achieved by the use of a multihole injector with novel spray shape, which allows ultra-lean stratification for a wide range of part-load operating conditions without compromising smoke and hydrocarbon emissions. In this regard, a key challenge for wall-controlled SIDI engines is the minimization of wall wetting to prevent smoke, which may require advanced injection timings, while at the same time minimizing hydrocarbon emissions, which may require retarding injection and thereby preventing over-mixing of the fuel vapor.

The results of an experimental investigation of the flow in the near wake of a generic Sport Utility Vehicle (SUV) model are presented. The main goals of the study are to gain a better understanding of the external aerodynamics of SUVs, and to obtain a comprehensive experimental database that can be used as a benchmark to validate math-based CFD simulations for external aerodynamics. Data obtained in this study include the instantaneous and mean pressures, as well as mean velocities and turbulent quantities at various locations in the near wake. Mean pressure coefficients on the base of the SUV model vary from −0.23 to −0.1. The spectrum of the pressure coefficient fluctuation at the base of the model has a weak peak at a Strouhal number of 0.07. PIV measurements show a complex three-dimensional recirculation region behind the model of length approximately 1.2 times the width of the model.

Free expansion of straight tubes is the simplest test to evaluate tube properties for hydroforming applications and to provide basic understanding of the mechanics of tube hydroforming. A circular cylindrical tube is sealed at both ends and fluid, usually water, is pumped into the tube to increase its internal pressure to bulge and burst the tube. Previous numerical simulations of the free expansion tube test were limited to modeling the midsection of the tube under various assumptions of deformation path. The simulation results obtained deviated from the experimental results under all simulation conditions considered. A new model is developed in this paper in which the whole tube is simulated instead of considering only its mid-section. Judged by the pressure-expansion relations, the model accurately predicted free expansion hydroforming tests results.

Condenser, fan, radiator power train cooling module (CFRM) proposed by Delphi Automobile Systems was evaluated in the context of vehicle thermal system analysis. The results from the CFRM configuration were compared with those from the conventional condenser, radiator, and fan power train cooling module (CRFM). The analysis shows that for a typical passenger vehicle, the underhood temperature for the CFRM configuration is more than 10°C lower than its CRFM counterpart when the fan is operating at the same speed of 2500 rpm. This is due mainly to the higher mass flow rate impelled by the fan in the CFRM configuration. At the equal mass flow condition, both the CFRM and the CRFM configurations give similar underhood temperatures; but the fan in the CFRM configuration uses 19% less power, due mainly to the reduction in the fan speed needed to impel the same amount of mass flow rate.

The results of an experimental investigation of the flow over a pickup truck are presented. The main objectives of the study are to gain a better understanding of the flow structure in near wake region, and to obtain a detailed quantitative data set for validation of numerical simulations of this flow. Experiments were conducted at moderate Reynolds numbers (∼3×105) in the open return tunnel at the University of Michigan. Measured quantities include: the mean pressure on the symmetry plane, unsteady pressure in the bed, and Particle Image Velocimetry (PIV) measurements of the flow in the near wake. The unsteady pressure results show that pressure fluctuations in the forward section of the bed are small and increase significantly at the edge of the tailgate. Pressure fluctuation spectra at the edge of the tailgate show a spectral peak at a Strouhal number of 0.07 and large energy content at very low frequency.

The concept of condenser, fan, and radiator power train cooling module (CFRM) was further evaluated via three-dimensional computational fluid dynamics (CFD) studies in the present paper for vehicle at idle conditions. The analysis shows that the CFRM configuration was more prone to the problem of front-end air re-circulation as compared with the conventional condenser, radiator, and fan power train cooling module (CRFM). The enhanced front-end air re-circulation leads to a higher air temperature passing through the condenser. The higher air temperature, left unimproved, could render the vehicle air conditioning (AC) unit ineffective. The analysis also shows that the front-end air re-circulation can be reduced with an added sealing between the CFRM package and the front of the vehicle, making the CFRM package acceptable at the vehicle idle conditions.

Cylinder pressure data is so completely integral to the combustion system development process that ensuring measurements of the highest possible accuracy is of paramount importance. Three main areas of the pressure measurement and analysis process control the accuracy of measured cylinder pressure and its derived metrics: 1) Association of the pressure data to the engine's crankshaft position or cylinder volume 2) Pegging, or referencing, the pressure sensor output to a known, absolute pressure level 3) The raw, relative pressure output of the piezoelectric cylinder pressure sensor Certain cylinder pressure-based metrics, such as mean effective pressures (MEP) and heat release parameters, require knowledge of the cylinder volume associated with the sampled pressure data. Accurate determination of the cylinder volume is dependent on knowing the rotational position of the crankshaft.

Critical speed induced by imbalance forces is a well-known dynamic behavior of rotating shafts. Such problems are typically found in flexible shafts or rigid shafts with flexible supports when the frequency of rotation reaches the natural frequencies of the shaft. This simple critical speed problem is well understood and formulated in many engineering texts. However, not all critical speed phenomena are induced by imbalance. A perfectly balanced shaft with certain inertial properties also reaches a critical speed condition at a rotational speed that is not equal to the natural frequency of the shaft. Several variables of the dynamic system play a role on the critical speed condition, which is mainly induced by the unstable gyroscopic moment acting on the shaft. The unstable gyroscopic moment forces the shaft bearings to deflect causing precession about the undeflected geometric centerline of the shaft, but the rotation and precession speeds remain synchronized at low speeds.

This paper describes the design and functionality of an in-situ air entrainment measuring device for analysis of the air entrainment and air release properties of lubricating fluids. The apparatus allows for a variety of measurement techniques for the aeration and deaeration of the lubricating fluid at various temperatures, pressures, and agitation speeds. This test apparatus is patent pending because of its unique ability to allow for continuous, in-situ measurement of the fluid properties and the rates of change of these properties. Most other measurement techniques and apparatuses do not allow for uninterrupted measurement. This apparatus is also unique in that it is capable of detecting minor fluid density changes at a lower level and with more accuracy than all other current techniques or apparatuses.

A computer assisted development technique for Quick Plastic Forming parts [1] is described, based on the simulation program PAM-STAMP [2]. The technique allows thickness changes during forming to be accurately considered in the development process without physical trials. Process pressure cycles, which provide for maximal material formability, can be determined with a single simulation. The paper describes new program features, which reduce modeling effort and increase simulation accuracy. Various validation examples and industrial case studies are also presented, demonstrating current capabilities.

Noise due to the flow over an automobile A-pillar rain gutter in isolation was computed using a two step procedure. Initially the flow solution was obtained by solving the Reynolds Averaged Navier Stokes (RANS) equations. Acoustical Sources were extracted from the flow solution and propagated to the far-field using the Lighthill-Curle equation. Experiments were conducted to evaluate the computations. Compared results include steady pressures, time dependent pressures, and sound intensity levels. Computed results and experimental data were reduced in a similar way to ensure a one to one comparison. Computed results are in good agreement with the experimental values. A-weighted noise levels are predicted reasonably well.