L'Automobile Hispano-Suiza

PART II : 1912-1950

AFTER Zuccarelli’s victory in 1910, the Hispano-Suiza team was absent from the Coupe de l’Auto in 1911. The cause of this absence was primarily connected with the disorganisation occasioned by the move to the new Paris factory, but in any case a team was again entered for the race in 1912. The regulations now stipulated a maximum engine capacity of three litres, and an entirely new set of Hispano-Suiza racers were prepared in the utmost secrecy. Shortly before the race, however, La Vie Automobile declared that the engine had dimensions of 85 by 132 mm., and had developed 100 h.p. on the brake.

This output from a 3-litre engine represented something of a phenomenon in 1912, and by degrees it became known to the inquisitive that these surprising results had been obtained by the use of a supercharger. A well-known English journalist who disclosed the fact, however, suffered for his “scoop” by becoming persona ingratissima, as far as Hispano-Suiza were concerned, for very many years. In fact, Monsieur Birkigt and his colleagues remained exceedingly reticent about the new engine, and as the cars did not start in the race, considerable obscurity surrounds it to this day, although it steems that one of these cars, presumably without the supercharger, appeared at Brooklands in 1914. It is known, however, that the supercharger consisted of a double-cylinder reciprocating pump, located at the front of the engine and driven from a forward extension of the crankshaft. Moreover, according to Mr. Alec Ulmann, the well-known American Hispano-Suiza enthusiast, this pump “not only supercharged the mixture on the intake stroke, but also, by means of an extra valve located in the head of the T-head motor, provided a scavenging operation during the exhaust stroke.”

How were these extra overhead valves operated ? If I were really competent to write a history of Hispano-Suiza, I should, of course, know, but as in fact I cannot find anyone to enlighten my ignorance, can only fall back on surmise. In about 1913, while the French factory was otherwise engaged, the works at Barcelona began to manufacture a standard model with a bore and stroke of 85 by 130 mm., dimensions which strongly suggest that this engine was based on that prepared for the 1912 Coupé de l’Auto. As far as the racing engine is concerned, incidentally, the dimensions are noteworthy. In that long-stroke era, one would rather have expected the 80-mm. bore of the Alfonso engine combined with the maximum stroke permissibIe with the 3-litre capacity limit, namely 149 mm., as in the successful Sunbeam engine. Evidently, however, M. Birkigt expected extremely high crankshaft speeds from his supercharged engine, and, in order to reduce piston speed, chose the shorter stroke, 130 mm. (I do not believe it was really 132 mm.), of his 80 by 130 mm., 15-30-h.p. model, combined with the new bore of 85 mm., of which, incidentally, more was to be heard in the future.

But to return to the 85 by 130mm. “Barcelona” model of 1915 and subsequent years. In about 1927 or 1928, I drove one of these cars, which the man who was trying to sell it to me said was built in 1916. As in those days there was no Veteran or Vintage Club and age was in no wise considered a virtue in a car, one may take it that it was not built any later. According to my contemporary note, the engine had “eight overhead valves operated by an overhead camshaft driven by a vertical shaft” ; but unfortunately I did not buy the car, and I no longer have a clear mental picture of what the engine looked like. However, in 1922. the Autocar published a photograph of the current edition of this engine which, it said, “has a similar overhead valve layout to the larger [i.e., 37.2-h.p.] model.” A glance at the photograph, on the other hand, is sufficient to reveal that the overhead valve layout is, in fact, quite different. The 37.2-h.p. engine has the vertical drive for the camshaft and the camshaft itself on the longitudinal axis of the cylinders, whereas in the 80 by 130 mm. engine, the vertical shaft, while still at the front of the engine, is offset towards the exhaust side of it, and there is a gear-driven fan oil the centre-line. It is not clear where the camshaft is, but the engine has a flat instead of a domed top, and the camshaft may not be in line with its driving shaft. In any case, this engine looks to me very much like the derivative of the 1912 engine, with its “extra valve located in the head.”

The point is of some importance, because early in 1914, it seems, M. Birkigt took the next important step forward by redesigning the four-cylinder engine of 100 by 150 mm., to which reference has been made already. In the new design the T-head was abandoned altogether in favour of vertical overhead valves operated by an overhead camshaft direct, and perhaps to accommodate this valve gear, the stroke was reduced to 140 mm. I cannot pretend to be familiar with this engine, but Mr. Peter Dale, writing in Australian Motor Sports in December, 1948, declared that “this engine was identical in design to the 37-hp. model [of 1919] except that it had four cylinders, a three-bearing camshaft and a five-bearing crankshaft . . . The valve adjustments, the valve operation, the enamelling of the cylinder block and the transverse drive for the two magnetos were all identical.” I do not think, actually, that any 37-h.p., 100 by 140 mm., six-cylinder engine ever had two, or indeed any, magnetos, but apart from this detail Mr. Dale is, I am sure, a reliable witness.

It does not seem that this engine, which was said to develop 80 h.p. at 1,500 r.p.m., went into production in 1914, but, with the outbreak of war and the transfer of demand from motor cars to aero engines, M. Birkigt proceeded to mount two of these blocks on a common crankcase to produce a V8 engine which was perhaps the most successful of any used by the Allied air forces during the 1914-18 war, and which was built by fourteen firms in France as well as seven in England, America and Italy.

By the end of the war, the designer had at least projected a 12-cylinder edition, still with the same bore and stroke. It seems rather doubtful whether this engine was ever put into serious production, but one thing at least is certain: when the post-war Hispano-Suiza car was exhibited to an almost dumbfounded public at the Paris Salon in October, 1919 (less than a year after the end of hostilities), the engine was seen to be in effect one half of this projected 2-cylinder, and the direct descendant of those that had won so great a réclame in the air. The new car was shown at Olympia in November of that year and in March, 1921, the Autocar tried a test-chassis in this country, doing a timed mile at Brooklands, four up, at 75 m.p.h., and speaking very favourably of the top-gear performance, the acceleration, and particularly of the power of the servo brakes. It is also interesting, in view of my own experience of these cars, that they remarked : ” . . . although large cars are not always handy in traffic, this car can be turned sharply in and out around other vehicles in a manner quite surprising.” Incidentally, this chassis with box body weighed 4,247 lb., of which 1,974 lb. were carried by the front axle and 2,273 lb. by the back axle.

The new car immediately caused sensation. Gone were the light cars descended from the voiturette racers of the Coupe de l’Auto and the Catalan Cup ; in their place was an outstanding example of the voiture de grand luxe et de grand tourisme. The six cylinder engine, with its overhead valves operated directly by the overhead camshaft and a light-alloy cylinder block, enamelled under pressure, and having screwed-in steel liners, as in the aero engine, had a bore and stroke of 100 by 140 mm., and a capacity of 6,597 c.c. [It is believed that just before the 37-h.p. model went out of production, wet cylinder liners were used, sealed at the bottom by rubber rings.—ED.] The bore, it may be noted, was the same as in the 20-24-h.p. model of 1906, and the stroke the same as in the 40-h.p. of the same date. The new engine gave some 135 h.p., at 3,500 r.p.m. and propelled the car so rapidly that something quite unusual in the way of brakes was considered necessary to match it ; and at this early date, M. Birkigt adapted front-wheel brakes operated by a mechanical servo using a drum driven off the gearbox. To crown it all, the radiator cap now carried as a mascot a reproduction of the famous Flying Stork, emblem of the squadron commanded by the ace pilot, Capitaine Guynemer.

During the next decade, the 32-c.v., or 37.2-h.p. Hispano-Suiza attained a degree of chic which has perhaps never been equalled by any other motor car, and Pierre Frondaie even wrote a novel called “L’Homme à l’Hispano.” But readers of MOTOR SPORT are likely to be more notably interested in the mechanical features of this remarkable car, and at this point I have sought the assistance of Mr. George Briand, of 47, Tamworth Road, Croydon, who is, I believe, the only remaining Hispano-Suiza specialist surviving in this country from among the company’s once fairly numerous representatives. It is he who has kindly supplied much of the mechanical information used in the rest of this article. The earliest engines of this type, he informs me, had the rather unusual firing order. 1, 4, 2, 6, 3, 5, while in later examples the more conventional 1, 5, 3, 6, 2, 4 order was adopted. Correct valve clearance with the engine cold was 2 mm. The valve timing was decidedly “classical” and typical of a relatively low-speed engine, the following figures being accurate in terms of millimetres on the circumference of the fly-wheel, while the corresponding figures in degrees are approximations only : Inlet opens 35 mm. (8°) after t.d.e., closes 190 mm. (44°) after b.d.c.; Exhaust opens 160 mm. (37°) before b.d.c., closes 35 mm. (8°) after t.d.c.

This timing diagram which was very similar to that of the aero engine, with vertical overhead valves and a compression ratio of 4 1/2 to 1, produced an extremely flexible engine with a relatively flat power curve, giving remarkable acceleration and comparative reluctance to attain high crankshaft speeds. With the standard back-axle ratio of 16 to 54, however, the 37.2-h.p. Hispano-Suiza was one of the fastest standard models of the early 1920’s and could, moreover, maintain a high cruising speed for as long as the greatest requirements of le grand tourisme dictated. This performance was matched by the efficacy of the servo-operated brakes, which at that period were nothing short of a revelation, and perhaps even more notably by the handling qualities of the car. Few designers have been able to emulate Marc Birkigt’s ability to endow a large car with the “feel” of a small compact sports model, although if stability is to be maintained at high speeds it is usually necessary to pay rather particular attention to accurate wheel balance. The steering has a wheel movement of approximately 2 1/4 turns front lock to lock, and may be assisted by the transmission system, which comprises a short torque tube, with its spherical joint mounted on a cross-member of the chassis and an open propeller shaft between it and the gearbox.

The latter had only three forward speeds, the overall ratios, with the 16 by 54 back axle, being 9 1/2, 6 and 3 3/8 to 1, and, indeed, granted the characteristics of the engine, it is difficult to see what purpose a fourth would have served. They were selected, of course, by a right-hand lever working in an open gate, but the latter was placed forward of the normal position of an off-side front door, with the lever inclined backwards, a simple arrangement which, if more generally adopted, would have solved the problem of direct access to the drivers seat without resort to the unpleasant expedient of a central gearlever. The gear-change, moreover, attained a standard of “honesty” which, if more widely emulated, might well have resulted in the still-birth of syncromesh mechanisms. The lever and gate, together with the pedal layout and the arrangement of the minor controls, were almost exactly similar to those employed on the pre-1914 models, and, indeed appear to have closely followed the lines laid down for the first models of 1906.

It was in the electrical arrangements of the new chassis, however, that M. Birkigt, himself originally an electrical engineer, exercised almost his greatest ingenuity. Ignition was by two coils, with two plugs per cylinder, and the advance and retard was controlled by a centrifugal governor, an arrangement which at the time was uncommon outside aero-engine practice. There was also an over-riding hand-control, giving an exceptionally wide range of spark-timing, by means of which the engine could usually be induced to start, if there was gas in the cylinders, without motion being imparted to it from an outside source. “We recommend this practice,” declared the Instruction Book, in contradistinction to those of some other makers, “which is convenient and economises the battery charge.”

At the same time it was recognised that the starter motor would frequently have to be used, and more particularly when the engine was cold. It is in these conditions that coil ignition is at a peculiar disadvantage, because the effort to turn a stiff engine makes time maximum demand on the battery just when it is also required to supply ample current to provide a powerful spark at the plug points. In order to overcome this it was decided to use two batteries. At first, only one of these was of sufficient capacity, and provided with heavy enough wiring to operate the starter motor, the other being comparatively small, and used only for ignition while starting as well as being a standby for side-Iights and other light work. Later, however, two full-sized batteries were fitted, both of them, in some instances, wired for the supply of current to the starter, in which case two starter buttons were provided, and the whole of the original intention of the second battery could be defeated by pressing both starter buttons at once!

In spite of the high performance of the 37-h.p. model, which had a maximum speed of fully 80 m.p.h.—or perhaps more exactly because of it—there soon arose a demand for something more. For some time the works resisted this clamour, but in 1922 and 1923 it was decided to run cars in the Georges Boillot Cup race at Boulogne, and a special sporting chassis was prepared for the purpose. Considerable obscurity surrounds this whole subject, but it seems that the cars in question had a short wheelbase, which I have seen quoted as 11ft. 1 1/4 in., and the cylinder bores enlarged to 102.5 mm. In this form, the winning car in 1923 averaged 71 m.p.h. for 287 miles, a remarkably high speed on the difficult Boulogne circuit.

Almost simultaneously, in 1922 or 1923, there was produced what was known as the “Monza” model, although I cannot ascertain that it ever made a public appearance on the then new Italian track. This model also had a short chassis, which was moreover made lower by modification to the shackles, while in this case the cylinder bore was apparently increased to 102 mm. There is a strong tradition that the “Monza” engine had a stroke of 150 mm., and indeed this fact was clearly stated by contemporary authorities. Mr. Briand has, however, kindly taken the opportunity in recent weeks to measure one of these engines. and his found the stroke to be 140 mm. In the absence of any evidence to the contrary, therefore, the “long-stroke Monza” engine must be dismissed as a myth. Incidentally, it was a “Monza” Hispano-Suiza which showed up a Stutz “Blackhawk” to such disadvantage during a 24-hour run at Indianapolis after a wager between G. T. Weymann and F. G. Moskovics. The “Monza” model, in any case, had a limited existence, as in the spring of 1924 Andre Dubonnet started in the Targa Florio with an Hispano-Suiza fitted with a new-type engine. [It is said that he drove one of these cars from Paris to Nice in 12 h. 35 m. by way of practice.—ED.] This exactly resembled the 37-h.p. in general characteristics, but the bore, instead of being enlarged by a millimetre or two, was now definitely increased to 119 mm., making the capacity 7,983 c.c. With an R.A.C. rating of 45-h.p., the sports chassis fitted with this engine became popularly, and almost universally, known as the “Boulogne” model. It is officially stated, however, that this is a misnomer, and indeed it has long been something of a mystery to the present writer why a model which made its first appearance in 1924 should be called after the venue of a race in which Hispano-Suiza ran in 1922 and 1923. The explanation would seem to be, however, that the new engine was fitted to what had previously been known as the “Boulogne” chassis and thus popularly retained the designation of its predecessor.

In any case, for the next ten years or so, the 45-h.p. Hispano-Suiza attained rather the same prestige, in its different class, as had the Alfonso model before 1914. In some instances the compression ratio was advanced from 4 1/2 to 1 to 6 to 1, and while the genuine “Boulogne” and “Monza” models had a maximum speed of over 90 m.p.h., the 45-h.p. could exceed the magic figure of 100 m.p.h. In 1924, for example, Barnato’s 45-h.p. Hispano-Suiza took records at Brooklands, covering 300 miles at over 92 m.p.h. In recent weeks I have had the pleasure of driving a fairly early example of this model which is, however, in particularly good mechanical condition, and I recommend a similar experience to any reader who is of the opinion that his small sports model is likely to feel handier in traffic or on winding roads than an 8-litre car weighing a couple of tons.

While the Paris factory was thus catering for the luxury trade of the world, the original Barcelona works were utilised for the production of more modest motor cars. The 85 by 130 mm., four-cylinder model was continued into the early ‘twenties ; but about 1924 or 1925 its place was taken by a chassis with a six-cylinder engine having the same bore but with the stroke reduced to 110 mm., so that the stroke-bore ratio was approximately the same as in the 110 by 140 mm. engine of 1924, while the capacity at 3,750 c.c. was about half as large. This engine was an almost exact scaledown of the larger model, except that the block was of cast iron instead of aluminium, and the valve timing was quite different, the inlet valve opening 10 degrees early and the exhaust closing 20 degrees late, giving an overlap of 30 degrees instead of no overlap at all. The chassis was also on exactly similar lines. The engine was rated in England at 27 h.p., and as such, or as the “Barcelona,” this model has usually been known in this country.

I have myself been running an 11 ft. 6 in. wheelbase example of this model for the past two years, and its performance has proved exceptionally interesting. While unfortunately I have no comparative figures to rely on, I am inclined to think that up to, say, 50 m.p.h., even the 45-h.p. does not accelerate very much faster than the 27-h.p., but the latter achieves its performance by relatively low gearing, the normal axle ratio being 12 by 59, which restricts its maximum speed to about 70 m.p.h. and renders the engine much more prominent at normal cruising speeds than is the larger edition. The acceleration, braking and handling qualities of the car, however, combine to produce a vehicle which achieves averages which are somewhat astonishing in relation to its maximum speed.

By 1931, Hispano-Suiza had taken over the Ballot concern, and this union was celebrated, rather curiously since Ballot had originally been an engine specialist, by fitting an Hispano-Suiza engine to a Ballot chassis. This model was given the rather unattractive name of “Junior,” and as a result of the Ballot influence, the more conservative were shocked by the sight of an Hispano-Suiza with a central gear-change. The engine closely followed existing practice, with a bore and stroke of 90 by 120 mm., and a capacity of 4,580 c.c. Unfortunately, I have never tried this model and am completely ignorant of the character of its performance.

In any case, more important developments were afoot at this time. M. Birkigt, in fact, had come to the conclusion, round about 1929, that as a voiture de grand luxe et de grand tourisme, his 1919 design was now out of date. He therefore went back to his aero-engine of 1918, but instead of using one half of it, he determined to use the whole, and to produce a 12-cylinder car engine. The bore was still to be 100 mm. ; but in fixing the stroke he cast further back still, to the “square” six-cylinder models of 1907, and determined that the stroke should be 100 mm. also, the resulting capacity being 9,424 c.c. From the days of the long-stroke Alfonso engine, the wheel had turned full circle.

The most striking change, however, was in the valve gear. For some time it must have been apparent that, even with the greatest care in manufacture, an overhead camshaft was inherently noisy, and, presumably for this reason, it was now abandoned, in spite of its mechanical advantages. Instead a single camshaft was now mounted in the 60-degree V of the twelve cylinders, and operated the valves by posh-rods and rockers. At least the short stroke reduced the length and weight of the push-rods. In the earlier engines ignition was by four coils, but in later editions these were replaced by two Scintilla Vertex magnetos. The sump held 3 1/4 gallons of oil.

The valve timing was rather similar to that of the 37-h.p., except for the very much earlier opening of the exhaust valves, the diagram being as follows : Inlet opens 7 degrees after t.d.c., closes 45 degrees after b.d.c. ; Exhaust opens 60 degrees before b.d.c., closes 5 degrees after t.d.c. Valve clearances : Inlet 0.014 in., exhaust, 0.018 in.; carburetter jets : mains, 160 ; pilots, 70.

On paper the 12-cylinder engine in 1931 was perhaps mechanically less attractive titan was its six-cylinder predecessor in 1919, but the finish and cleanliness of design were again superb, and if there had still been any room for criticism, it would have been silenced by the performance. With a compression ratio of 5 to 1, the engine gave 190 h.p. at 3,000 r.p.m., and with a ratio of 6 to 1, the output was increased to 220 h.p. at the same crankshaft speed. In March, 1934, the Autocar tested a 54/220-h.p. chassis of this type fitted with Vanooren drop head coupé body and weighing approximately 39 cwt. The overall gear ratios were 5.44, 4.10 and 2.72 to 1, and I expect that the engine compression ratio was 6 to 1, although this is not stated. This car lapped Brooklands at over 95 m.p.h., and covered half a mile at 100 m.p.h. From rest, it attained 60 m.p.h. in 12 sec., and 80 m.p.h. in 19 sec., and from 30 to 50 m.p.h. it took 6 sec. in top gear and 4 sec. in second. Such figures can seldom have been excelled by a car of this type. Incidentally the price, like that of a 1950 Rolls-Royce, was £3,500 and the Cygogne Volante still rode proudly on the radiator cap. To drive a 12-cylinder Hispano-Suiza is for me a pleasure still to come ; but my friend, Mr. C. R. Abbott, tells me that it once fell to his lot to take one across London. He expected to feel rather like the captain of the Queen Mary during the process, and within a very short distance he found himself, on the contrary, tempted to emulate the proverbial minnow in the traffic stream. This feature, in a large car, is, to my mind, the ultimate test of genius in a designer. The 12-cylinder was ably introduced by Charles Faroux, who drove one from Paris to Nice and back as fast as possible, whereupon it was driven on to a large sheet of white paper in the Paris showrooms, not a drop of oil falling on to the paper.

Not unnaturally, the 12-cylinder soon displaced the larger six-cylinder models, and in about 1934 the Junior was also superseded by a new six-cylinder chassis. Known as the 39/120-h.p., the new engine combined the 110-mm. bore of the 1919 engine with the 110-mm. stroke of the 27-h.p. model, the resulting capacity being 4,900 c.c. In general design, however, it resembled the 12-cylinder, with push-rod operated overhead valves, but the timing presumably allowed for higher engine speeds, the inlet valves, as in the 27-hp., opening before t.d.c., while on this engine the dual coil ignition was retained. Unfortunately, as some may think, the Ballot influence not only resulted in the adoption of a central gear-lever on this model, but, also spread to the 12-cylinder, which lost its traditional Hispano-Suiza layout in this respect soon after its introduction. The Autocar tested a “30/120” with a Vanvooren pillarless saloon body in November, 1934. the car weighing 35 cwt. The overall gear ratios were 7.25, 5.40 and 3.65 to 1. The timed 1/4 mile came out at 82.95 m.p.h., and the tester estimated that the “normal or short-wheelbase version with 3.4 to 1 top gear would reach 90 m.p.h. As it was, 50 m.p.h. was attained in 12.2 sec., 60 m.p.h. in 19.6 sec., from rest, and from 30 to 50 m.p.h. took 8.0 sec. in top gear and 5.5 sec. in second. The price of the complete car, incidentally, was £1,895, or about that of the basic cost of to-day’s 2 1/2-litre Lagonda.

At Barcelona, moreover, there was still a demand for something simpler than the Paris products, and at about this same date the 27-h.p., 85 by 110 mm. Overhead camshaft six-cylinder model was replaced by a chassis having a six-cylinder engine rated at 20 h.p. The dimensions of this engine, which had a capacity of 3,016 c.c., were 80 by 100 mm., the stroke being the same as in the 12-cylinder, and the bore harking back to the days of the Alfonso model. This engine was of the new push-rod type, but the model remained practically unknown in England.

Before the outbreak of war in 1939 Hispano-Suiza had practically abandoned the manufacture of motor cars, and the first post-war Paris Salon revealed no successor to the famous 37-h.p. of 1919. And yet the lying jade declares that it is on the way. Its character, however, is still shrouded in mystery and those who know won’t tell, while those who tell probably don’t know. All the same, it may have been gleaned from these notes of Hispano-Suiza history, that M. Birkigt is not a man who fails to profit by his experience. In the 1914-18 war he used to build eight-cylinder aero-engines ; he does not seem to have produced an eight-cylinder car so far.