From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
Article from 'The Autocar' magazine discussing car body design principles, including streamline form, weight distribution, and frame distortion.
| Identifier | ExFiles\Box 160\2\ scan0002 | |
| Date | 8th August 1914 | |
| 262 THE AUTOCAR, August 8th, 1914. Considerations of Body Design. Wind Resistance, Weight Distribution, and Frame Distortion. By W. G.{Mr Griffiths - Chief Accountant / Mr Gnapp} Aston. THE Editor of The Autocar has from time to time been indulgent enough to allow me to make some criticisms on current body design. In most cases the criticisms made have been “destructive,” but it has occurred to me that destructive criticism alone might, perhaps, be not what was needed, and I have therefore embarked on the present article, which is primarily constructive. Now there are so many different kinds of body work, and the field of utility which they cover is so great, that it is quite impossible to cover the whole ground in one, or even a series of articles, and one is forced, therefore, to concentrate on a single type. On looking through “The Autocar Fuyer's Guide,” and making the necessary allowance for the proportionate numbers of the various cars turned out, one reaches the conclusion that the medium-powered car, that is to say, with an engine between 15 and 25 h.p., possesses an enormous numerical preponderance over all others. Furthermore, the majority of these medium-powered cars are fitted with a five-seated body, which, as most of their catalogues will tell you, is the standard type. It is then to the five-seated body of the medium-powered car that I propose to direct a few sharp criticisms. My first suggestion is, briefly, that this five-seater body design is entirely wrong. In accordance with natural laws, the sting should be in the tail of what one writes, but in this case I think it is, perhaps, best to start off with an outrageous statement, and then proceed to justify it. I would merely ask, as Sherlock Holmes so often asked Dr. Watson, that as we go along you will kindly correct me if I am at fault. 1.—Streamline Form. I take this consideration first, because it is really the most trivial, but is generally regarded as the most important, which reminds me that the other day I saw a very special arch-knut driving a car answerable to his highest ambitions. Its principal feature was that it had cone-shaped covers in front of the head lamps, in order to lessen their head resistance. These, elegant as they were, contrasted rather ill with the perfectly square panel at the back of the body. The latter would probably absorb about 3,000 per cent. more power than the unenclosed head lamps. It looks from this as though I rather cast ridicule on the idea of streamline form being of real importance in touring body design, but, as a matter of fact, I do not do so in the least: I only suggest that it should be about the last thing that should claim consideration. Other things being equal, the streamline body is superior to one which outrages all notions on the flow of fluids. But, after all, the average speed which the average car does is so extremely low that wind resistance only accounts for quite a small percentage of the power given off by the engine. The only thing which justifies one in considering this aspect of the matter in connection with touring body design is this; that if one or two horse-power can be saved, there is at least that much to the good, and if it has a negligible effect under ordinary circumstances, it may at least show its value when driving into a head wind, and in consequence, the effect of the design is more marked than would otherwise be the case. Figs. 1 and 2 are designed to show the contrast, and the former illustrates nature's idea of streamline form and represents in a very crude manner the plan view of a fish. It is true that this travels in a fluid which is considerably heavier than air, yet this fact has really only a slight effect on the actual shape when working in a lighter medium. As a matter of fact, in air, a streamline body can be made considerably more blunt and fat than if it has to pass through water, and this, of course, is an advantage from the body maker's point of view. On looking at the plan view of a typical five-seater as shown in fig. 2, one realises that if body designers had tried to make something that was not streamline form, which was, in fact, as far removed from it as possible, they would have achieved a very notable performance. As a matter of fact, from this point of view, the average car is far more efficient when it is driven on the reverse than when it is travelling forward. As that great scientist Froude observed, and his work in connection with the flow of liquid remains classically valuable to this day, it is the blunt tail and not the blunt nose that causes eddies. If this means anything, it means that the wise man who has to effect a compromise, will devote more attention to the tail than to the nose. The modern body designer not realising this has, of course, done the very reverse, and in this ridiculous enterprise he has been encouraged and abetted by the chassis designer, members of which have literally fallen over one another in their endeavour to devise some form of radiator which shall, at least, give the appearance of being somewhere near “streamline form.” The tapered bonnet and the scuttle dashboard are also cases in point, but the latter at least has some foundation in sense. The idea of a pointed radiator offering less wind resistance than a flat one is, of course, utterly absurd, and strongly suggests the old preparatory school question which asks whether you can build more houses on the sides of a hill than on the area of ground which the hill covers. It always seems to me that the designers of so-called “streamline radiators” have an idea at the back of their mind that, by making something of suitable outward appearance, they will be able to cheat the forces of Nature, and persuade Nature to do something which she would be otherwise averse from doing. In short, because they call it a streamline radiator, they expect Nature to treat it as such. Nature does not. The bottle-shaped body, as illustrated in fig. 2, is narrow where it ought to be broad, and broad where it ought to be narrow. It is square where it ought to be sharp, and sharp where it ought to be blunt, and all these faults are aggravated when one considers the manner in which the passengers are disposed, an accepted arrangement being two in front and three behind. THE AUTOCAR, August 8th, 1914. 263 Considerations of Body Design. Like the rest of the body, this is the exact reverse of what it ought to be. Three persons sitting in front of two will shield them against a current of air, but two persons sitting in front of three will not shield more than one, and therefore the other two, being unshielded, will add very materially to the total wind resistance. I am aware that on most cars a glass wind screen effectually protects the front passengers, but a point which perhaps may not be generally recognised is that in doing so it merely places a premium upon the resistance which can be exercised by those in the rear. At moderate speed, the disturbance in streamline form, caused by the windscreen, is very purely local; that is to say, the air currents join together again, as suggested in fig. 3, before they reach the rear passengers, each of whom, therefore, exercises approximately as much resistance to the air as though he were stuck in the front of an entirely unprotected car (see figs. 3 and 7). So much for wind resistance. Other points suggest themselves in this connection, but they may be safely left as being more than trivial. We now come to 2.—Weight Distribution. In the modern five-seated body this is all wrong. A statement naturally begs a question, and to which the American mode I would wish to add another. To wit, Which type of flywheel has the greater effect, both being of the same weight, the one of small diameter or the one of large diameter? Now if we turn to fig. 4 we shall see a side view of the ordinary five-seater body, in which I have sought to represent the massing of the various weights. These are, namely, the weight of the passengers and the weight of the mechanism. The former is represented by round patches of black, the latter by square patches. The distribution of the latter by square patches shows the distribution of the gravity of the whole vehicle, including passengers, is marked approximately with a cross. Now when a car strikes a bump on a road, the fact that it hits the inequality first with one pair of wheels and then with the other causes it to indulge in a series of oscillations in a vertical plane, which have to be damped out by the springs, fore and aft. These oscillations consist more or less of a circular movement about the centre of gravity of the car, and this point is therefore alogous to the pivot of the spindle of a flywheel. Since the two opposite spokes, as it were, of the flywheel are of large diameter, and their weights at the extreme ends, we have what is called a considerable moment of inertia, and when you start them swinging about the centre of gravity of the car they show very little tendency to come to rest, and consequently the oscillations up and down continue very much longer than they would do if the whole of the weights concerned were massed closer to the centre of gravity. This ill-effect has been increased by virtue of the fact that many modern designers have adopted the practice of making their engines in a very compact unit, and transmitting the power therefrom through a long propeller-shaft to a heavy back axle, which not infrequently also incorporates the gear box as well. This means that the principal mechanical weights of the vehicle are placed one at one end of the chassis and the other at the other; that is to say, they are as far apart as they can get. This arrangement, together with that of the passengers, not only has its effect in the vertical plane, and results in the springs having to get rid of a great deal of energy in the damping out of oscillations, but it has its effect in the horizontal plane, and tends to prevent a car from deliberately following the course set by the driver when turning corners. The lamentable phenomenon of “rolling” results from this cause as well as from the fact that springs are sometimes attached to the chassis con-point at which they are attached to the axle, and as they have a certain amount of elasticity in the sideways direction, as well as in a vertical one, they tend to stop the “flywheel” and then turn it back again, thus enabling it to continue its horizontal oscillations. In one or two cars which I have driven, in which the weights have been very much fore and aft, this effect has been decidedly noticeable, and, as I have pointed out above, it is exaggerated by the disposition of the passengers. Whilst public demand discourages striking departures from conventionality, designers are almost compelled to follow, what I consider, the now unnecessary practice of putting their engines under bonnets in the front of the car, it is too much to hope for a betterment of this state of affairs so far as mechanical weights are concerned, but the body builder could at least do something by massing the weights of his passengers a little more symmetrically. Thus there will be a very distinct improvement if he place three passengers in the centre of the vehicle, that is to say, one in the front seat, and two in the rear as in fig. 6, and this, as we have seen, is the correct principle from the streamline point of view. It will not only reduce wind resistance, but it will make the car more comfortable to ride in over bumpy roads, and it will make it certainly very much more easy to steer round corners. In consequence it will result in less wear and tear, not only to the tyres, but to the car as well. Fig. 1.—True streamline form. Fig. 2.—Plan of average five-seater. Fig. 3.—Five-seater in plan, with air currents. Fig. 4.—Weight distribution on five-seater. Fig. 5.—Weight distribution with back axle gear box. Fig. 6.—Suggested form of five-seater. | ||