THE DEVELOPMENT OF STREAMLINING
THE DEVELOPMENT OF STREAMLINING Cecil Glutton outlines the History of Wind-Drag Defeat and suggests a New Formula by which the efficiency of Streamlining can be assessed
PEOPLE began to worry about wind resistance quite early in the history of racing, and even in 1898 we find CharrOn decorating the front of his Panhard with a sort of pyramidal projection, which was doubtless intended for the purpose of cleaving the atmosphere.
Generally speaking, there seem to have been two schools of thought : those who set about reducing frontal area to a minimum, and those who shaped their cars to cut through the air. A combination of the two did not seem to occur to anybody.
Some early cars had remarkably low frontal area, achieved by the use of horizontally opposed engines, and notable examples were the Welseley Beetles and Winton Bullets of 1903. The latter was reminiscent of nothing so much as an inverted punt, and in both cars the steering wheel was the only thing that stood higher than the wheels ; the only thing, that is, execpt for the unconsidered trifle of the drivers, who were most mortally exposed to the elements.
The wind cutters, following on Charron's tentative opening bid in 1898, took example from rowing boats, and the racing Boll& of 1899 has a most amphibian appearance. Later on the tendency was to keep the boat, but turn it upside down, and a good example of this was Gabriel's winning Mors in the 1903 Paris-Bordeaux race. The radiator was placed low between the wheels, and the whole of the rest of the car was nothing more nor less than a capsized canoe.
The first serious advance in streamlining came with the early electrical record breaker3, which, of course, had the advantage that they could spread their machinery about as seemed best to them. The Jeanteaud of 1898, which took the flying kilometre record at 39.2 m.p.h., had quite a good outline, but better still was Jenatzy's famous " Jamais Contente," which was exactly cigar shaped, and raised the record to 43.7 m.p.h. during 1899. But the first scientific thought devoted to the matter, as opposed to purely empirical experiment, seems to have come from the ever-fruitful brain of Laurence Pomeroy (pere). Working on the basis of the tremendously Successful 1908 trials car, which had a four-cylinder engine of 90 x120 mm. (3,050 c.c.), in 1909 he fitted it with the first genuine streamlined body, for record-breaking purposes at Brooklands. This body at once reduced frontal area to a minimum and fulfilled the main
essentials of good streamlining. The largest cross-section is very near the front of the car and the atual wind-cutting is looked after by a radiator cowl with only a very narrow opening. At the other end we have what is probably the first appearance of the tapering tail, sticking out some distance behind the back axle. The engine at this time gave 52 b.h.p., and this was sufficient to traverse the flying kilometre at 88.26 m.p.h. and 10 miles et 81.23 m.p.h., which Was faster than any previous car up to 60 h.p. rating.
The frontal area of this car was the exceptionally low one of 7.5 square feet. This outstanding performance put 100 m.p.h. within reach of a 3-litre car, and with the data gained Mr. Pomeroy was able to calculate the frontal area and b.h.p. which would actually achieve this figure. The new body was a slight improvement aerodynamically on the 1909 edition, and the engine, which had been reduced to 89.7 x 118 mm. to bring it within the 8-litre class, now gave 60 b.h.p. With this, in 1912, the flying kilometre Wi s covered at 101.4 m.p.h. and the mile at 99.61. Ten laps of Brooklands were
travelled at 92.13 m.p.h., raising the record for Class E by 14 m.p.h., and the World's record (unlimited) for 50 miles was taken at 97.1 m.p.h. The 3.05-litre car had, however, clocked 100 m.p.h. as early as 1910.
Mr. Pomeroy's notebook contains some very interesting data of experiments made with this car on April 27th, 1912. First of all, it was run with all wind-cutters, disc wheels, radiator cowl and minus the spare wheels. The maximum speed was then 81.74 m.p.h., the water tempere hire 195° F. ana the oil temperature 175° F. Taking off the discs reduced the speed to 80.42 m.p.h. and carrying two spare wheels brought it down to 80.01 m.p.h. The radiator cowl was then removed, reducing speed to 79.02 m.p.h., but the water now ran at 159° F. and the oil at 146° F. Run as a bare chassis the car could only achieve 71 m.p.h. These very interesting relative speeds appear to have been taken over a distk.nce of 10 laps. Following Vauxhall's lead, other makes, especially Sunbeam and Talbot, began to take an interest in streamlining for use at Brooklands, but it was not until
1921 that designers began to use it at all on road-racing cars. A problem which hardly confronted early exponents of streamlining was the wind resistance set up by the wheels and axles. Tyres seldom had a greater section than 444", axles were flimsy, and the front axles were unembarrassed by brake drums and operating gear. The front axle itself was frequently, however, padded out to a streamlined shape, of which Vieux Charles's is a good contemporary example. So far as the body was concerned, however, ti ere was no question of making it anything but the narrowest which would
as years went on and tyre sections grew to 6" and 7W, in conjunction with large brake drums, perrot shafts, heavy axle beams and the like, the resistance set up became a serious menace and was frequently as much as the whole of the rest of the cur put together. Designers were now faced with the revolutionary step of increasing the frontal area in order to decrease wind resistance. Bugatti was probably the first man to realise that a perfect streamlined body should envelop the whole car, wheels included, under an uninterrupted contour, and exemplified this in the queer tanklike racing machines of 1923. Ila.ving regard to the small tyres still used at that date, he was probably before his time, and the cars were unsuccessful on other grounds, so that he seemed temporarily to lose interest in streamlining ; but his 1937 Le Mans winner was still palpably related to its precursor of 15 years earlier. One or two other people, such as Chenard-Walker in 1925, made fitful excursions into the realms of streamlining, but it can really be said that no
serious advance was made from 1909 until the appearance of the revolutionary Mercedes and Auto-Union Grand Prix machines in 19:34. In these machines the overall streamlining of the bodies showed a marked improvement upon the current Grand Prix Alfa-Rorneos and Bugattis, which had previously held the field, while the suspension was largely enclosed and increased attention paid to the removal of any avoidable excrescences.
For record-breaking purposes these cars appeared with complete streamlined shapes of exceptional aerodynamic efficiency, but these would not have been practicable for road-racing. Perhaps the greatest recent advances in streamlining of everyday utility have been made at Le Mans, where the regulations put a limit on the amount of engine modifications which can be made, so that additional speed can only be sought by
streamlining. In view of the tortuous nature of much of the course the cars must, nevertheless, be compact and manceuvreable. Adler achieved splendid results with very paltry motive power, but undoubtedly the greatest advance of all was achieved by B.M.W.'s in the 1940 Mille Miglia Race, when their s-litre (a development of the "Type 328 ") appeared with a large saloon body enclosing all the wheels under one smooth shell. The tail of this car is by no means egregious, visibility is good, the doors are wide, three passengers could sit abreast, and altogether it would be an entirely practicable motor-car in which to go shopping. In this instance, the engine (unblown) had been tuned to give the very fine output of 135 b.h.p., which was sufficient to urge the car along at 135 m.p.h. Despite the capacious coachwork, the total weight was only 15 cwt. and the petrol consumption was roughly 18 m.p.g. at 100 m.p.h. A splendid example of production streamlining was the " Special" 1,100-c.c. Fiat, marketed at E375, of which Gordon Wilkins wrote that inspired road test in
the Motor of July 25th, 1939. Here was a three-seat-in-line saloon which only 42 b.h.p. would carry along at 90 m.p.h. and whose petrol consumption at a cruising speed of 60 m.p.h. was 40 m.p.g. The tail was rather overdone, but this could perfectly well have been reduced without loss of performance.
It has always struck me as unfortunate that there is no generally accepted yardstir k by which the aerodynamic efficiency of different ears can be compared. Such a yardstick can, nevertheless, I think, be quite easily devised.
It is a fairly accurate formula which says that the speed of any car varies as the cube root of the b.h.p. per square foot of frontal area. It therefore follows that for any given chassis-plus-coachwork the speed divided by the power (as above) will always be a constant. To take an unstrearnlined example in
the first place, the famous Peugeot which won the Grand Prix of 1912 gave about 130 b.h.p., had a frontal area of 16.2 square feet and a maximum speed of about 105 m.p.h. The power per square foot is, therefore, 8 and the cube root of 8 is 2. Our fraction is therefore 105/2, which equals 52.5. Now, whatever the horsepower of any engine which might be inserted in this car, the maximum speed would wry proportionately, so that the fraction would always work out at 52.5; 52.5 may therefore very conveniently be taken as the coefficient of aerodynamic efficiency of the 1912 G.P. Peugeot. Going to the other end of the scale, by taking the 1940 Mille Miglia B.M.W., we have a frontal area of some 22.5 square feet, giving 6 b.h.p. per square foot. Adopting the above formula we get 75 as the figure of merit for the B.M.W., representing an improvement of 25% over the Peugeot and a higher maximum speed by 23 m.p.h. despite 2 b.h.p. per square foot less. Incidentally, it only requires 55 h.p. to drive the B.M.W. at 100 m.p.h. I The efficiency of the 1,100-c.c. Fiat just men tioned is not much lower than the B.M.W'
Forrest Lyeett's Bentley affords an interesting example of a moderately streamlined body of small frontal area with entirely unfaired 'wheels and axles. The maximum of this car in touring trim cannot be less than 130 m.p.h., the power output can fairly safely be assessed at 240 b.h.p. and the frontal area at about 16 square feet. The figure of merit on this data is only 52.5, showing that the turbulence set up by fat tyres, etc., more than offsets the inferior streamlining of the 1912 Peugeot.
It is interesting to note that with cars like the Bentley, having unenclosed wheels, an undershield produces a most advantageous result; but in completely streamlined cars, like the B.M.W., the wind has no temptation to get underneath the car, and an undertray can quite well be omitted.
To take yet one more common type, the 1938 T.T. Delahaye works out at 60, with the windscreen flat.
This formula works fairly accurately for speeds between 80 m.p.h. (when streamlining first becomes really valuable) and 150 m.p.h. (after which speed frictional losses between tyre and road begin to upset things by assuming quite alarming proportions). Nevertheless, two apparently identical cars may conceivably yield different figures of merit because the frictional losses of the transmission systems are unequal. It is, however, instructive to see how the formula works for very high speeds, such as were attained by the 1934 G.P. Auto-Union. In roadracing form it had a frontal area of 12.2 square feet, gave 310 b.h.p. and attained about 185 m.p.h. This gives no less than 41.75 b.h.p. per square foot, despite which the figure of merit is only 53.5. This is partly accounted for by the increased frictional losses just mentioned and partly by the very large wheels used. As against that, the record-breaking car had a frontal area of 16.5 square feet, gave 525 b.h.p. and attained 242 m.p.h., giving a figure of merit of 75, which is identical with the Mile Miglia B.M.W. At speeds of less than 150 m.p.h. its efficiency would therefore be quite fantastic—probably somewhere in the order of 100, or even perhaps 120!
In the foregoing figures of frontal area the area of the front wheels only has been taken where all the wheels are exposed, although in point of fact the rear wheels set up almost as much resistance again. The usual method of calculating frontal area is, however, to take one pair of wheels only, and it has therefore been followed for the sake of comparison.
Some of the data which has been given about the Grand Prix Auto-Unions his not, I believe, been previously published. When I started to write this article I appealed to the technical staff of the Motor for much of the technical data which I have employed, and they kindly gave me everything I needed, including the Auto-Union figures, and the original for an extremely interesting and instructive line drawing showing half the 1909 Vauxhall against half an Auto-Union, while the outline of the record-breaking Auto-Union is shown by dotted lines— the beginning and end of streamlining in a nutshell.