CV joint made with muscle
Tom Shelley reports on a new design of constant velocity joint primarily designed for heavy industrial and commercial vehicle use
A new constant velocity joint uses cam shapes and cylindrical-surface contact to maximise the ability to transmit torque and achieve longevity. The centring mechanism is put on the outside of the construction rather than the inside so that it can be made more robust without restrictions caused by lack of space.
Initially targeted at heavy duty industrial use, interest so far has included applications as varied as commercial and other heavy vehicles, car steering columns, novel bicycles and miniature machine tools.
The joint is the latest development to come out of Transmission Systems, whose shuffling rack differential featured on the cover of Eureka’s January 1988 issue. Although the differential never achieved volume sales, it was subsequently offered as an option on Vauxhall Vectras and became standard equipment on Caterham sports cars. The distinguished designer of the differential and the present product (who wishes to remain anonymous), who has been working in mechanical transmissions for over 20 years, says he has spent seven years developing this latest idea.
His stated object was to come up with something both rugged and simple, which would achieve at least a B10 or L10 25,000 hour rating and which would still allow the smooth transmission of torque through high deflection angles (over 45 degrees is possible). This would require it to last much longer than a Rzeppa joint under similarly severe conditions, and to work better than the many alternatives offered in the patent literature.
A typical Rzeppa joint, as used in the vast majority of front wheel drive cars, has six balls. Each has one point of contact with the eccentric inner race or groove and one point of contact with the eccentric outer race or groove. In addition, the balls are in sliding-point contact rather than rolling contact so that they skid in order to move with the cage and satisfy the requirement that the cage must deflect by one half of the joint angle. The joint certainly solves the problem of constant velocity and is made in various versions in vast numbers. However, it only maintains a reasonable degree of reliability because cars spend most of their lives going in a straight line or near to it. Large deflections, encountered when turning tight corners or parking, are normally only associated with relatively low speeds and torques. Its weakness for heavy duty use is that it still depends on the skidding balls and point contacts.
The other solution commonly adopted, based on Cardan and double Cardan joints, is not truly constant velocity. One problem, with the Thompson Coupling, described in Eureka’s May issue, and others like it, according to the designer of this joint, is that there are too many joints for it to be commercially interesting, and the size of the joints is restricted by their having to be inside the mechanism. Other ideas, such as a development, patented in 1999, in which knob ends slide inside tubes are even less viable.
Simplistic cam design
This new joint, on the other hand, is very simple. Each of the input and output shafts end in claws, the peripheries of whose outer ends are formed into partly circular cam shapes with a 10mm eccentric. The claws pivot on stub axles projecting perpendicularly from the sides of cross shafts which are mounted in a cage. As the joint turns, with the shafts at an angle to each other, the cage reciprocates on the cross shafts. In order to do so, it moves through a distance of a few mm (sin 22.5 degrees x 10mm in the case of the prototype). The connection between the claws and the cross shafts is by circular internal bearing cups in the claws and the quite substantial stub axles. The cross shafts consist of a pair of elements with cross sectional profiles consisting of 135 degree sectors. The cross shafts are able to rotate relative to each other about their common axes through two cylindrical elements placed between them.
The whole device comprises eight parts, plus the two cylindrical elements which, in production, could be incorporated into the cross shafts. It presently exists in the form of a working demonstration prototype made of hard aluminium. The connection between the claws and the cages is by engagement between the external cam shapes of the claws and slippers on the insides of bridging elements. All moving parts therefore engage each other across substantial contact areas. In a production version, the slippers could be made of phosphor bronze, ceramic or any other low friction material with good compressive strength. Any point on the cross shaft in contact with the cage describes a cylindrical ellipse, which is good for lubrication. There are no stops and starts, unlike the track of a piston ring on the inside of a cylinder. The joints are virtually immune to the adverse effects of axial loads, unlike the Rzeppa joint.
The designer is convinced that, based on the results of experiments carried out during development, the joints in production units need to be equipped with rolling element bearings rather than plain bearings, to accommodate repeated off-axis loadings. However, he is prepared to admit that it may be possible to use plain bearings manufactured with some of the more modern polymer composite-based plain bearing materials, with considerable consequent cost savings, but he has not had a chance to test any of these. Whatever the bearings used, the joints are easy to assemble using automated or manual techniques. The prototype has only eight screws and requires use of only a single Allen key.
Main applications are seen in special purpose machines for automated production lines, heavy goods vehicles, military vehicles and vehicles for non-recreational off road use – construction plant and the like. One of the more interesting enquiries has been from a major automotive component manufacturer, engaged in the task of designing a steering column with four or six joints to get past a front wheel drive engine in an unusually restricted space.
The design has been patented. Transmission Systems is based in Jersey, and specialises in the selection, patenting, development and licensing of inventions. The company is currently discussing the technology with a number of companies and is interested in hearing from others. Working models can be provided to potential licensees as part of an option or licensing arrangement. FOR MORE INFORMATION: Transmission Systems Ltd., Chalet Abaco, Green Road, St. Clement, Jersey, JE2 6QA tel: 01534 769460 Fax: 01534 769466