Search Results

The purpose of this study is to propose an effective model to estimate the excitation force accompanied with stick-slip between shoe and disc, considering the strain distribution on contact surface of the shoe, and then to propose an effective concept to design the brake which reduced the brake squeal under practical use. In order to investigate the influence of configuration of the hole, three types of discs were prepared in which the size of holes was different. The SPL (Sound Pressure Level) and the frequency of squeal for three types of discs were measured when the brake squeal was observed at conditions of low sliding speed. The change of stability of the brake shoe passing on hole was analyzed by 2-D simplified brake system model.

A prediction method was proposed for crack initiation in one-piece type brake discs under extreme braking conditions. Braking tests under extreme loading conditions were conducted by using sample discs. The variations of internal temperatures and surface strains at several locations were measured by using thermocouples and strain gages. In order to obtain the S-N curve of the disc material, specimens with a hole in their center were used for the fatigue test in which an alternative strain was cyclically applied. The numbers of strain cycles, when a 0.3mm crack initiated from the hole, were analyzed by the Weibull plot. The maximum and minimum strains at the hole edge were estimated by calculation considering the temperature variation with respect to time as well as the strain induced by friction due to braking pads. The number of cycles corresponding to the strain amplitude range was estimated by the rain-flow method.

The mechanism of permanent deformation of pressure plates has been investigated under very severe conditions, which are used for automobile clutches. Sliding tests were performed, where an extreme overload was applied to the pressure plate. It is found from the test results that the residual stress in the circumferential direction varied from compression to tension after the sliding test near the sliding surface of the plate. FE (Finite element) calculations were also conducted taking both plastic deformation of the plate due to thermal expansion and geometrical non-linearity into account. The FE calculations agreed well with the experimental results. The permanent deformation mechanism for pressure plates was summarized as follows. Compressive yielding of the plate occurs at the heating stage near the sliding surface. The residual stress in tension appears in the sliding surface when the plate becomes cool, and it deflects the plate like a bowl.

The aim of these experiments is to determine the best way to obtain high mechanical properties for phenol resin and glass fibers based composites. Various ways of fabricating the material were studied, as well as its best composition. The conditions of drying, molding processes were optimized. From the most conventional method, using ethanol as a solvent to newer ones, including continuous ways of processing and the use of water instead of ethanol, a lot of possibilities exist to produce such a material. This paper explains the advantages and drawbacks of a whole range of manufacturing processes.

Tiny (hair-like) cracks initiated around small holes in the flange of one-piece brake discs were observed at an overloading condition. Thermally induced cyclic stress strongly affects the crack initiation in the brake discs. In order to show the crack initiation mechanism, the temperature distribution at the flange was firstly measured. The temperature distribution under overloading was analyzed by using the finite element method. Based on the experimental and calculated results, the crack initiation mechanism for one-piece brake discs at the very severe braking condition was explained. In addition, the effective methods are suggested for reducing the initiation of tiny cracks around the holes.

Creep tests under three-point bending loading were conducted at elevated temperatures for long glass fiber reinforced polypropylene in order to examine the effect of maleic anhydride modification for polypropylene matrix on the creep resistance. The high acid modification (1 wt%) significantly reduces the creep deflection at both low and high sustained loads. The residual tensile strength after 1000 hours creep also increased with an increase of maleic anhydride content. AE observation indicating an occurrence of debonding decreased with increasing acid density. The fractured surfaces show that the interfacial strength between glass fibers and matrix can be increased owing to an increase of maleic anhydride fraction.

Braking tests under overloading were carried out using large one-piece brake disks having eyebrow-shaped holes as decoration. When the number of braking cycles was more than five, permanent deflection of the disks was observed. When the number of braking cycles was less than six, no appreciable deflection occured. The experiment revealed that such deflection occurrs during the cooling process after the final braking cycle. The mechanism for this phenomenon is explained based on the deflection-time record in conjunction with the temperature distribution of disks and its variation with respect to time. The key for this phenomenon is yielding in tension at the bridges between holes. The deflection occurs due to elastic-plastic buckling caused by shrinkage of the flange. Numerical simulations were successfully conducted by using a general 3D FEM in consideration of geometrical and material non-linearities.

An advanced numerical model is proposed to analyze the power transmitting mechanisms of a CVT using a metal V-belt. By using the present model, forces acting on the belt are well estimated not only at steady states but also during transitional states where the speed ratio is changing. The numerical results show that blocks are in compression in both strands when the speed ratio is rapidly shifted. A complementary model is also developed to analyze the load distribution among bands which form the ring. The load distribution in the ring is governed by the difference in coefficients of friction among elements.

The effects of temperature, crack tip opening rate and rubber content on static fracture characteristics of CNBR (Cross-linked acryloNitrile Butadiene Rubber) modified epoxy adhesives were investigated under mode I loading. Loading-unloading tests were statically performed by using DCB (Double Cantilever Beam) specimens. The fracture toughness increased with increasing the rubber content. The fracture toughness of CNBR modified and unmodified epoxy adhesives was much influenced by temperature and crack tip opening rate. The surface topology of fractured surface was changed by temperature and type of adhesive.

The effects of inclusion of glass beads and rubber modification on the fatigue fracture characteristics of an epoxy adhesive were investigated. Hybrid effects were also investigated when not only the epoxy adhesive was rubber modified but also when glass beads were mixed simultaneously. Fatigue crack growth resistance was greatly increased due to glass beads, CTBN and CNBR modifications at the second stage of crack growth (da/dN=10-4-10-3 mm/cycle). However, the energy release rate at threshold for both CNBR and CTBN modified adhesives were lower than that for the unmodified adhesive. No significant hybrid effects were distinguishable.

A new parameter was proposed to evaluate the fatigue life of toothed belts. The parameter is the frictional work spent on the belt tooth surface for driving and driven pulleys. It can be estimated only with the 2D finite element model of the belts previously developed by the authors. As well as the frictional work, an alternative parameter, maximum tooth load (widely used in the literature) was also used to evaluate the fatigue life of toothed belts. In order to prove the effectiveness of the present parameter, fatigue tests were conducted using S8M belts at a constant power. The test results show that the maximum tooth load can explain the fatigue degradation of the toothed belts to some extent while the proposed parameter, the frictional work can evaluate the fatigue life of the belt due to wear of the belt facing fabric more appropriately than the maximum tooth load.

Some theories on the behaviour of CVT using metal pushing V-belts have been recently drawn. However, our previous experiments did not well prove their prediction. A numerical model which can calculate all block motions of the belt was developed in this paper. Using this model, some steady states of power transmitting of CVT were calculated and compared with the previous experimental results. Satisfactory agreements were obtained between two results in all ratio. This model is effective to estimate the CVT response at steady states.

It is very important to know the load distribution in pulleys to predict the life of toothed belts. In this study, a simple model consists of springs and friction elements has been developed for numerical analysis of the load distribution. A sample problem with steady pulley motion for a two pulley system was analyzed. The analytical result was compared with the experimental result. It was also compared with the numerical result by the alternative model using FEM. Relatively good agreements between them were obtained. A typical problem for the two pulley system subjected to fluctuating torque loadings was also analyzed by the present model. The calculated result shows a large difference in tooth load distribution between the steady state case and the unsteady state one.

The effects of CNBR (cross-linked acrylonitrile butadiene rubber) modification on the fracture characteristics of epoxy adhesives were investigated under Mode I static loading. Fracture tests were conducted by using DCB (double cantilever beam) specimens. Rubber content, adhesive thickness and loading rate were changed. The crack extension resistance (given by energy release rate) of the epoxy adhesives was much improved by CNBR modification. For the rubber modified epoxy adhesives, the crack extension resistance becomes high with an increase in adhesive thickness as well as loading rate. The reason why the CNBR modification improves the crack extension resistance was explained based on the surface observation and fractal dimension of the fractured surface.

Six forces act on the block of a metal pushing V-belt. Previously, we successfully measured these forces at steady states using devised blocks and a telemeter system. In this paper, six forces are measured using the same testing and measuring systems at transitional states where a speed ratio varies from low to high, or vice versa. The experimental results reveal that distributions of four forces at transitional states except normal and frictional forces between rings and blocks are different in shape from those at steady states.

A concept and a (nonlinear finite element) model of how to analyze load distribution of toothed belts having curvilinear tooth profiles for automotive engines at steady states was developed by utilizing a general nonlinear finite element program considering contact problems as well as geometrical nonlinear problems. A toothed belt in the model consists of circularly linked beam elements for endless tension members and two dimensional solid elements for a belt body. A curved pulley surface is supposed to be rigid. Interaction between surfaces of belt teeth and pulleys is considered as moving boundaries. A quite good agreement between experimental and computed results for frictional forces and tooth load confirms that the proposed model is presently the only one practical approach for analyzing load distribution of toothed belts which none of the existing theories can do. Some numerical simulations were performed by changing parameters such as belt pitch, dimensions of teeth and so on.

Block compression force and ring tension of a metal pushing V-belt type CVT have been experimentally measured at steady states. The peculiar transmitting mechanisms for this type of belts has also been outlined based on the experimental results in the previous works. In this paper, other forces simultaneously acting on a block at steady states were measured using newly developed devised blocks. These forces are frictional force between blocks and rings, normal force between blocks and pulleys, frictional forces between blocks and pulleys in radial and tangential directions. The transmitting mechanisms for the metal pushing V-belt type CVT were drawn in detail based on new experimental data. The following conclusions are emphasized in the present work. (1) A cohesive point where the block coheres with the ring exists in the pulley having a larger pitch radius at all conditions. This is not dependent on speed ratio and transmitting torque.

In this study, distributions of block compression force on the driving and driven pulleys were measured using a tiny load-cell inserted between two blocks and a telemeter system, under several constant speed ratios. Ring tension distributions were also measured using a specially devised block. From the experimental results, the following conclusions were drawn: (1) Block compression force distribution on the driving pulley is significantly different from that on the driven pulley. (2) Ring tension takes different value at each side of strings. It is considered that this phenomenon is caused by difference of saddle surface speed between two pulleys.

Transmitted torque, thrusts of driving and driven pulleys, and axial force between two pulleys were measured on a metal pushing V-belt type CVT. Thrust ratios between driving and driven pulleys at several different speed ratios were plotted with respect to torque ratio and compared with each other. It was found that the relation between thrust ratio and speed ratio was almost independent of rotational speed of the pulley and the maximum transmittable torque at a constant torque ratio. The thrust ratio is primarily a function of speed ratio. It also depends on torque ratio and coefficient of friction between blocks and a pulley. An empirical equation for pulley thrust balance was derived. The equation is expressed in an explicit form. It is so simple that it can be applicable for electronic control of CVT.

The purpose of this paper is to analyze the force distribution between pulleys and blocks of a newly developed CVT belt. Three components of the force (transmitting force, normal force and frictional force) were measured directly using a newly devised pulley. The experimental results reveal that the transmitting force distribution on the driving pulley is similar to that on the driven pulley as long as blocks do not slip while the distribution of the normal force component for both pulleys does not resemble each other as well as the distribution of friction force in the radial direction of the pulley. It is also found that no idle arc exists in the contact arc of both driven and driving pulleys even in the case that the transmitting torque is low. The experimental force distribution is compared with a theory based on the discrete spring model taking no consideration of slippage between the pulley and the blocks.