Search Results

The recent trends of increasing driveline torque and use of traction control devices call for increasingly higher durability capacity from driveline components. Bench and vehicle durability tests are often used to validate designs, but they are not cost-effective and take months to complete. Traditional finite element analysis (FEA) procedures have been used effectively in the re-design of driveline components to reduce stress, and occasionally, to predict fatigue life. But in the case of certain rotating components, such as the Axle Differential Case, where the component sees large stress/strain fluctuations within the course of one complete rotation, even under constant input torque, historical fatigue analysis (when conducted) yields very conservative results. The axle differential case tends to be one of the weakest links in the rear axle assembly. Therefore, there is a crucial need for analytical methods to more accurately predict fatigue life to reduce testing time and cost.

In this paper, responses from a vehicle's suspension, chassis and body, are used to demonstrate a methodology to optimize physical test results. It is well known that there is a variability effect due to an increase of wheel unsprung mass (due to loads measurement fixturing), tire pressure, speed, etc. This paper quantifies loading variability due to Wheel Force Transducer (WFT) unsprung mass by using a rainflow cycle counting domain. Also, presents a proving ground-to-test correlation study and the data reduction techniques that are used in road simulation test development to identify the most nominal road load measurement. Fundamental technical information and analytical methodology useful in overall vehicle durability testing are discussed. Durability testing in a laboratory is designed to correlate fatigue damage rig to road. A Proving Ground (PG) loading history is often acquired by running an instrumented vehicle over one or more PG events with various drivers.

The staircase fatigue testing method is a recognized method for determining the fatigue limit of powertrain components. The purpose of this paper is to improve upon existing standards by adding common practices that will ensure a higher degree of statistical accuracy in the data. This includes specifying appropriate sample sizes, stress increments and initial load conditions, as well as making suggestions for appropriate methods of analyzing the data. Two methods (Dixon and Mood method and probit analysis method) are selected and compared in terms of relative percent difference on four parameters (mean, standard deviation, B10 fatigue strength and B50 fatigue strength). The staircase data are obtained by simulations from normal and lognormal fatigue limit distributions.

During the initial vehicle design phase and as the first prototypes are built, extensive on-board instrumentation and data acquisition is required at the proving grounds (PG). The data is used for various types of testing and analysis. During this phase of development very few parts and assembly components are available for physical test. The objective is to develop a component test for the truck box. This test can be run without suspension parts during the early stages of the vehicle development. A further objective is to correlate the test to FEA models and actual Proving Ground full vehicle test results.

In a vehicle NVH development and refinement phase, it is necessary to understand the source of the noise and vibration from various powertrain and drivetrain mechanisms. The noise and vibration generated by a drivetrain in a vehicle is a complicate but significant source of physical mechanism, which might become important issues in early or later phase of the vehicle development. For the diagnostic purpose of the drivetrain, a rear-wheel drive (RWD) vehicle in early development phase has been used to measure the bending and torsional vibration of the drivetrain, as well as the vehicle interior noise simultaneously, while the vehicle is running up and down under quasi-steady state on a chassis dynamometer. The lower frequency resonances of torsional and bending vibrations from the drivetrain are correlated with the vehicle interior boom or overall loudness.

Historically, pump noise can be a contributor to customer dissatisfaction with automatic transmissions. In this paper, a Shainin experiment was conducted to identify all probable root causes for pump noise on a production RWD transmission. Sample transmissions were selected following subjective evaluations. Noise was objectively measured in the lab using a microphone and an accelerometer. The study was conducted following a systematic Shainin statistical engineering methodology, which included the following major steps: selection of the test measure using the isoplot technique, selection of Best of Best (BOB) and Worst of Worst (WOW) transmissions, assessment of assembly variation, component search, and pair-wise comparisons. The study successfully highlighted the key variables on the drive gear involute profile, which are now being tightly controlled for improved noise characteristics.

A Draw Bead Simulator (DBS), with modified draw beads, was employed in this study to understand the springback behavior of sheet metal subjected to multiple bending-unbending cycles. The investigations were carried out in both the rolling and the transverse rolling directions on four types of materials: Electro-Galvanized DQ steel, light and heavy gauge Hot-Dip Galvanealed High Strength Steels, and Aluminum alloy AL6111. The sheet geometries, thickness strains, pulling forces and clamping forces were measured and analyzed for the purpose of establishing a benchmark database for numerical predictions of springback. The results indicate that the springback curvature changes dramatically with the die holding force. The conditions at which the springback is minimized was observed and found to depend on the material properties and the sheet thickness. Analysis with an implicit FEM showed that the predicted and the experimental results are in very good agreement.

The springback behavior might be different due to different gap clearances between die and punch. A study using FEA simulation is done to investigate the die gap effect. A 3D brick element and an explicit-implicit method are employed to investigate a few simple problems. A draw form, a crash form with an upper pad and a flange form are investigated separately. Numisheet’93 2D draw bending springback problem is also investigated using an explicit dynamic code. Comparisons between springback simulation results on several different die gaps are illustrated. The Kirchhoff assumption of C° shell element and the Mindlin/Love assumption of thin shell element are also examined on different cases. A case study then is performed on a rail type panel. Conclusions and recommendations for future studies are summarized.

A springback study on three rail type panels is summarized. Numisheet'96 S-rail, A/S P rail II and a Daimlerchrysler rail are presented with experiment data and FEA simulation predictions. The details of the measurement on experiment samples and simulation models are illustrated. The comparison between the experiment data and the simulation results from four different softwares is made on separate cases. The correlation between experiment data and simulation results is analyzed.