F actors Affecting
Page 34
Page 35
Page 36
If you've noticed an error in this article please click here to report it so we can fix it.
ACCURATE CONTROL
By J. Pickles, A.M.I.Mech.E.
FOR a long period the search for the ideal method of insulating the load of a vehicht from shock lay in improving the method of springing. To this end, some efficient, as well as some totally impossible, schemes have been evolved, but invariably it was on the subject of springing that most . thought was concentrated. The direction of travel and the manner in which the vehicle answered to the steering wheel were considered to be steering problems.
To-day, however, the tendency, which is based on the experience of older workers, combined with the ever-greater use of scientific investigation methods, is to attempt to improve the suspension. This somewhat abstract term is used to describe not only the actual springing, but includes, as a study, its effect on general handling. Frame design, steering, axle position, weight distribution and tyres must all be considered by the conscientious suspension engineer. He must bear in mind the safety, as well as the comfort, of vehicle users.
Divergence from Theory It is a simple matter to demonstrate on a drawing board the geometrical accuracy of a steering layout, but even if this be so—and with existing linkages, it can never be—at other than the lowest speeds, the vehicle will follow a different path from the one which is theoretically correct.
Knowing the factors involved, the behaviour of a chassis may be closely predicted in the design stage, Lut the ultimate test is, of course, that of operation on the road. In the past there have been many vehicles on which the most elementary errors have appeared, and even on some having the hallmark of careful thought, a driver, unused to his charge, could have some exciting moments.
A vehicle of Continental origin, which I examined, came in this category; it was excellently designed, A32
solidly built and had an ingenious suspension system. The amply powered engine offered every encouragement to the attainment of speeds somewhat higher than the average, but when pulling out to pass a stower-moving vehicle the tail
appeared to swing violently towards the kerb, with a similar effect towards the opposite side when straightening out to pass. This rapid tail wag, although causing no danger under the particular conditions, would, in many cases, result in some vicious sliding.
It is not enough to say that one should not travel so fast that a vehicle becomes unsafe, because safety does, on occasions, rely on speed. What occurs at high speeds in dry weather may well occur at low speeds in wet weather, or during heavy braking.
Any one interested in motor vehicles can explain , how, with Ackermann steering, the wheels are turned at different angles to give. true rolling when turning a corner. Quite a few, too, could point out that lines produced through the stub axles of the steered wheels should intersect, at the centre of the turn, a line produced through the back axle..
The accent is, however, on the way • that the front wheels are turned to.. provide steering, and the fact that the back axle is as vital as is the front is usually ignored.
Turning Rear Axle
For instance, let us suppose that. with the front wheels fixed in the straight-ahead position and the chassis in motion, the rear axle is turned in a steering sense. The vehicle would thus follow a curve, but, in this case, the steering would be instigated from the rear, and the radius of the turn would be governed by the distance from the centre of the chassis, of the outer section of lines drawn through the axes of front and rear wheels.
To digress for a moment, let us consider the orthodox semi-elliptic spring. As it is deflected, the centre distance . between the spring eyes must vary so that only one eye can• be pivoted directly to the frame, the other end being suspended on a shackle.
The axle, in consequence, does not rise and fall in a straight line, but on an arc, having a centre adjacent to the fixed spring eye. The spring may thus, for geometrical purposes, be replaced by a hypothetical lever pivoted on the virtual centre of the spring and fixed to the axle. The axle rise and fall must be a partial rotation about the pivotal point, and the longitudinal distance will be greatest when the hypothetical lever is flat, decreasing in increasing measure as the angle changes from the horizontal position.
.During roll, the spring towards the outside of the turn is depressed and the opposite one is lifted. If both springs be so loaded that the virtual lever is flat—that is, the virtual centre of the spring and the axle centre are equidistant from the ground—both axle centres will move forward by a like degree. The axle will thus maintain the correct angular disposition with the longtitudinal eentre line.
Skewed Axle
Supposing now that the two imaginary levers slope down towards the axle. The depressed spring will thus, in rising, cause the axle centre at that side to move up the arc towards the flat position, when the lorcitudinal distance between axle and virtual centre will be at its greatest. The opposite spring, on the other hand, moves so that the virtual centre and axle-centre distance decreases. In consequence, the axle is forced to skew in relation to the frame (Fig. 1).
In the case of springs having the normal camber, this condition, in which the virtual centre is higher than the axle centre, must occur, so that when turning a corner an additional proportion of steering action is instigated by the rear axle.
To minimize this trouble a certain amount of reverse spring-camber is necessary to achieve the required geometry. An excessive degree of spring camber in the static loaded position will cause the reverse effect on the steering, resulting in a reduction in the amount of steer that would be correct from a theoretical consideration of the geometry of the system.
Ideal Spring Camber
Even with the spring cambered to give the best condition, only partial correctness can be achieved, as accuracy depends on the loading being constantly the highest for which the vehicle was design& whereas, of course, a wide range of loadings must be catered for.
With a normally cambered spring the result is to give what is termed an " oversteer " effect, and an excess of reverse camber, an "understeer." Axle skewing, however, can cause only a comparatively small proportion of understeer or oversteer effect, as other conditions may serve to augment or more than counteract, such faults.
To counteract understeer or oversteer from other causes, the rear springs are sometimes sloped, so that the deliberate skewing which is introduced may overcome the initial fault. This form of cure is, at best, uncertain, because of the variation in loading already referred to, although partial success has been achieved by this method, it is considerably better to minimize the original trouble rather than to cure a fault by introducing another.
Front axles are not so susceptible to these troubles, as the springs are stiffer than those at the rear, and the load variation is usually small compared with that at the rear.
In the case of a narrow wheel there would be no difficulty were it to run at any angle to the road, but a road wheel has, of course, considerable breadth. When running perpendicularly to the road, the distance between road surface and stub axle will be identical at any point between either face of the wheel. When the wheel lies at other than 90 degrees to the road, the distance between road and stub axle must vary between the two outside faces of the wheel; that is, the rolling radius is not constant over the complete width of the wheel.
Thus, as the angular velocity of the wheel cannot change to suit the rolling radius, it must follow that, for true rolling, that part of the wheel having the smallest rolling radius should have the shortest distance of travel for each revolution. This condition is obtained when the ratio of radii of turn to rolling radii over the full width of the wheel is identical. A taper-roller bearing is proportioned so that this requirement is obtained.
Untrue Rolling
With a perpendicular wheel, true rolling cannot occur when turning, but with a heavy positive camber, matters are made even worse. Increased tyre wear is a result of such untrue rolling, but, in addition to this, the tyre tends to move about a path which would permit correct running. A perpendicular tyre tries to run in a straight line, and a definite force must be exerted to constrain it to run on any other path.
The force exerted by the tyre tending to follow its natural rolling path is termed camber thrust, and this is but one of the forces tending to modify the direction of travel from that which, in theory, is correct.
With the orthodox beam-axle suspension, all wheels remain sensibly perpendicular, although twin tyres cause fairly large variations in ratio between the rolling radii of the inner and outer faces of the two wheels and their respective distances from the turning centre (Fig. 2).
Side Sliding
The road wheels of a vehicle during high-speed cornering are called upon to resist side sliding, and their capacity to do so is dependent on adhesion. There is, therefore, a limited force regardless of direction, which can be applied before sliding occurs.
Although various designs of tread will permit somewhat higher side loads to be borne than others, once sliding is instigated, only a small force is required to change the direction of slide, no matter what type of tyre be concerned.
Thus, supposing that the engine power transmitted to the rear wheels was so great that the wheels were on the point of spinning, a very little centrifugal force would be sufficient to cause side sliding. It follows that the greater the engine power transmitted by the wheels, the lower will be the resistance to skidding.
With rear-wheel drive there is the advantage that a rear-end slide can be controlled, whereas, with frontwheel drive, the initiation of side slide leaves the driver helpless.
The importance of the behaviour of a tyre is only now beginning to be fully realized, although there is still widespread misapprehension that a pneumatie-tyred vehicle behaves like a railway coach with
its flanged wheels to provide guidance against the rails.
As an instance, the rear wheels, being driven by the engine, are required to transmit the driving effort to the road, relying on adhesion between tyre and the road to prevent spinning. When turning a corner at speed, this adhesion can be overcome to cause sideslip, but at speeds lower than the danger point a somewhat different phenomenon occurs. In addition to the driving force on the tyre there is the effect of centrifugal force tending to move the vehicle bodily away from the turning centre, the magnitude of which will depend on both speed and radi us.
Effect of Side Wind The normal downward static load on the tyres is modified, being greater on those on the side away from the turning centre and correspondingly less on the near side. Similar conditions occur when a side wind strikes the vehicle and, in considering the behaviour of the tyres, the effect of a side wind will no doubt be that which is easiest to appreciate.
With the vehicle travelling in a straight line, imagine a pneumatictyred wheel being subjected to a side wind concentrated at the centre point. This will have the effect of moving the wheel bodily sideways, whilst the wheel will tend to maintain its original direction of motion, so that the actual path will be at an angle to that which was originally traversed by the wheel.
Drift This side action, superimposed on the forward direction, must in no sense be confused with skidding, as quite large angles of deviation can occur without excessive tyre wear. Side drift is analogous to that occurring to a man when walking and subjected to a strong side wind; if resistance be not imposed he will find that his direction of motion will bt modified.
Drift or slip angle from a constant pressure depends on the tyre section, tyre pressure and load carried. The last factor is of vital importance, in that it is this which most greatly affects the particular degree of handling accuracy.
Side drift, as already stated, is not accompanied by any undue tyre wear, so it follows that the part of the tread in contact with the road is aligned with the actual direction of travel. This means that it is skewed in relation to the wheel, which remains oriented in the original direction High resistance to side drift, therefore, must result from tyres having high deformation resistance, high tyre pressures being better than low pressures (Fig. 3).
Cornering or side wind causes a transference of load to certain tyres, thereby compressing the section between rim and road. Obviously, the greater this deformation the less will be the resistance to skewing and consequently the lower the resistance to side drift. In engineering parlance its cornering power is reduced, cornering power being the capacity of the tyre to resist side drift.
Another factor in the cornering power of a tyre is its angle to the road. In the case of present-day commercial-vehicle chassis, this .factor is not of great importance, as all wheels remain sensibly vertical, but should independent suspension become general, then the position will be different.
With a true parallel-action suspension the wheels remain parallel with the vertical centre line of the body and must, therefore, make an angle to the road other than one of 90 degrees. Positive camber of the wheel to the outside of the curve— that is, camber in the usual direction —results in reduced cornering power, arising from increased camber thrust (Fig. 4).