THE RACING CAR ENGINES OF 1935-1940
THERE ARE, I think, few more interesting subjects to the motoring enthusiast than the progress of racingcar design through the years, and to my mind the series of articles on “Racing Car Evolution ” which appeared in MOTOR SPORT some two years ago are amongst the most interesting that have ever appeared. Of these Laurence Pomeroy’s three particularly stand out, covering, as they do, one of two periods when development has been most spectacular, and it is a matter for much regret that the period 1983 to 1989, which saw equally spectacular development, has not been covered by the same pen in the same manner. Presumably, the necessary facts and figures for such an analysis are still more or less closely guarded secrets. Another difficulty is that progress was made in four parallel directions -in Grand Prix racing and in the three
small-capacity classes, which renders comparison difficult.
In an effort to get some idea of the direction in which progress was being made, I spent some hours one evening last winter collecting as much data as I could regarding racing engines of the period 1985 up to and including the very short Italian racing season of 1040. Since then I have loosed my slide rule on these figures, with results that are interesting but of very limited valite, for two reasons. First, without casting any aspersions upon our excellent semi-technical Press,
I feel the figures are not too reliablein fact, in some cases one finds two completely different sets of figures for one car. Secondly, little or no indication is given of the state of tune for the particular output quoted. However, for what they are worth I give selected examples from my delvings in an accompanying table, and have followed it by certain remarks on the various engines listed therein. In the By Capt. John Moon Last month Capt. John Moon described his idea of a I4–litre G.P. car. In this article he reviews the racing engines of I935-40, and the figures he quotes make an interesting comparison with those he would expect to obtain from his ideal
table I have attempted to give in. the remarks column some indication of the state of tune by writing “long race,” indicating state to withstand 200 miles or more of racing, “short race” to cover races up to 25 miles, or ” sprint ” to cover hill-climbs or short-distance record runs.
(1) The 250-c.c. Benelli engine, which is, of course, a motor-cycle power unit, is rather outside the usual sphere of MOTOR SPORT, but is included because it is an excellent example of the way in which the power output, judged on a swept volume basis, can be increased by reducing the size of the cylinders. While the Benelli’s piston speed is much the lowest on the table, being less than 75 per cent. of the generally accepted limit for racing engines of around 4,000 ft. per minute, and the brake mean effective pressure is by no means the highest, the horse-power per litre figure is the highest quoted. While the war has brought us 24-cylinder aero engines and 80-cylinder tank engines, the prospect of 24-cylinder 11-litres, and 48-cylinder cars to the 1938-39 G.P. formula is awe-inspiring, to say the least. (2) These figures refer to the II-type M.G. engine, fitted with the McEvoyPomeroy twin camshaft head, the development of which has been described in the war-time pages of MOTOR SPORT, and
it is unnecessary to add to the comments made there, except to record my belief that these figures apply to a state of tune for sprint events only, and that that power output could not be maintained throughout a long-distance race. (8) I have always been a fervent admirer of T. Murray Jarnieson’s brilliant twin overhead camshaft Austin, representing, as it does, the only racing engine planned ab iieitio as such, to be designed and produced in England in the last decade. The figures that I quote were given to me by the Austin Motor Company in early 1938, and are quite likely to have been improved upon during subsequent development. An interesting point that does arise is that if the blower pressure is assumed to be about 18 lb./sq. in., which is about the highest value that it can have in view of the use of a Roots blower, the b.m.e.p. of an equivalent un
blown engine comes out at 124 lb./sq. in. This leads one to suppose that had the Austin had anti-friction bearings throughout, it would have shown itself at least the equal in this respect of the 1i-litre straight-eight Delage, held by many to be the most efficient engine yet produced in cylinder head design and porting layout. Incidentally, should this engine be developed to run up to 12,000 r.p.m., as was its designer’s intention, piston speed will reach 5,140 ft. per min., a pretty staggering figure.
(4) The 1,100-c.c. 6-cylinder .engine fitted to Major Goldie Gardiner’s 200m.p.h. car is unique amongst engines of ultra high output in being without the advantage of an inclined valve head, in spite of which the mean effective pressures are commendably high. It is thought, however, that the engine is probably tuned to the ” nth ” degree and that it would not have sufficient stamina to last through even a short race. (5) Figures for the 1926 lit-litre Deluge,
in the form in which it was raced in 1936 by the late Richard Seaman, are included largely for the sake of comparison. Of chief interest is the equivalent unblown m.e.p., which at 134 lb./sq. in. has few rivals. This, I think, is due to three reasons. Firstly, to the Delage’s magnificent porting layout, which gains a lot from Lory’s decision not to use a detachable cylinder head. Secondly, the rather high stroke-bore ratio enables a 7.5 compression ratio to be used without having to have an odd-shaped combustion chamber, and, finally, the well-laid-out mechanical design ensures that a minimum of the power developed in the heads is wasted.
(6) The figures for the 1938 11-litre E.R.A. are interesting as showing what can be done in the way of developing an existing design, even though it has unsatisfactory features. In this case the small bore and long stroke put the piston speed way above the 4,000 ft. per minute figure, and the horse-power per sq. in. of piston area is again very high.
(7) The stroke-bore ratio of the 1939 E.R.A. is much reduced, though the piston speed is still rather above average. It will be noted that two sets of figures are given, the first being those quoted when the car was described in the Press and which, it is assumed, are those for sustained output. The second power output comes from a Raymond Mays article in the Autocar in which the ’89 E.R.A.’s output is described as 25 h.p. less than that of Mays’s black 2-litre, which is credited with 330 h.p. The speed at which this power is developed is my own estimate, and figures deduced therefore should be viewed accordingly.
It is interesting to note that the 1939 E.R.A. can sustain over a long race the same output as its predecessor could for a short race only.
(8) The 1i-litre Ma,serati is remarkable in view of the large power output obtained from the four large pots. The design of a piston 78 mm. in diameter transmitting 60 h.p. must be quite a problem. Four valves and four ports are used per cylinder in order to assist the induction and exhaust. Although this represents one of the most successful pre-war 1i-litres, it is difficult to foresee much development for the Maserati brothers until they return to a 6or 8-cylinder 1i-1itre.
Incidentally, a 3-litre 22-valve engine was also produced, giving much the same performance figures. (9) I quote two sets of figures for the
T 308 Alfa-Romeo in order to indicate the sort of difficulties that one comes up against. Although the power outputs are similar, the m.e.p.s vary considerably, due to the differing speeds. • In view of the use of a single-stage Roots blower, probably blowing at about 15 lb./sq. in., one can only say that either figure is possible, though I am inclined to think the former is more correct. (10) Unfortunately, my researches did not reveal any figures for the interesting 1 i-litre Mercedes, though I believe that the bore and stroke have been published and are remarkable—in fact, I think, so far unique in that the stroke-bore ratio is less than unity. , (11) The Alta figures are included as, if accurate, they show that more than a little is known of cylinder head design at Tolworth, as is indicated by the high equivalent unblown m.e.p.s, though these
must be aided by the low crankshaft speed at which maximum power is developed. Nevertheless, things augur well for the 1 f litre V8, with the possibilities of which Geoffrey Taylor was known to have been toying. (12) The Mercedes 3-litre gives a rather low equivalent unblown m.e.p., which indicates either that there is still room
for development or that the head and porting are not so efficient as they might be. (13) I do not know whether the 3-litre 16-cylinder engine designed by Ricardo for the 1940 Grand Prix Alfa-Romeo
exists except on paper, but in any case it is an extremely interesting project. The boost pressure, which is probably about 25 lb./sq. in., indicates a low equivalent unblown m.e.p. of about
90 lb./sq. in., due in some measure to the high engine speed and the consequent short space of time available for filling the cylinder. There appears to be, however, very considerable scope for development in
this engine, and thermo-dynamically it would appear that an output of 650 h.p. is within its capabilities. As I have stated before, it is difficult and, indeed, dangerous to attempt to draw far-reaching conclusions from the above. It does appear that the limits for engines for sustained output are at present about 290 lb./sq. in. m.e.p., and fairly well under 4,000 ft. per minute piston speed. It also appears that these two figures are seldom reached simul taneously. One fact does, I think, stand out, and that is that no engine yet
appears to have been built to equal the 1926 Delage as regards efficiency of cylinder head and porting layout.