Search Results

The objective of this research is to create a new brake system with fewer mechanical parts, higher performance, greater flexibility for adaptation to new functions, and lower cost. A simple/series electro-hydrostatic brake system is investigated as an inexpensive, reliable, and redundant integrated brake system that can include the functions; Boost, ABS, TCS, VDC, etc. Production issues are considered. The required motor power is the most critical and is estimated by simulation based on data from experiments. To reduce this power a flow boost self-energizing mechanism with computer control is explored, and it is found that the effect is significant. Robustness of the control for pad friction fluctuation is also analyzed, and the limitation is estimated. The result of analysis shows that a competitive commercial product can be developed.

As an aid in conceptual development of sophisticated integrated control strategies for torque, brakes, steering, and suspensions in ground vehicles, reasonable physical system models are necessary. These models must be sufficiently complete to include the important dynamics, but not so complex that insight is obscured. This paper develops several vehicle models of varying complexity that can aid in the design synthesis of controllers. The potential uses of these models is described.

“Beaming” in trucks is due to a bending vibration of the frame resulting in a fore-aft motion at the driver location. It has been found that beaming motions are relatively unaffected by the suspension type or components, and, in particular, by the shock absorbers. Since beaming is due to a distributed dynamic effect, it is believed that modal nodes exist near the suspension attachments to the frame thus rendering the suspension ineffectual at controlling beaming. Another explanation of beaming is that an overall system dynamic mode exists such that at a specific frequency the relative velocity across the suspension is near zero thus producing little damping for that mode. This concept is developed here through the use of some simple models. The sensitivity of beaming to system configuration is then demonstrated through use of a rather complete overall tractor/trailer model.

A torque control policy for four-wheel drive road-going vehicles is developed, based on the use of a compact variable ratio unit (VRU) located at each wheel. Since the appropriate hardware is not yet available, a computer model is developed to examine what gear ratio range and frequency response might be required of the hardware to allow for improved performance and stability over current four-wheel drive systems. A comparison is then made to a front-wheel drive (FWD), rear-wheel drive (RWD) and four-wheel drive (4WD) to determine the effectiveness of the derived control policy.

The ultimate objective of this work is to develop an optimized control strategy for all wheel, independent torque control for road-going vehicles, and to use this control primarily for directional stability. Thus far only front/rear torque control has been investigated although a complete model suitable for All Wheel Drive development has been formulated. A control policy has been postulated for front/rear torque control that performs better than front wheel drive, rear wheel drive, or 4 wheel drive, under some conditions. Based on the results obtained so far, an all wheel drive system with independent wheel torque control is anticipated to yield greatly improved handling under all conditions.

Bond graphs are a concise pictorial representation of the interactive dynamics of all types of energetic, engineering systems. Their virtue lies in the relatively few symbols required to model dynamic systems with many coupled, interacting energy domains, in the ease with which physical state variables are identified, and in the manner in which equation formulation is dictated. The procedure for deriving state equations is so unified and repetitive, that a digital computer can automatically derive the state equations directly from the bond graph structure. This paper develops some of these bond graph virtues for application to vehicle dynamics with emphasis on 3-D rigid body sprung mass motions, suspension dynamics, and steering dynamics. The methods of including control in the system model are presented, and the development of control policies for nonlinear vehicle systems is discussed.

A three-wheel car concept is described, in which steering of the front wheels is limited to small angles for packaging reasons, so rear-wheel-steering is needed. Arguments for minimizing weight and aerodynamic drag suggest that this particular configuration is the optimum two-passenger vehicle for maximizing fuel economy. The limitation on steering of the front wheels requires that rear-wheel-steering is dominant in tight turns at low speed. The feasibility of this proposed steering scheme is supported by experimental evidence, as well as by the literature on driver behavior and all-wheel-steering cars. Preliminary tests with a prototype vehicle indicate that rear-wheel-steering alone is sufficient below 10 meters per second (22 mph), which would permit a front steer angle limit of .1 radians (6 degrees).

A previously developed mathematical model of an operating two-stroke engine was modified to predict fresh air/fuel mixture loss through the exhaust port. At some operating conditions, losses as high as 30% of the newly scavenged mixture were predicted. A servovalve is proposed for placement downstream of the exhaust port. By sensing the temperature of the exhaust flow and closing the valve when this temperature falls below some set temperature, the loss in fresh mixture was predicted to be reduced to virtually zero. This was accompanied by an increase in power.

The characteristics of large amplitude wave propagation in variable area ducts were investigated in order to gain a more thorough understanding of exhaust systems for power tuned two-stroke engines. An apparatus to simulate the power stroke of a two-stroke motorcycle engine was developed, using a stock Yamaha 360 MX barrel, cylinder head, piston and ring. In this way, a single pulse of pressurized air could be introduced into virtually any duct geometry. An ocilloscope trace of the subsequent exhaust pulse and duct interaction was photographed. The resulting pressure-time histories were used to experimentally evaluate the effects of several duct terminations on the shape and timing of the reflected wave. The investigation resulted in quantifying the effects of an orifice plate and converging cone termination and in determining the best termination for a Yamaha 360 MX engine.