THE TRUTH AFTER EIGHTEEN YEARS
J. L. Wyer finds some authentic data relating to the 1921-26 Sunbeam-Talbot-Darracq racing cars
Ti IE recent series ol art ales ill MOTOR SPORT dealing with the 200-mile races in the early nineteen-twenties have. I suppose, drawn all our minds back to that vivid phase of British nicker. not the least, outstandiar. feature of which seems at this (late to have been the wide diversity of entries, hoth amateur and professimad. It is, to ale at least, of surpassim, interest to have recalled in this way tlw salima points of these earlier ears, and to ninsi upon the hopes which went into their preparation.
It is interesting, too, to be reminded again of the dominant position which the Talbot-Darracqs, or. more properly, the Sunbeam-Talbot-Darracqs, ocelipkd in this period. They, one feels, were the real professionals. Their appearance in any event automatically raised it to it higher level of importance, and it is hardly too mueb to say that they occupied the same positioa. in relation to their eomial it ors, as Ferrari in the early nineteen-thirties. or the German teams in the immediate pre-war years. All this gives added point and interest to some data which (‘nine into my possession quite recently. and which gives a faseinating glimpse not only of the 1.’,litre Talbot-Darracqs, but of the whole of S.T.D. racing development during the period 1921-1920. With Antlarny !feat. I made a pilgrimage to the old Sunbeam works at Wolverhampton, till’ whole expedition forminga small . part of Anthony’s indefatigable quest for anything relating to, or bearing upon. the Sunbeam racing cars. ()I’ NVIlit’ll his viii mil monopoly has yet to be attacked by the daily Press. We had hopes of obtaining working drawings of parts which might need reidaeement, and even, possibly, that some of the parts themselves iniAt be lying in a disused corner. In this we were singularly unsuccessful. The racing car spares, of Nvhich there were very larfre quantities, were reduced to scrap metal, by order of’ the Booty!: organisation, as recently as 191.2, and all the drawings had gone at the same time for paper salvage. Such patriotism, we felt, was possibly a trifle misguided. From my own point of view, however, the trip was not entirely barren, since I renewed acquaintance with a number of old friemls, and, in particular. .1(dm Dabbs, quondam fellow draughtsman in the Sunbeam drawing °Mee. itow ehief desimer of Sunbeam. mumercial vehicles. After smut’ cogitation, Dabbs r:amanhered he had some power curves and other data relating to tla. racing ears somewhere aiming his effects, and in due course he sent then) ou to me. Quite what prompted Dahby ” to preserve these fragments fn an Ill.. holocaust is !moldy
matical. but it is fortunate he did, since they may well be the only authentic data of the kind still in existence. As far as possible I will try to reconstruct something of the S.T.D. racing picture, and the first ears to be dealt with in natural chronological order are the 4-cylimler Talbot-Darracqs. There are only two curves dealing with these highly successful little cars, but one of these is of great intere:t. showing plotted on the saute ,Yraph the power curves for all three types of engine used—the 1921 10-valve jobs. t lie 192:1 8-valve, and the 1924. supercharged version of the 8-valve engine. Incidentally, in a recent issue of Murcia Seoul. the Editor mentioned that there were three entirely separate designs involved, and while this may be so,
have always understood that the 1921. cars were a sill wrcharged and ” soupedup ” edition of t he 1923 machines, in the same way that the 1923 Grand Prix Sunbeams were built into ” the 1021. cars. [It is perfectly true that the 102:3 67 by 105 mm, ears were virtually the same as the supercharged 1924-5 type, but certain outward changes and a vastly higher power justified the division. –En. I
Be that as it. may, here are the respective figures. The 1921 16-valve engine produced 53 b.h.p. at 4,000 r.p.m.. at which speed the curve was flattening out. Maximum b.m.c.p. was 122 p.s.i. at :3,000 r.p.m.. with 110 at the horse-power peak. Two t laudel-lIobson carburetters were used, with 28-mm. chokes, and the compression ratio was 6.40 to 1. ‘The cylinder block and head casting of this engine was, ()I’ course, in aluminium. Bertarione’s I 923 8-valve design, built on the well-known welded-up principle, with thin sheet steel water jackets on cylinder barrels machined from individual
billets, %vas at considerable improvement. Maximum power shown is 65 bit.p. at 4,500 b.m.e.p. curve VMS absolutely flat at 135 p.s.i. between 2,500 r.p.m. and 4,000 r.p.m., falling to
125 at the horse-power peak. Compression ratio was 7.12 to 1, and the two Molex carburetters used 33-mm. chokes. The output from the 1924 supercharged engine is less than might be expected. In the September, 19-14, issue of MoToa Sewer dealing yin the 1024 200-Nlile Race, I see the figure quoted is 108 b.h.p. at bet veen .1,001) and 5,500 r.p.m. This curve, however. gives the engine credit for only 92 h.h.p. at 4,500 r.p.m., and while the curve is certainly still rising at this speed. and it is very probable that readings were taken at higher revolutions, a projeta ion of the curve would not
realise t ligure mentioned. The ban.c.p. curve rises fairly steeply to 190 p.s.i. at 3.:;00 r.p.m. and falls to 180 at •1..500. Compression ratio was now 0.10 to 1. ‘Hie carburetter is quoted as a 1924 GI,. sotex, with .t I-non, choke, mud was probably the same type as that. used on the 2-litre Grand Prix Sunbeams. No details are given regarding supercharger pressure or speed, but. the supercharger is quoted as ” 1925 compressor 130 by 128 min.” Ineident ally, the date of this particular curve (No. :371) is April 9th, 1920, and coining so late in the life of the eie,ines may, I think, he taken as ” best ever perfiwinanees.
The only other curve relating to the I-cylinder engines is dated April 12th. 1921., and shows the 8-valve unsupercharged engine converted for marine I urposes. Compression ratio is given as 7.06 to I, hut no ot her details are supplied of any modilicat ions. The power figures are identical with those already quoted. and the curve is of no special interest.. Conteniporary with the 10-valve TalbotDarracqs were flue 8-cylinder :3-litre Sunbeams ‘vhich ran at Le Mans in 1921, and the Isle of Man in 1922. Never partictifatly brilliant. there is very little information relating to these ears. One curve, dated April :iiith, 1920, naively headed ” facer,” and, in brackets, “SI ockholm Marine,” gives the maximum power as 98 h.h.p. at 4.000 r.p.m. Another curve. already in my possession, credits this engine with 108 b.h.p. at I he same speed. and since no details are given of the modifications for marine work, the curve is of no particular interest. Two other curves show the amplitude of crankslitift vibration for this engine. ‘rests were made with the flywheel flange only, weighing 13 lb., and with from one to four flywheels added. Beyond showing a shattering vibration period at about 1,800 r.p.m. tlwrc is not much to be
learned. The vibration was least without any flywheel, and worst with only one flywheel. For comparative purposes, the 30-h.p. 8-cylinder production car is shown. This engine had a bad period at 1,300-1,450 r.p.m., but of only half the intensity of the racing engine. These curves were made in November, 1926, and were possibly for the purpose of providing a comparison for the production car. There is, naturally, a good deal more data relating to the 2-litre 6-cylinder Grand Prix Sunbeam, most spectacular of all S.T.D. racing cars. The earliest curve is only a valve-timing diagram, which I will deal with later. The next curve, in date order, is of intense interest, showing, plotted on one graph, power curves for the engine in the following forms : (a) unsupercharged, and as it won the French Grand Prix at Tours in 1923; (b) =supercharged, but with the compression ratio reduced to the level of the later, supercharged version ; (c) supercharged through the carburetter ; and
(d) supercharged, drawing from the carburetter, in the form in which it was always raced. This curve is, I think, important for two reasons ; first because it gives authentic figures for the rather elusive 1923 Grand Prix-winning engine and, secondly, because probably it is not generally known that Sunbeam experimented with supercharging through the carburetter before adopting the more conventional layout.
The following figures were obtained : (a) It appears that 102 b.h.p. at 5,000 r.p.m. was good enough to bring Sunbeam their highest reward in racing–rather less than has sometimes been claimed, although the curve is still rising. Compression ratio was 7.3 to 1, and as is fairly widely known the cars ran on =doped petrol or petrol-benzole. Maximum b.m.e.p. was 137 p.s.i. at 4,700 r.p.m., falling away slightly towards the horsepower peak.
(b) With a compression ratio of 6.2 to 1, as originally used for the supercharged cars, the power fell to 86 at 5,000 r.p.m., and the b.m.e.p. to 121 p.s.i. at 4,200 r.p.m., falling away considerably at the higher speeds.
(c) Supercharged, but with the carburetter between the supercharger and the engine, maximum power recorded is 116 b.h.p. at 5,200 r.p.m., although the curve shows little sign of falling off. B.m.e.p. rose steadily from 130 p.s.i. at 1,000 r.p.m. to 153 at 4,200, and fell to 147 at 5,200.
(d) With the carburetter on the suction side of the blower, power rose to 138 at 5,500, and the b.m.e.p. to 179 at 4,300, falling to 170 at 5,500. Compression ratio for curves (c) and
(d) was 6.2 to 1. The date of this curve is May 12th, 1924, and in view of the proximity to the French Grand Prix it is reasonable to suppose that little or no further development took place before the race. 138 b.h.p. was, I think, the figure claimed for the Alfa-Romeos that year—an interesting fact, as although there was little to choose between the cars, the advantage in speed appeared to lie with the Sunbeams, or was it only Segrave’s driving which gave them the lap record at Lyons ? The next curve, dated December 19th,
1924, takes us a stage further in the development of the 2-litre engine. Here five curves are plotted, and I will deal with them in the order in which they appear on the drawing. (a) Carburetter is quoted as Solex with 44-mm. choke, magneto as Carbon Brush HL60, plugs as K.L.G. 180, compression ratio as 6.3 to 1, and compressor pressure
5 lb./sq. in. The maximum power shown with this combination is 130 at 5,000, still rising steeply. Apart from the fact that this curve terminates at 5,000 r.p.m. it is practically identical with the May figures given in (d) above.
(b) All details identical with (a) except that the magneto is now quoted as Jump Spark FHA and the plugs K.L.G. C.B. It is notable that this modification alone raised the power to 138 at 5,000 and 150 at 5,500.
Presumably the FlIa6 magneto was the B.T.H. subsequently used on these cars, but it is interesting to speculate if the III460 is the Bosch used at Lyons in 1924. Perhaps someone can come forward and confirm this point.
(c) Details are identical with (b), except that the compression ratio has been lowered to 6 to 1, and the power fell accordingly to 128 at 5,000. (d) The compression ratio is still 6 to 1, but the compressor pressure is raised to
6 lb./sq. in., and the carburetter is now quoted as “Sunbeam 47 mm. choke” (actually a modified barrel-throttle Solex). The curve now goes up to 153 at 5,500.
(e) One feels that, at this stage, Capt. Irving and the rest of the Sunbeam backroom boys really decided to lead out trumps. The ” Sunbeam ” carburetter and the 47-mm. choke were retained, together with the FHa6 magneto and K.L.G. C.B. plugs, but the compression ratio went back to 6.3 to 1 and the blower pressure up to 6.5 lb./sq. in. This sent the power up to 162 at 5,500, the best figure recorded in this series. This curve, however, is almost identical (except only for the use of a 48-mm. choke) with another curve already in my possession, and which—apart from a curve taken on E2 fuel, which gave 171 b.h.p.—is the best I have ever seen recorded for this engine.
It would seem there are two conclusions to be drawn. First, that very considerable development took place between the 1924 Grand Prix, when the car was, by common consent, the fastest road-racing machine in Europe, and the 1925 race at Montlhery, when it had decisively lost that advantage to AlfaRomeo and Delage. And, secondly, that this ultimate stage in the development of the engine was reached very much earlier than has sometimes been thought.
B.m.e.p. figures are given only in connection with curve (e), and show the characteristic b.m.e.p. for this engine, rising sharply almost throughout the speed range, and peaking at 197 p.s.i. at 5,000 r.p.m. Another curve, dated July 8th, 1925, shows the effect of different pistons and compression ratios. All the power figures terminate at 5,000 r.p.m., and so do not show the absolute best of which the engine was capable, but, nevertheless, they provide an interesting comparison. With a compression ratio of 0.15 to 1, aluminium pistons weighing 7 oz. 11 drms., gave slightly better results than electron pistons weighing 5 oz. 14 drms. (145 against 143 b.h.p.). Rather surprisingly aluminium pistons weighing 8 oz. 3 drms., and giving a compression ratio of only 6.05 to 1, gave still further improvement,
to 147 b.h.p., and finally aluminium pistons, with the weight down to 7 oz. 11i drms., and the compression ratio up to 6.375 to 1, gave only 148 b.h.p. Presumably the increased ability of the heavier piston to absorb heat very nearly balanced the greater inertia losses and drop in compression ratio, and it
would be interesting to know if figures were taken with pistons both heavier and giving a higher compression. Other data relating to the engine used for this series of tests were : Shell fuel,
150-mm. supercharger, and 54-mm. Solex carburetter set 48 choke, 250 main jet, 85 auxiliary jet. The only remaining curves relating to the 2-litre engine deal with valve timing, and here I will go back to the earlier curve mentioned above, which is dated April 3rd, 1924. This curve (No. 259) is headed “2-litre, 1924 valve lift diagram” and gives the following figures : inlet opens 100 before t.d.c., closes 29° after b.d.c., period 219°, maximum lift 7.25 mm. Exhaust opens 22° before b.d.c., closes 14° after t.d.c., period 216°, maximum lift 7.35 mm. The overlap was
thus a modest 24°. Then we have a much later curve, dated February 15th, 1926, headed “Comparison—Standard and Lowlift Camshafts,” but the “Lowlift ” camshaft shown is, in fact, identical with the 1924 timing quoted from curve No. 259. The ” Standard ” (or highlift) camshaft gives the following timing : inlet opens 10° before t.d.c., closes 52° after b.d.c., period 242°, maximum lift 8.95 mm. Exhaust opens 52° before b.d.c., closes 14° after t.d.c., period 246°, maximum lift 9.05 mm. Thus, while the periods and lifts for both inlets and exhausts were considerably increased, the overlap surprisingly remained unchanged. The reference to “Standard ” for the highlift camshaft. is puzzling, as is the fact that experiments were still being made in February, 1926, after the effective Grand Prix life of the cars was over, and some_ considerable time before they were run at Brooklands, where they did not appear until 1927.
The whole thing becomes a great deal more obscure when we consider the power figures provided by the various camshafts. With lowlift inlet and exhaust the maximum power shown is 100 b.h.p. at 4,500 r.p.m., an absurdly low figure for the supercharged engine. With lowlift inlet and highlift exhaust, the power goes up to 108 at 4,750, with highlift inlet and lowlift exhaust to 112 at 4,750, and, finally, with both highlift camshafts, to 138 at 5,000. The only conclusion one can draw from all this is that, notwithstanding the 1924 curve, the ” Standard ” or highlift, camshafts were employed throughout the development of the engine, which leaves one speculating as to why Sunbeams should have apparently been interested in detuning the engine at this stage. Is it conceivable t I t t, they were toying with the idea of prod tieing a sports version of the supercharge( i 2-litre—surely an idea
fascinating in its possibilities, when one reflects upon the effect it might have had upon sports-car races of the late ‘twenties and early ‘thirties ? There is still one curve dealing with the S.T.D. racing cars, the last of the line, namely, Bertarione’s brilliant but erratic Ii-litre straight-eight Talbot of 1920. This curve is dated April 23rd, 1926, comparatively early in the development of the engine which was still far from au point when it first appeared later in that year. It was, it will be remembered, extremely fast but very unreliable, and never attained that pitch of perfection which later allowed Delage to carry all before them. Two interesting sets of figures are given for b.h.p., b.m.e.p. and imluction pipe pressure and temperature. For the first, the compression ratio was 6.5 to 1, the fuel 11.P. “Special,” the
choke tube 49 mm. with a 400 jet, and the induction pipe diameter 55 mm. Maximum b.h.p. was 140 at 7,000 r.p.m., induction pipe pressure 14 p.s.i., and the temperature reached 105° C. The b.m.e.p. peak was 198 p.s.i. at 5,000 r.p.m., falling away to below 180. For the second curve, the compression ratio was raised to 7.0 to 1, fuel was the ethanol base E.2, a 55-mm. choke was used with 500 jet, and the induction pipe diameter increased to 59 mm. Improved breathing reduced the induction pipe pressure to 13.5 p.s.i., and temperature to 50° C., aided, no doubt, by the alcohol content of the fuel. Power went up to 145 b.h.p. at 6,500 r.p.m., and b.m.e.p. to 208 at 5,400 r.p.m. Little seems to be known, except by inference, about the ultimate performance of this engine, but assuming that the breathing was improved still further, the second curve can be projected to well over 160 b.h.p., a not improbable figure.
That is all the information that remains in comiection with the racing ears, all there is to show for those years of intensive work and the expenditure of many thousands of pounds. I have given the bare figures, without attempting to draw any very weighty conclusions or to point any morals. If some of the figures quoted are disappointingly low when compared with those which have sometimes been claimed, there is nothing so very unusual about that. All these curves bear the stamp of the Sunbeam Experimental Departnlent, who were less prone to exaggerating their conclusions than the Publicity Section, but were unlikely to have suffered from sham modesty in relation to their product. There is also a n tt her of curves relating to the production 3-litre car, giving valve timing and power figures for various carlat rct I ,_er set t hugs. I do
not feel I hey are of sufficient interest to he quoted in full.