The Design and Development Story, as told by Alex Moulton, the Originator of this Hydraulically Interconnected Rubber Springing System Used on the  Morris I 100 Car

Each of the rear Hydrolastic units has its axis approximately 15 deg to the horizontal and the piston of the diaphragm-type displacer is deflected by a short arm projecting from beneath the trailing link of the rear suspension. Hydraulic fluid presssurised by the weight of the car, is used as the interconnecting medium between this unit and that at the front wheel on the same side.

In view of the considerable success of the ADO 15 vehicles, which have been introduced in the last three years by the British Motor Corporation, it has been obvious that a range of larger cars of similar layout, designed by the same team, headed by Alec Issigonis, would follow this now-familiar arrangement – a transversely installed in-line engine built integrally with the transmission unit, front-wheel drive and a long wheelbase; independent suspension, with rubber springs and small diameter wheels offers considerable benefits in terms of road handling performance, fuel economy and the ratio of the interior to overall volume of the car. A factor that has contributed to both the excellent road-holding and the high efficiency in terms of space utilization has been the use of Moulton cone-type, rubber springs for the road wheels-they occupy a relatively small space by comparison with that required for orthodox steel springs, and their progressive-rate characteristics can be brought closer to the ideal than could those of the conventional springs.

By virtue of the installation of the Hydrolastic system, the ride of the Morris 1100 is virtually free from pitching, and its roll-stiffness is high. This car is shown rounding a corner at speed over rough ground.

The larger car now introduced, the ADO 16, is to be known to the general public as the Morris 1100. From the technical view point its most interesting feature is the ingenious system hydraulically  interconnected rubber springs – also Moulton design-which are hardly less compact than those of the ADO 15. The standards of road-holding and ride afforded by this layout have already been the subject of wide acclaim. The layout, termed the Hydrolastic supension system, was developed through the collaboration of three organizations – British Motor Corporation,
Moulton Developments Ltd, and Dunlop Rubber Co. Ltd.
In each car there are two separate, identical systems, one for the left- and one for the right-hand pair of wheels: there are no levelling mechanisms, pumps or accumulators. and no need for adjusting headlamp angle to compensate for vehicle loading; the two systems are pressurized by the weight of the car. The cost is said to be comparable to that of an orthodox springing and damping system on a car with all-independent suspension.
Judged in absolute terms, Hydrolastic springing is an innovation of the very highest technical merit. In overall field of vehicle springing, its introduction on a small family car in volume production is an especially noteworthy aspect and great credit is due to the British Motor Corporation for sponsoring the work. Experience with this new vehicle has already demonstrated convincingly that a highly efficient springing and damping system does not necessarily carry penalties in respect of bulk, complexity, weight, reliability or cost.

The fact that a totally original concept has been brought from the patent stage to production in six years is also creditable. For, although the characteristics of rubber as a springing medium are now well known, a completely new technology in respect of hydraulically deflected rubber elements has had to be developed.  Furthermore, economy of production had to be borne in mind throughout every stage of the development programme, because this high quality springing system was planned, from its inception, for a low-cost  vehicle  destined  for  very  high  rates  of production.         

Above: In addition to the anti-roll bar be­ tween the trailing links of the rear suspension, there are small – diameter tor­sion bars to limit the static pitch attitude of the vehicle. Below: The front Hydrolastic unit is identical with that at the rear, but operates at a different leverage.

The compactness of the unit, which weighs 9lb, is demonstrated here. In service, the flexible hose allows for the deflections of the spring and for slight movements of the rubber-mounted sub-frame  relative to the hull. Each unit is held firmly in place in its sub-frame by the pre-load on the spring. The damper is well protected.

Since a prerequisite for obtaining the maximum space for the occupants is a long wheelbase relative to overall length, some form of interconnection of front and rear springs was considered essential for maintaining an acceptably high standard of ride.  The need for adopting interconnection to reduce pitching frequency has, of course, been appreciated for many years and has been acknowledged in practice by the adaption of this principle, but with mechanical inter­-connection in the Citroen 2CV and AMI 6 models.  An article giving the theory of interconnected suspension was published in the January 1957 issue of Automobile Engineer. Orthodox springing layouts have the limitation that the front and rear suspension frequencies have to differ to minimize pitching, which would otherwise occur owing to coupling of the pitching and bouncing modes of vibration. As a result, the rear-seat ride is frequently inferior to that at the front. On the ADO 16, however, by virtue of the interconnection of the front and rear springs, pitching is minimized and the motion is harmonic and jerk-free. In practice the permissible lowness of the frequency is limited by the need for avoiding excessive variations in the static attitude of the vehicle when the fore and aft distribution of its load is changed. The small Citroen car, for instance, has a low pitch frequency of the order of 50 c/min­ and since its attitude varies widely with load distribution, provision is made for adjusting the angle of the headlamp beams relative to the road.

Readers may be interested to know that a prototype car with an interconnected springing system was working as long ago as 1935. This was designed by Issigonis in collaboration with J. N. Morris, now Chief Engineer of S.U. Carburettor Co. Ltd.  The car had double-wishbone suspension and  torsion-bar springing at front and rear; on each side of the car, the pair of torsion bars extended forward or rearward to a differential gear mounted on the chassis midway between the wheels; the end of each bar was splined onto one of the bevel gears.

Description of the Hydrolastic system

The spring element for each wheel comprises tapered inner and outer sleeves between which is bonded an annular, natural rubber spring of a highly resilient mix. Thus, when deflected axially, the spring is subjected to both compression and shear forces, and, like the cone springs used on the ADO 15 car, has a progressive-rate characteristic. Imme­diately beneath each spring, and forming part of a common assembly, is a hydraulic piston, diaphragm and cylinder unit, termed the displacer unit.

Both sleeves are steel pressings.  The inner one is of cupped form, and the outer one is part of a fabricated canister serving as the hydraulic cylinder.  The  upper end of this canister is held, by the spring pre-load, firmly against the suspension sub-frame. Hydraulic fluid is contained between the lower face of the rubber spring and a diaphragm that closes the bottom end of the canister. Interposed between this diaphragm and the spring is the damper assembly, comprising a conical cup-shape pressing, termed the port-plate, on the centre of which are mounted the rubber flap type damper-valves. There is also a permanent bleed hole in this port-plate.

The outer edge of the diaphragm is, of course, clamped around the lower end of the canister, while a conical piston assembly, connected to the suspension arm, seats on the centre of the lower face. As the suspension moves up and down, the motion of the piston deflects the diaphragm, which rolls within a steel skirt pressing of approximately conical form, the upper end of which is also secured to the lower end of the canister. As can be seen in the accom­panying illustrations, the taper of the skirt pressing is in the opposite sense to that of the piston. Nylon-reinforced moulded rubber is employed for the diaphragm. Further reinforcement is afforded by two con­ centric beads of steel wire, to which the nylon cords are anchored. One of these beads reinforces the periphery and the other seats around the crown of the piston.  A thin liner of butyl rubber is fitted over the upper face of the diaphragm,  to ensure impermeability.

There is one system on each side of the car. When both wheels are disturbed to the same degree, as in pure bounce -rarely experienced- or in situ no interflow of fluid occurs. Consequently, the progressive-rate springs are deflected together and the ride is firm. If only one wheel is disturbed fluid passes through the conduit to move the other wheel in the opposite sense. This reduces pitching and also, because the springs only partially deflected, produces a softer ride. The throttling effect of the conduit is beneficial to the quality of the ride at high road speeds.

There are two separate seals. One is between the periphery of the diaphragm and that of the port-plate, while the other is between the port-plate and the lower end of the canister. During assembly, the first is effected by clamping the peripheries of the butyl lining and the diaphragm between the flanged upper end of the skirt and the  rolled-over  periphery  of  the  port-plate. Then  the second seal is made by pressing the rubber lined lower end of the canister around the rolled-over portion of the port- plate.  This bonded lining is an extension of  the spring material within the canister.  In this way, a remarkably sturdy unit, weighing approximately 9 lb, is formed. The four damper ports punched in the port-plate are oval. They are identical and are equally spaced on a common pitch circle : one diametrically opposed pair works during bounce and the other during rebound strokes. Each of these two functions is controlled by a separate rubber flap component which, as viewed in plan, is of waisted form. There is one on the upper face of the pressing, this controls bounce. The other disposed at 90 degrees to the first mentioned one-is on the underside and controls rebound. Each flap component has a bonded-in ferrule and a single rivet is passed through it and the centre of the port-plate to secure two small steel pressings, straddling each flap. The ends of each of these two saddle pieces extend into the punched holes of the other damper valve to afford angular location. Extending laterally the central portion of each retainer saddle are two tongs which during assembly are bent down, by means of special equipment, on to the projecting ends of the flap components, to set the blow-off pressure of the valve. Claimed that rubber flap-valve dampers have longer working lives and are more reliable than orthodox metal ones, that the mechanized setting process affords a much closer control over damping characteristics than has hitherto been possible. In  addition to these pressure-controlled va lve there is, as was already mentioned, a single fixed orifice that functions as a bleed control; this hole is situated on the tapered face of the port-plate.

On each side of the vehicle, Bundy tubing, ½ in  diam interconnects the two Hydrolastic units.  It is acccomodated in the floor-tunnel along which the exhaust pipe brake line are taken. To allow for movements of both the spring and the suspension sub-frame, the ends of conduita are connected to the spring units by a hose approximately 1 ft long. One end of each hose is fitted over a serrated extension-piece butt-welded to the inner pressing of the spring and, at the other end, the connection is a standard threaded union.

There is a Schrader charging valve in the pipeline: it is close to the front Hydrolastic unit and accessible through the engine compartment. Through this valve, the system charged with hydraulic fluid-a mixture of 49% demineralized water, 49% alcohol, 1% triethanolamine phosphate and 1% sodium mercaptobenzthiazole.  This fluid is, of course, an anti-freeze solution of constant viscosity, containing a rust inhibitor and agent that is added to make the fluid distasteful – a requirement. Before the system is charged with fluid a pressure of about 205 lb/in2, the standardized pressure for an unladen vehicle- an 80 per cent vacuum condition is induced by an exhauster pump. During service, pressure may rise to 450 lb/in2 at full bump, and may drop to as low as 70 lb/in2 on conditions of full rebound.

Operation of the system

As was previously stated, the function of interconnection between front and rear springs is to reduce the pitching frequency of a vehicle equipped with road springs of a bounce frequency affording good handling characteristics. For the Morris llOO car, the frequencies are 90-92 c/min bounce and 65 c/min pitch. A reduction in pitch frequency can be achieved by shortening the wheelbase, of course, but this, as already mentioned, is undesirable.

Clearly, the most frequent disturbances to a car travelling in a forward direction are those due to road irregularities that are contacted first by the front wheel and then by the rear wheel on the same side of the vehicle. In a conven­tionally sprung car, each of these movements causes angular acceleration in the pitch mode.  When a front displacer unit of the Hydrolastic system is deflected by a movement of the road wheel, fluid is impelled through the conduit to the rear displacer, which raises the rear end of the car and thus reduces the angular acceleration.  As a result, pitch is virtually eliminated and the car maintains a substantially level attitude. Because of the interconnection of the front and rear springs, the dynamic loads on any one wheel are shared by both, the actual proportions distributed between the two being determined by the throttling action of the conduit. Full wheel movement, therefore, is not normally accompanied by full spring movement, and a relatively soft ride is produced. At higher road speeds the resistance of the pipe to fluid flow increases, as the cube of the fluid velocity, and this steadies the car in the pitch mode. The effective diameter of the conduit is critical. On the Morris 1100, the conduits are of larger bore than is necessary, but each incorporates an orifice plate which determines the ride characteristics.

If both displacer units are deflected simultaneously-as in pure bounce or roll-no interflow occurs, so the pressure of the fluid is conveyed entirely to the springs, which deflect simultaneously, giving an increased suspension rate. Since a condition of pure bounce is rarely experienced, the full effect of two springs acting together is normally felt only during roll, when the outer wheels deflect in concert. The figure quoted for the roll stiffness of the ADO 16 is 6,000 lb-in/ deg, which is remarkably high for a saloon car.

The siting of the flap-valves adjacent to the diaphragms ensures that all wheel movements are damped. Small displacements of fluid are throttled only by the constant­ diameter bleed hole. During more arduous conditions of operation, the main dampers take over; these are progressive in action, and an adequate level of damping is afforded for all types of terrain. Because of the volume of fluid avail­able for dissipating heat-3.84 pints per system, or 7.68 pints per car-damper fade is not encountered.

Mention has been made of the taper configurations of the piston and the skirt against which the diaphragm rolls. This feature, in addition to being of value in respect of supporting a large area of the diaphragm, helps to stabilize the vehicle in the pitch mode and also contributes to the non-linear springing characteristics. As the piston deflects, the effec­tive area of the diaphragm is increased. In pitch, the “‘displaced fluid is conveyed to the other unit-in which the diaphragm’s effective area progressively decreases-and the resistance to pitching therefore rises; in bounce, however, the rate of increase of pressure in the fluid falls, so the apparent spring rate rises.

To minimize changes in the static pitch-attitude, a small­ diameter torsion bar is incorporated between each rear trailing arm and the chassis frame.  This avoids both the unpleasant appearance of a car with an excessive nose-down or tail-down attitude, and the need of adjusting the head­-lamp setting.  Another anti-roll bar is fitted between the rear trailing arms, to counteract the inherent understeering characteristic of the vehicle, which arises from the fact that 62 per cent of the unladen weight is carried by the front wheels. Since – to rationalize the production of primary equipment and spares – all four units on a vehicle are identical, different leverage ratios have been adopted at front and rear. That at the front is 3-95: 1, and that at the rear 4-4: 1; wheel rates are therefore greater at the front than at the rear, and the anti-roll bar reduces the resultant excessive understeer to an acceptable level.

The ends of the anti-roll bar are bent through 90 deg and drilled, so that they can be bolted in the usual manner to the inner faces of the trailing arms. Sandwiched between each drilled portion and the trailing arm is a forward­ extending lever with a square hole near its end, to take the squared end of a pitch-control bar, which of course is installed laterally. The inner ends of the pitch bars are bent forward through 90 deg, and register in steel bushes carried in the cross-member of the sub-frame.

Three features therefore control the pitch of the vehicle. They are the taper configuration of the displacer, the geometry of the suspension linkage, and the pitch bars. Bounce is controlled by these three features and the road springs, and roll is controlled by all four, plus the anti-roll bar. An inherently good aspect of the vehicle design is that the pitching moment due to braking is counterbalanced by the torque reaction on the rear trailing arms, with the result that there is very little change of pitch attitude during braking. The manufacturers contend that, despite the absence of provision for control of ride or levelling, the qualities of ride and road-holding, even at their present level, are sufficiently in advance of current standards to justify the decision not to incorporate such features.

Development history

The first of many Moulton patents dealing with hydrauli­cally interconnected rubber springs was taken out in 1955; in the specification, figures for bounce and pitch frequencies 85 and 60 c/min respectively were quoted as being achieved on a rig arranged to simulate the inertia of a vehicle with hydraulically interconnected rubber springs. It was for the purpose of developing this and subsequent patents, in exclusive application to B.M.C. cars, that the firm of Moulton Developments Ltd. an associate firm of B.M.C. was incorporated.

The development history of the Hydrolastic system is remarkable for the close and harmonious collaboration between the three interested companies, Moulton Develop­ments Ltd, the designers; British Motor Corporation, the users; and Dunlop Rubber Co. Ltd, the manufacturers. Throughout the development period, weekly visits were made by Alex Moulton to both the Dunlop factory, where the meetings were under the chairmanship of Mr. H. Trevaskis, the Technical Director, and also to the Suspension department at the Cowley works of B.M.C. The work of the Suspension department was entirely devoted to damping requirements and to problems connected with the applica­tion of the system to vehicles. Over 250,000 miles of road­ running were carried out on Morris Minors converted to independent rear suspension. Concurrently, rig-testing took place at the Moulton establishment at Bradford-on-Avon. In this way, the technical responsibility was vested in Moulton Developments Ltd, but Alec Issigonis, having overall technical control and approval, was kept informed of all changes and decisions. To keep the evaluation funcion separate from the manufacturing function, Moulton Developments Ltd. were responsible for the testing and evaluation of all components. During the whole of the development period, the rig shop at Bradford-on-Avon ran at a very high factor of utilization: the figure for one particular year including nights, week-ends and holidays-was 82 per cent.

From the outset, there were three basic aims. First, the system was to be completely sealed, so that there should be no maintenance problems of the type associated with pumps. Secondly, for simplicity, rubber was to be used as a springing medium in bounce.  Thirdly, the dampening mechanism was to be integral with the hydraulic inter­-connection. Most of the stages in the following account of the development programme are depicted in the accompanying illustrations.

Stage one, July 1956.  One of the first experiments to take place after the incorporation of Moulton Developments Ltd involved a layout in which a 9 in diameter annular rubber spring- in which the rubber was stressed in combined compression and shear, was coupled hydraulically to a pair of 2½ in diameter displacer units of the rolling diaphragm type; there was one displacer at each wheel, of course, and the springs for the two systems were installed beneath the front seats in the car. The stroke of the pistons was virtually the same as wheel movement, and static hydraulic pressure was 100 lb/in2. Rig tests showed that the diaphragms had a fatigue life of only 0.25 million cycles; the target figure was 1 million cycles.  Failures of these components-which were of only single ply construction, to accommodate stretching during rolling-were caused by inadequate burst strength, pinching due to inadequate radial clearance, and abrasion due to the high-velocity rolling action; hence, displacers of long stroke and small diameter were abandoned.  In this layout, a single damper valve, serving two wheels, was con­tained within each spring casing, and was of the orthodox spring-loaded metal type.

Stage  two, December  1956.  Operation of the next displacer, of 7½ in diameter was based on flexing rather than rolling.  The  diaphragm was reinforced by six plies of nylon tyre cord, and withstood 2 million cycles on the rig. Lever operation of the displacer was adopted and retained thereafter to shorten the stroke of the unit. Despite its acceptable fatigue life, this diaphragm was found to be too cumbersome and stiff for production purposes. It incor­porated two beads, however, the inner attached to the piston by a riveted-on collar, the outer clamped in an annular fold of the casing by a pressing operation. The sealing of the inner bead was not satisfactory, and subsequent diaphragms were therefore made continuous across the centre, where in each case support was afforded by the head of the piston.

A development of this displacer was tried, in which the diaphragm was 6 in diameter and reinforced by four plies of flat nylon cord. This was subject to puncturing and ply separation, however, and had a fatigue life of only 0-6 million cycles. A modified reinforcement was therefore developed, in which three plies were wrapped into a cylinder, with their cords at a small angle to the axis. One end of the cylinder was expanded by a balloon, until the cords were almost radial, and the diaphragm was moulded in this condition. This diaphragm had a life of 2 million cycles. Stage three, March 1957. With stronger diaphragms, a change was made to higher working pressures, and con­sequently to displacers of smaller diameter. It is interesting to reflect that the pursuit of higher working pressure was a central theme of the Moulton development work, with the result that, today, production diaphragms are subjected to unusually high pressures with complete safety. To give burst protection to the diaphragm, a short, tapered, skirt was incorporated in the displacer housing, extending down­ wards from the annular groove in which the bead was clamped; the piston was given a conical form.

Concurrently, in an effort to make reductions in the space needed for the spring, a change was made to a central nitrogen spring contained in an 8 in diameter steel casing of oblate shape. Tests showed that this spring, although able to withstand 5 million working cycles, was unduly sensitive to changes of temperature; additionally, there was the feeling that a pressure vessel-particularly one situated close to the occupants of the vehicle-was a potential hazard in a collision, and the idea was abandoned.

Stage four, September 1957. Hitherto, damping in the pitch mode had been afforded by the viscous flow of the fluid in its interconnecting pipes, and damping in bounce and roll by mechanical valves in the spring unit. These valves were, of course, not progressive in action, and their opening and closing produced perceptible shocks in the test cars. Accordingly, rubber flap-valve dampers were developed, and sited within the displacer units. These had the virtues of being silent in operation, more reliable than orthodox valves, able to cope with the inherently high displacements of fluid, because of their large port areas­ progressive in action, simple to make, able to operate without lubrication and of low cost. Wheel-hop, which occurred before the re-siting of the dampers, was obviously as a result of this modification.

To reduce space required for installation, the interconnection pipes were then used as tubular rubber springs: each pipe consisted of an outwardly fluted rubber tube surrounded by braided steel tubing 17/8 in diameter. As fluid pressure increased with wheel deflection, the rubber tubing distended, and its flutes were elastically compressed against the casing. This spring, however, was subject to abrasion and kicking ­which caused the braiding to break up.

In the interests of perfect sealing, a diaphragm impermeable to the hydraulic fluid was essential. However; the production of an impermeable diaphragm was round to be unfeasible, so a separate liner of butyl rubber was incorporated in the displacer. It was moulded to the same shape as, but was not attached to, the diaphragm, and was similarly trapped at its periphery. Fluid pressure held it against the diaphragm. Owing to potential production difficulties with the type of diaphragm in use, a new construction was tried in which the reinforcement consisted of two semi-circular flat pieces of tyre cord ply wrapped into 60 deg cones, and placed one inside the other with the joints on opposite sides. There was no central bead, so the head of the displacer piston was dished in the centre, to locate radially the apex of the cone.  The  displacer unit could now be made more compact, since the dishing made room for the damper valves, which were mounted on a dished pressing; the piston was now a simple pressing.  In fact, at the end of 1957, all of the  main  components  in  the  displacer  were  pressings. Hitherto, normal experimental practices had been employed, with consequent  reliance on threaded, brazed and  welded constructions.
Stage five, February 1958.  A passing attempt was made to incorporate a simple Hydrolastic system in the smaller ADO 15 car, which was entering its final design stages. In this layout, a cone-type spring was used as a displacer, and a smaller version  of  the fluted spring was employed to augment the springing. Unfortunately, the cone spring had insufficient hydraulic stiffness, in that it became distended under  hydraulic  pressure;  the  rate  of  increase  of  fluid pressure bore no relation to the deflection of the displacer, and unreliable  damping  conditions  were created. Consequently, the car went into production with mechanically operated rubber springs and dampers of the orthodox hydraulic type.

Stage six, August 1958. After experiments had been carried out with a development of the fluted spring-now encased by a solid-drawn steel tube-this line of investigation was abandoned.  The extrusion of rubber tubing to the required  dimensional  accuracy was not feasible, and moulding was a forbidding problem. Additionally, accommodation of two long pipes of 1 7/8 in diameter presented difficulties on a small car.

Meanwhile, the type of hydraulic fluid was being decided on. Among others, paraffin and brake fluid were examined. but the first was ruled out for its incompatibility with rubber and the second because of its unsuitable viscosity characteristics and cost.  During this period, a degree of harshness in the ride –  experienced mainly over rough roads –  was cured by enlarging and tapering the outlet of the displacer unit,  to  eliminate  choking  of  the  fluid  discharge  when wheel velocities were high.  At about this time, a screwed insert was incorporated in the rubber moulding, to stop the point of the  cone-type diaphragm fretting in the dished piston.

Stage seven, December 1958.  Experiments were carried out on the damper flaps, to assess their stiffness and the ability of  the  rubber  to resist  being forced  through the punched holes.  When these were completed, the finalized spacing of the cords than was afforded in calendered plies, developed a technique of constructing the reinforcement from a single nylon filament, and the resultant diaphragm was found to have the required standards of strength, flexibility and fatigue life-over 2½ million reversals­ together with excellent dimensional consistency. In the production of this member, a tube of longitudinally disposed strands is created by threading the continuous filament around the teeth of two coaxial, toothed wheels placed about 5 in apart. When one wheel is rotated relative to the other, a hyperboloid of revolution is formed, and removal of the wheels leaves a hollow casing of waisted shape, with a short parallel skirt at each end.

The inner bead ring is slipped over the end of the casing, to rest at the small-diameter portion, and one half of the casing is then folded back axially on the other, to enclose the ring and form a two-ply reinforcement of approximately conical shape, in which the spacing of the cords and the relative angle of the plies are extremely consistent. In the moulding of the diaphragm, the parallel skirt is wrapped round the outer bead ring, and a layer of rubber fills the central orifice. A combination of flexibility and unusually high strength in the diaphragm has been one of the most important factors in arriving at the final configuration of the Hydrolastic unit, for the remarkable and characteristic degree of compactness could be achieved only by using dampers were calibrated against orthodox units at Cowley, and fatigue tests began at Bradford.  Initially, the damping practice was conventional, in that more damping was applied in rebound than in bump, but the production settings are now similar in both directions.

A rubber spring, of inverted dome shape, was incorporated in the upper part of the displacer unit, now tapered slightly, and the conduit tubing was changed to steel throughout, leading from a union in the side of the canister.  By virtue of these changes, a slight harshness over small irregularities on the road was eliminated, for there was unrestricted flow of fluid to the spring. To help increase the life of the diaphragm, a study was instituted to develop the optimum profiles for the piston and skirt. The purely conical shapes gave rise to an unacceptably small radius of bending at the bottom of the stroke, and a large radius-with attendant circumferential strain-at the top. Eventually, profiles were developed to reduce the severity in these extreme conditions, and also to provide a smooth transition phase between the two.

Final stage, June 1959 onwards. When it was found that the inverted dome-spring was weak in bending, a metal insert­, first a solid cone, later a cup-shape pressing, was bonded into the centre of the rubber element; this modification, coupled with an increase in the degree of taper of the cannister surrounding the spring, modified the stress distribu­tion and raised the fatigue life of the spring from under 1 million to over 2½ million cycles. The union for the conduit was transferred from the side of the canister to the end of the pressing in the centre of the spring. By virtue this move, the hoop-strength of the canister was improved, the opportunity was taken of shortening the unit, and a flexible hose, attached to a serrated projection on the inner cone, was interposed between the unit and the end of the steel conduit, which was now re-routed, this last modification considerably eased the problem of installation. The two-ply cone type of diaphragm was meanwhile giving good results on rig-tests, but it was felt that, because of the type of construction of the reinforcement, there would be difficulty, in production, of maintaining consistency in the spacing of the calendered cords during the moulding process.

However, the Dunlop Rubber Co, to obtain more uniform much higher fluid pressures than were hitherto associated with diaphragm-operated mechanisms of comparable size. The production item withstands a static pressure of 2,000 lb/in2. During the arduous rig- and road-testing that followed the finalization of the design, a fatigue weak­ ness was revealed around the -punched holes of the port­ plate; the gauge of the metal was therefore increased from 0-080 in to 0-104 in, and the diameter of the plateau portion was reduced.

The later prototypes of the Morris 1100 were subjected to the standard acceptance tests imposed on all B.M.C. vehicles of new design. This rigorous programme includes sustained running over the various rough surfaces at the M.I.R.A. proving ground and the Long Valley course at Chobham. Because of the high standards set during the development of the suspension components, no failures were experienced during these tests.

All of the rigs used in the dynamic tests were constructed in a manner such as to make dynamic creep of the rubber spring measurable, with the consequence that enough information on this subject was amassed to allow an accurate design factor to be introduced into the calculations relevant to the ride height of the vehicle. Static creep of the rubber was also assessed, of course. In addition, the impermeability of the butyl rubber liner for the diaphragm was appraised on a separate static rig. Over a period of 2½ years, it was found that any settling due to permeability could be ignored. Clearly, in the interests of maintaining approximately equal ride height at all four wheel, the deflected height of Hydrolastic units must be held to close tolerances. During manufacture, the top of the canister is machined, to provide a tolerance of ±0-015 in on the distance to the end of the piston rod. During this operation, the fluid pressure is held at 200 lb/in2, and a load of 2,000 lb is applied to the piston. On the vehicle assembly line, a novel method-protected by patent, as are most of the manufacturing processes used for the Hydrolastic system-is used to eliminate primary creep in the rubber springs. The Hydrolastic systems are pressurized for 30 min to about 400 lb/in2, which brings the suspension against its rebound stops and then reduced to the standard static pressure, 205 lb/in. By virtue of this technique, the springs have no opportunity of regaining the free state, as the weight of the car is applied throughout.

Design features

Among the noteworthy features not already mentioned the following :  

• Of the main metal components, all are pressings except the piston rod, a forging. The characteristic taper of some pressings is, of course, of benefit in the metal forming process. All metallic components in the system are cadmium plated, primarily to enhance shelf-life.
• Precision machining of the type associated with orthodox damping techniques is entirely unnecessary, and no lubrication is required during service.
• No gaskets or sealing compound are needed in the assembly of the unit.
• During the life of the vehicle, any settling attributable to secondary creep in the rubber can be compensated by restoration of the static hydraulic pressure.
• Angular movements of the actuating rod are catered for by the flexibility of the diaphragm.
• Because of the sturdy construction of the canister, the damper valves are extremely well protected; accidental dropping that might ruin an orthodox telescopic damper has no effect.

Service considerations

Arduous testing on rigs and on the road have shown that no noticeable drop in the hydraulic pressure can be expected during a twelve-month period. Tests carried out during development have shown that instantaneous dumping of the hydraulic fluid-simulating failure-from one of the systems in a car travelling in a straight line calls for no correction in steering; both springs are affected to the same degree, and the car merely sags to one side, with movement restricted as the bump-stops are contacted.

Similar tests on a car travelling round a comer at limiting tyre adhesion caused no alteration of cornering behaviour: in this instance the fluid in the system on the side of the car remote from the centre of turn was dumped. The hazard involved cannot be compared to that incurred by failure of brakes or tyres.  Gradual loss of pressure has no noticeable effect on handling or steering; however, it would appear advisable for users to have the pressures checked at least annually. Each B.M.C. main dealer will keep a small manually operated portable pumping unit for the purpose of charging the system when such a pressure drop occurs, or when work on a car is sufficiently drastic to justify uncoupling a spring unit. The pumping unit consists of an exhauster pump to evacuate the air, and a hydraulic pump to inject the Hydrolastic fluid and pressurize it.  During this operation, the transition from evacuation to charging is effected instantaneously by turning a three-way valve. Thermal expansion due to changes of ambient temperature in service does not give rise to any significant changes in ride height. However, vehicles exported to territories where extreme temperature conditions are experienced will have their pressures normalized on arrival at the distributors.

Larger Engine for Ford Consul Classic

At the end of last month, the Ford Motor Company began installing a re-designed, larger capacity engine in their Consul Classic and Capri models. Originally, the swept volume of the four-cylinder engines in these cars was 1,340 cm3 and this has been increased to 1,499 cm3 by lengthening the stroke from 65.07 mm to 72.7 mm.  At the same time, a change has been made from a three bearing to a five bearing design for the new crankshaft which, as before, is an iron casting.  A Zenitll type 33 VN carburettor is fitted to the new engine : the diameter of its choke is greater than that of the instrument on the 1,340 cm3 version.

The reason for the change to a five journal shaft was to eliminate main bearing thump which the previous unit had exhibited and which was found to be accentuated on a small number of 1,499 cm3, three-bearing engines that were built initially.

In the latest engine, since the loading at each bearing is lower than that on the three bearing version, it has been possible to change from copper-lead to white metal main bearing shells. The dimensions of each bearing have not been reduced and, in fact, the same components are used as in the 997 cm3 Anglia engine, of the 105E model, with consequent production economies.

Other changes to the engine are few. The cylinder block­ crankcase casting is, of course, deeper to accommodate the longer stroke and the new connecting rods are of slightly heavier section. The oil capacity of the sump has been increased by 50 per cent to avoid an increase in oil running temperatures and the diameter of the inlet valve heads has been made a little larger. The compression ratio has been reduced from 8.5: 1 to 8.3: 1.

Maximum net power output of the 1,499 cm3 engine is 59.5 b.h.p. at 4,600 r.p.m, and the maximum net torque, of 81.5 lb-ft, occurs at 2,300 r.p.m. The corresponding figures for the superseded 1,340 cm3 unit are 54 b.h.p. at 4,900 r.p.m. and 74 lb-ft at 2,500 r.p.m.

A new gearbox having baulk-ring synchromesh on all forward gears is now fitted to these models. Previously the three upper ratios only had synchromesh. The overall ratio for first gear has been changed from 16.99: 1 to 14.61: 1. Sealed bearings and joints, which require no periodic lubrication, are now fitted at the front suspension assemblies, the steering tie rods and the propeller shaft universal joint.

Perbury Continuously Variable Gear

Interim Report Issued by the National Research Development Corporation

An interim report has been released on the Perbury gear, the development of which is being sponsored by the National Research Development Corporation.  Basically the work of F. G. de B. Perry, this gear is, in principle, the Hayes continuously variable ratio transmission which was offered by Austin as optional equipment in 1933 on their 16 h.p. and 18 h.p. cars. The Hayes transmission was described in the Automobile Engineer of November 21st, 1933.

The engine torque is transmitted from two driving discs, spaced apart on the input shaft, through two sets of three rollers to a centrally disposed double-faced driven disc. This driven disc is connected to the output shaft by a drum­ shaped member. The six rollers are mounted on roller bearings on two stationary three-armed spiders and their convex peripheral surfaces roll along concave circular tracks formed on the faces of both driving and driven discs. In the Hayes transmission the axial closing force required to establish drive between discs and rollers was provided by the reaction from a series of steel balls acting on ramps at a coupling on the input shaft, the axial force being propor­tional to transmitted torque. For the experimental Perbury gear this force is provided by hydraulic means.

The speed ratio between input and output shaft, which can be varied between a 2-7: 1 reduction and a 1-75: 1 increase, is, of course, dependent upon the angle of the rollers relative to the driving and driven disc. To change ratio, the rollers are tilted simultaneously, by means of a hydraulically actuated linkage, so that the axes of the rollers do not intersect that of the input shaft, the resultant precession causing the rollers to be steered to a different attitude.

A study of the operation of oil film drives, of which the Perbury gear is an example, has been made by Perbury Engineering Ltd, and a gear installed in a Hillman Minx has covered 25,000 miles. A theoretical investigation into the conditions in the oil film is being carried out in the Department of Mechanical Engineering at Leeds University, and use is being made also of earlier investigations.

Perbury gears are being tested in pairs on a recirculating power test rig designed and built by Plint and Partners Ltd, of Wargrave.  These tests have established that when 100 h.p. is being transmitted at an input speed of 5,000 r.p.m. the overall transmission efficiency is 94 per cent at a reduc­tion ratio of 2: 1. Losses at the oil film amount to approxi­mately 3 per cent, the remainder being accounted for by bearings, seals and oil churning losses. Variations in overall efficiency are small throughout the range of ratios which can be obtained.

Endurance testing will begin shortly on a second test rig and it is expected that, as with the Hayes transmission, the limiting factor will be surface fatigue of the rollers. However, it is suggested that the normal roller bearing formulae relating life to stress will apply and that when 100 h.p. is being transmitted stresses are not high by industrial ball bearing standards.

A design study to assess the automotive use of the Perbury gear has been made by Norris Brothers Ltd, Burgess Hill. There are also possibilities in the use of this type of transmission  in  machine  tools,  test  houses  and  laboratories.