71 replies to this topic

[Roger Clark]

Maybe they didn't notice- it may have been after scrutineering, and it was a French team.

An interesting point, but I was asking where the pistons came from.

[fivestar]

From my reading various articles, the original engine #108 C, seized during practice and replacement pistons were ordered from the factory and delivered. A grinding shop was found to clean up the bores. I would guess that the "oversize" pistons would probably only have been +.254mm to max .762mm oversize. Not likely to have made much difference to the total cubic capacity.
rgds

.+

An interesting point, but I was asking where the pistons came from.

Edited by fivestar, 17 March 2012 - 12:01.

[Roger Clark]

From my reading various articles, the original engine #108 C, seized during practice and replacement pistons were ordered from the factory and delivered. A grinding shop was found to clean up the bores. I would guess that the "oversize" pistons would probably only have been +.254mm to max .762mm oversize. Not likely to have made much difference to the total cubic capacity.
rgds

.+

Thanks for that, although the top end of your guess would give a total capacity of closer to 3.4-litres.

Is it common for piston sizes to vary by that much?

[fivestar]

Oversize pistons are normally rated at:-+.010,+.020,+.030 and +.040. I did a metric conversion which could be wrong.
I have a Type 57 - 57S Maintenance & Overhaul Manual which states the block can be rebored to a maximum of 1mm.
rgds

Thanks for that, although the top end of your guess would give a total capacity of closer to 3.4-litres.

Is it common for piston sizes to vary by that much?

[D-Type]

Oversize pistons are normally rated at:-+.010,+.020,+.030 and +.040. I did a metric conversion which could be wrong.
I have a Type 57 - 57S Maintenance & Overhaul Manual which states the block can be rebored to a maximum of 1mm.
rgds

Presumably in metric countries these would be 0.25, 0.5, 0.75 and 1 mm.
I have found the the bore and stroke of the T57 quoted as 72 x 100mm giving 3257cc. Going with the original (imperial) figures and increasing the bore to 72.762 (3rd oversize) gives 3327cc.

Incidentally I also found these snippets in Anthony Pritchard's "Sports Racing Cars". Referring to the 1936 cars he says:

... Industrial unrest in france resulted in the cancelletion of the 24 hours' race but there were still three events in which these cars could compete, including the 621 mile (1000km) French Grand Prix to be held at Montlhéry on 28 June.
The Bugattis, officially known as the 57S but known as the 57G at the factory, were in bclear breach of the French Grand Prix rules which demended that cars be of a type of which 20 had been constructed or laid down by prior to June 1936. The other French manufacturers, Delahaye and Talbot,also broke the rules but not so extremely. Bugatti built a car that was based on the Type 57S, but in reality it was technically more closely related to the Type 59 Grand Prix car less the supercharger.

He later says

In 1939 Bugatti entered a single car at Le Mans, a version of the production supercharged 57C. The chassis was much-modified, the supercharged 57C engine had been tuned to develop around 200 bhp, the gearbox was 57S45 and the body was a smoother, cleaner version of the 57G. An unusual feature was aspotlight recessed in the body on the right- hand side and directed at the verge.

Edited by D-Type, 19 March 2012 - 08:51.

[fatbaldbloke]

Last week I was speaking with someone for whom these cars is an all involving passion which he has been studying for 40-odd years. He has written a book on the subject but as some of his findings don't agree with the approved history he is finding it a little difficult to find a publisher at the moment and may have to self publish.

[Jon Petersen]

Off topic really, and no excuse sought for that.

But - to me, the real Bugatti Tank is the 32.

The 57s and G were "just" sportscars looking pretty much ahead of their time - the 32 Tank was out of this world, and looked more like someting from a 1920s Science Fiction film....


Regards

Jon

[cpbell]

Off topic really, and no excuse sought for that.

But - to me, the real Bugatti Tank is the 32.

The 57s and G were "just" sportscars looking pretty much ahead of their time - the 32 Tank was out of this world, and looked more like someting from a 1920s Science Fiction film....


Regards

Jon

Not sure I'd agree that the 57s weren't real tanks, but the 32 certainly had an HG Wells feel to it, IMO. Strangely, to me, the T30 always looks slightly scary in a weird way to me with its Strasbourg body (which according to Conway in Grand Prix Bugatti was made by a Strasbourg sheet metal workshop at the insistence of Pierre de Vizcaya, despite Ettore wanting to use a Brescia-style body with exposed bolster tank. In comparison, the T32 has an exciting, futuristic appearance, which belies the fact that it was by all accounts an awful car to drive.

Edited by cpbell, 19 March 2012 - 20:25.

[fivestar]

Now that i have found the key to my display cabinet, attached a couple of photos of the T57 Tank models I nhave from the early 1970s. Quality nothing like todays ready made IXOs.



Uploaded with ImageShack.us



Uploaded with ImageShack.us

[MTAnorak]

I have noted many of the questions and answers on this bulletin board about the Bugatti
Type 57 Tanks. If the following about the Type 57G 1936-37 is of interest, I shall be happy
to continue covering the 1939 Le Mans winning car, the 1937 T57S45 Tank, as well as
other T57Gʼs.
Part 1: Type 57G 1936-37
There are several books available about the successes of the T57G Grand Prix sports
cars, yet there are massive gaps in the story of their evolution. I have had a passionate
interest for the furious beauty of the Bugatti ʻTanksʼ since 1971 and have developed my
knowledge of them along with strong convictions about their development. Some of the
following opinions do not necessarily agree with more formal Bugatti historians or their
publications. Nevertheless, this is my interpretation of the Type 57Gʼs known data, history
and development, along with observations and a little analysis of Bugattiʼs 1934-35
competition record which lead to these cars. For a technical overview of the racing T57G,
it is important to recognize its origins as well as to understand its siblings.
The T57 touring model was the progenitor, with a wheelbase of 3.3 metres and a track of
1.35 metres. The T57S was a lowered and shorter sports car version of the T57, with a
wheelbase of 2.98 metres and track of 1.35 metres. Its low slung chassis frame was
characterized with deep sections at the rear, permitting its rear axle to pass right through it.
Its competition sibling, the T59 was a Grand Prix racing car with a wheelbase of 2.6
metres and track of 1.25 metres. The T57 and T59 were designed concurrently from 1932,
and the T57S in 1935. They shared a clearly common engine design architecture, yet each
version is different in many details and materials. The T57 engine has a wet sump
lubrication system, whereas the T59 and T57S a dry-sump version for improved
performance. Now we come to the origin of some confusion.
In all Bugatti publicity material the ʻTankʼ Grand Prix sports cars of 1936-37 were always
known as T57S, whilst their factory designation in documents and records was T57G. The
ʻGʼ was originally ʻGPʼ for Grand Prix. That is to say, this particular T57S was the Type 57
Grand Prix racing version of it, sharing its lowered chassis on a wheelbase of 2.98 metres.
It used the T57S hollow front axle which was specially adapted to accept GP T59 wheels
and brakes, where likewise, the T57S rear axle was also modified to accept the GP T59
wheels and brakes. My personal opinion regarding all Bugatti Type numbers, is that they
should be considered as project numbers. Thus, ʻT57GPʼ almost certainly began as a
racing engine project linked to the T57S project, where its ʻGPʼ designation in the drawing
office could embrace any additional parts specially required for its marriage to the standard
T57S chassis and T57S engine, such as special material valve seat, or an increased
capacity radiator etc. The ʻSʼ in T57S probably meant ʻsurbaisseʼ - ʻspecially loweredʼ,
although some believe it stands for ʻsportʼ. On the other hand, in 1937, Bugatti ran a T59-
framed, T59/57G-powered car with a chassis number 57248. This car was formally
designated as a T57 ʻSportʼ, so it seems unlikely that the ʻSʼ of the longer wheelbase T57S
was intended to be ʻSportʼ. The T57 and T59ʼs respective engines however, although
originally being different in capacity, firing order and component materials, had a generally
identical design pattern. As a result, there was always an inherent potential for some
degree of crossover of individual parts between them, even though the chassis design of
the three was distinctly different, as too their respective crankcases; wet or dry sump. I
strongly believe that Jean Bugatti - director of design during the creation and development
of the T57, T59 and T57S - considered all three cars as ʻT57ʼ variants.
1
A historical overview is also important to consider. During 1934, it was apparent that the
performance of Bugattiʼs T59 Grand Prix cars was no match for the technically superior
state-sponsored German racing machinery. This recognition and frustrating acceptance
got to a point where Bugatti disbanded the racing team with conspicuous publicity at the
end of the year, including selling off four of the nine, now clearly redundant, Type 59 racing
cars already built. For 1935, Jean Bugatti decided to concentrate his limited resources on
the development of a super sports car to replace the outdated Type 55. The outcome was
the creation of a short wheelbase and technically refined version of his 3.3 litre Type 57
touring car; - the Type 57S model. Traditionally, the House of Bugatti would always ensure
there was a more competitive version of their road-going sports cars; hence the T57G.
Consequently, the ʻGʼ engine was a ʻGrand Prixʼ version of the T57S power unit, where its
refinement evolved as a hybrid, combining the very best characteristics of the T57 and
newly redundant Grand Prix T59 engines; albeit on an extremely limited budget.
The prototype T59 began in 1932 with a 2.6 litre power unit, given a bore and stroke of 68
x 88mm, but for production this was altered to 2.9 litre using the T57 engine block with its
bore of 72mm along with the 88mm crank. In 1933 however, the T59 crankcase was
cleverly redesigned, simply by tilting the previously vertical oil pump and drive outwards to
provide clearance for a 100mm stroke crank throw. This modification needed only the
minimum of change to the original wooden engine block and sump patterns. By now, the
3.3 litre capacity of the T59 engine was effectively the maximum the design would ever
permit, where realistically, increased power could only ever come from higher engine
revolutions. Unfortunately, as engine speeds increased, racing mechanicsʼ note books
record crank and block breakages, and subsequent replacements. Retrospectively, the
reason is relatively straight forward. The T59 crankshaft utilized a 4-4 array whereas the
T57 used a 2-4-2 version; both having distinctly different firing orders, yet the 2-4-2 was
much smoother running. The 4-4 crank layout originated from Ettore Bugattiʼs first
successful straight-eight, by coupling together two of his Type 13 four-cylinder engines in
1912, placing each crank plane of four at 90 degrees to each other. This arrangement
worked extremely successfully at the relatively modest engine speeds used during the
1920ʼs, but as speeds began to exceed 4,000 rpm for prolonged periods, a secondary fore
and aft rocking couple placed increasingly excessive stress on crankshafts and engine
blocks; draining power too. The mechanicʼs note books indicate breakages in areas
consistent with extreme fore and aft rocking couple forces. These mainly take the form of
near-centre journal crank breakages, and/or cracking of engine block casings by twisting,
as well as sometimes breaking the camshaft drive tower lower flange.
For Jean Bugatti, technically more sensitive to modern developments than his father, the
smoother running of the 2-4-2 crank of the T57 was an obvious improvement which would
enable higher engine speeds. It seems that rebuilding them using specially lightened T57
pattern crankshafts along with high-lift profile T57 camshafts, prevented further breakage
problems. It is possible that the first prototype T59 engine to be fitted with a 100mm stroke
crank was modified using 2-4-2 T57 specification components, and so became an
example found capable of revving higher with reliability. Significantly, it is apparent that as
early as mid-1934, lightened Type 57 pattern crankshafts were being successfully fitted as
replacements into some of the Type 59 racing engine crankcases. We may see therefore,
that the original T59 racing power unit had incrementally evolved into a T57 specification
engine within a T59 dry-sump crankcase. Effectively, by the end of 1934, the embryonic
T57S and G engines had almost arrived in the latest form of T59 engine. As an aside, the
2-4-2 crank array of the T57 was so smooth running and reliable at high speeds that Rolls-
Royce chose the 2-4-2 arrangement for their own straight-eights in the Post-War period.
2
Interestingly, during 1935 and having officially abandoned racing, Bugatti entered just one
or two of his cars in numerous competition events throughout the year. Externally, these
were either GP T59 or T57 Grand Raid types. The latter version was a long-wheelbase
T57 with a tuned engine, having its radiator and bonnet line lowered by 80mm, and its
steering column dropped down correspondingly. A version of the GR included the Ulster
TT entries, although for the Brooklands BRDC 500 Mile Race for normally aspirated cars,
a T59/57 (G prototype?) powered GP T59 appeared. Whilst looking just like a Grand Prix
T59, it was formally entered as an unblown Type 57, so we can justifiably assume it had a
chassis number to comply with British import regulations. A T59-framed car exists with the
chassis number 57248, this being a T59/57G hybrid, known as the T57 ʻSportʼ or ʻKing
Leopold Carʼ. In addition to cylinder head and manifold refinement, we find special material
billets for crankshafts and camshafts, as well as high strength steel connecting rods, and
Rolls-Royce specification alloy aluminium components. Over a period of twelve months,
these competition engines had altered significantly in their material parts; far more suitable
for higher revolutions.
That these engines and associated components were being systematically modified during
1935 can be assumed from Bugattiʼs employment of the talented development engineer,
Piero Taruffi to drive in several races. By now the increased speeds and power demanded
a refinement of the standard T59 magneto, which was previously driven from the left-hand
camshaft. The result was a revised camshaft drive tower where the magneto drive came
from a sprocket in the cleavage between the camshafts; being directly powered from the
intermediate camshaft drive gear. This meant that the magneto would now run at engine
speed, being twice that of the original, providing a much more powerful spark. Also, being
closer to the rotational stability of the flywheel, it was less susceptible to torsional vibration
at higher revolutions. This special modification enabled precision with increased ignition
flux, more suitable for developing greater power through higher engine revolutions. With
greater power came increased engine heat, so a bolted steel top plate was fitted to the
cylinder head to ensure free passage of coolant around the valves and combustion
chambers, as well as increased coolant capacity above them. Some T59 engine photos
show seven-branch coolant risers; a modification which could only be achieved with a
bolted top plate. Additionally, an increased capacity oil cooler also helped to maintain a
reliable engine lubrication temperature. By the beginning of 1936, Bugatti had evolved a
reliable racing version of the Type 57S dry-sump engine - the T57G - which even in
unblown form, matched the road-going performance of the original supercharged T59!
The T57G super-streamlined ʻTankʼ made its maiden appearance at Montlhery for
functional and performance testing on the 8th and 9th June 1936. By the end of these two
days of successful exhibition, officials and observers were totally astounded by the carʼs
performance and reliability, prompting motoring writersʼ praise and surprise that its
development had been such a well-kept secret! However, I believe that during the evening
of the 9th, as everyone was packing up to leave, the prototype left the road at Virage de la
Ferme and crashed. This is the first acute bend at the end of the long straight returning to
the Autodrome after passing the water tower. The wrecked prototype is illustrated on page
60 of Bernhard Simon and Julius Krutaʼs book, ʻThe Bugatti Type 57Sʼ. The damage to its
alloy bodywork was such that the car could never have been repaired in time for the ACF
three weeks later. Consequently, it is my view that four cars were originally constructed,
where only three were ever seen at the ACF, and then later the Marne GP. Affirmation that
four cars were originally constructed may be established from the observation in
photographs, that the racing pit counter at the ACF show four bays allocated, along with
four race number headings. These being, 82, 84, 86 and 88, although only numbers 82,
3
84 and 86 were actually used by the three ʻTanksʼ seen. Surely Bugatti was not so
extravagant as to finance four bays and numbers for just three cars? Undoubtedly, the
fourth Tank was written off, so that when the factory accountant, Pracht, allocated three
chassis numbers the day before the ACF, only three ʻTanksʼ really existed. The observant
will note that of the three Tanks, each is slightly different in detail, so can be individually
identified from some angles. Most obvious, is that Wimilleʼs car has high front wheel
cooling air intakes, whereas the other two have low ones. Additionally, each has a different
bonnet shape where it meets its fasteners, and the fastener position varies too between
them. It may also be noted that the two-tone blue ʻflashʼ is slightly different at the front of
the cars as well. From these little details, it is possible to establish that in the ACF, car
number 82 was race number 14 in the Marne Grand Prix. Car number 84 in the ACF was
race number 12 in the Marne, and car number 86 in the ACF was race number 44.
The three cars were weighed-in by officials at the Marne, but curiously, each had a unique
weight. It is known that the surviving car, which is the Le Mans winner of 1937, also seems
to be car number 12 in the Marne, and race number 84 at the ACF. Moreover, when
stripped for refurbishment in the early Sixties, it was seen to have a chassis lightened with
holes and utilized light-alloy cross-members. However, at the Marne GP, this car was
originally the heaviest at 1265 kg. The lightest car was Benoistʼs, being number 82 at the
ACF and number 14 at the Marne, weighing 1225 kg. Car number 86 at the ACF and
number 44 at the Marne weighed-in at 1245 kg. Each car was different by around 20 kg,
so the range between the heaviest and lightest was a considerable 40 kg. This range of
weight is conducive with the difference between a standard T57S chassis and a drilled
one, as well as conducive with the difference between one with a drilled chassis and one
with a drilled chassis and alloy cross-members. Additionally, there would be further slight
values between cars using standard T57S components, and those enjoying the benefit of
magnesium castings. Overall, given that the crashed prototype and Wimilleʼs car are
essentially similar in appearance, it seems probable to me, that these two Tanks were the
first ones constructed using early T57S components, and the other two with low tyre
ventilators, were constructed later; one with a drilled chassis, and the other with a drilled
chassis and alloy cross-members, with some magnesium castings. Inexplicably however,
Benoistʼs lightest car disappeared after the 1936 Paris Salon and was never seen again.
The two cars seen at Le Mans a year later, now greatly modified, have body details which
suggest that the winning Wimille/Benoist car number 2, was previously 84 at the ACF and
number 12 at the Marne in 1936. Likewise, the Veyron/Labric car, number 1, was
previously number 86 at the ACF and 44 at the Marne. The modifications included a new
fuel tank and spare wheel stowage, clearly necessitating that the cars be completely
dismantled and rebuilt. Given that the successful Wimille car had originally been the
heaviest in 1936, and has ended up with an extensively lightened chassis frame, alloy
cross-member and several magnesium castings, and given that we justifiably suppose it to
have originally been an early version using many standard T57S components, then it
follows that the lightest Benoist car which disappeared completely after October 1936, was
extensively utilized in the rebuild, leaving just the Wimille Tank body to be added to it. This
would make sense of the historical photographic evidence, and would be an appropriate
amalgamation for a car to satisfy both its drivers, Wimille and Benoist. This left only one
car, the Veyron/Labric one, to modify to similar standards. This scenario might also
reconcile the otherwise irrational rumour that the 1936 Wimille ʻTankʼ was rebuilt as a T57S
Works demonstrator! If the original Wimille car had an essentially unmodified standard
T57S chassis frame and running gear, then of the three cars, it was ideally placed to be
returned to the production line or rebuilt as a demonstrator.
4
External body details changed during the Tanksʼ racing period. The original streamlined
prototype seen at Montlhery on the 8th and 9th of June 1936 had a pristine body shape. It
had only one central fuel filler, smooth rear wheel covers without louvres, no rear wing
ventilators, no rear jacking beams, no driverʼs flyscreen, and no leather bonnet straps. The
engine oil filler cap was located below a small access cover. By the ACF on the 28th June,
all three cars had acquired two fuel fillers, ventilated rear wheel covers, rear wing ventilator
exits, external rear jacking beams, a driverʼs flyscreen, a small flip-up cover above the
radiator filler cap, and leather straps to additionally secure the bonnet. Also, there were
long vertical ventilators each side to help cool the cockpit footwell, as well as small
ventilators each side of the scuttle, probably ducted to cool the magneto. For the Marne
Grand Prix a week later, an additional long louvre was cut into each side of the bodywork
just ahead of the rear tyres to aid cooling of the rear wheels and brakes. For the World
Record attempts at Montlhery between September and November later in the year, the
Wimille car had the long rear louvre welded back up, was fitted with larger diameter
headlamps, a higher and narrower driverʼs cowl, a wider passengerʼs seat cover to narrow
the cockpit, as well as special lockable covers for the fuel fillers. Additionally, the car
acquired a second rev-counter and other gauges, the radiator was moved about 100mm to
the left (as you sit in the car) to allow cool air to reach the carburetor; the small radiator
cap cover being correspondingly widened. Small baffle plates were added across the
radiator and oil cooler to maintain engine temperature during prolonged runs in these
cooler months. Rear wheel covers were fitted on some runs and omitted on others, as well
as the occasional fitting of larger diameter rear wheels to increase speeds. For the Le
Mans 24 Hour Race regulations, doors were cut into each side, the rear wheel was stowed
on the tail; the fuel tank being re-designed to suit. Also, an external engine oil filler cap
made access easier, a large central headlamp was fitted into the radiator intake tunnel,
and the radiator opening was increased in height to compensate for this restriction to flow.
In addition, the rear jacking beams were modified with ʻmushroomʼ heads to prevent the
car slipping off the jack, a large lamp was fitted into the right hand side of the body, and
the Wimille car acquired yet another revised dashboard layout.
I hope this information helps to clarify things for the 1936-37 T57G Tanks; especially for
model makers. Now to work on part 2.
5

[Roger Clark]

It certainly is of interest. I hope you will post on the later cars.

[D-Type]

Fascinating!

Please, please, keep it coming.

[MTAnorak]

I'm working on it.

[MTAnorak]

Part 2 - Type 57C Tank 1939
Here again we face a lot of confusion. The 1939 Le Mans winning ʻTankʼ was a 3.3 metre
long wheelbase Type 57C chassis converted with Grand Prix wheels and brakes finished
with a beautifully streamlined body. The car was so reliable and fast that it finished 26
miles ahead of its rivals. The ʻCʼ in its designation stood for Compressuer or supercharger,
which was permitted for Le Mans race entrants that year. The long wheelbase Tank was
entered in this event because it was the only T57 model being manufactured by Bugatti;
production of the shorter T57S having ceased in May 1938. Publicly, the car was
presented as a standard 1939 T57C model taken from the production line, although many
modifications had been carried out on it. However, factory records called it a T57G, and as
previously discussed in Part 1, this is because it was a ʻGrand Prixʼ version of the long
wheelbase T57 and used a grand prix tuned, and now, supercharged ʻGʼ specification
engine. Its engine and chassis numbers are currently unknown.
Undoubtedly, the car was as extensively modified beneath it shell as its 1937 forerunner.
Factory records and memories of mechanics recall that the car was indeed selected from
the production line, but was subject to detailed preparation. Unfortunately, eight weeks
after Le Mans, the car was wrecked whilst testing it on a public road prior to the La Baule
Grand Prix, tragically killing Jean Bugatti in the process. Subsequently it was buried in the
factory grounds and has never been recovered since, so there is no opportunity to
examine it in detail. Nevertheless, some comparative conclusions may be drawn from the
way the 1937 car was put together, as well as considering the known build information.
Froude dynamometer records show that its supercharged engine produced 200bhp, so it
was clearly tuned to Type 57SC specification! For power and reliability, I believe the
engine would have been a supercharged version of the usual ʻGʼ unit, with a dry-sump
lubrication system. Photos of the wrecked car show the characteristic central magneto
drive of the T57G engines. Given the nature of this unit, the power was almost certainly
transmitted through a standard T57SC twin-plate clutch rather than the normal T57 singleplate
one. It is known that the car was fitted with the T57S45 all-synchromesh gearbox.
This gearbox was designed to accept a bell-housing for touring applications or as a
separate entity for Grand Prix use. It also had two top plate options; to suit a Grand Prix
chassis with a side gear-change lever, or a touring version with central gear lever. A photo
of the Tank cockpit during the race clearly shows a central gear lever. The T57S45
gearbox had its own pressurized lubrication system with an option for an external oil
cooler. For reliability and to counter the endurance stresses of the Le Mans event, the
obvious option would have been to use an oil cooler version, which was very simple to
install. As before, the wheels and brakes were grand prix type, this time being the latest
T50B versions with wider drums and hydraulic operation. Wheel locking nuts too, were the
latest pattern with flared ears to reduce the risk of damaging the delicate wire spokes with
the mallet. The suspension would have been made stiffer, but used standard 1939 front
and rear telescopic dampers which were very good for the period. The oil viscosity within
them could have been easily adjusted and tuned to suit the lighter weight of the car.
The 1939 Series 2 chassis frame had great potential for weight reduction, which almost
certainly could have included trepanning holes in the beams like the 1937 Tank. Here the
1939 frame was reinforced with large cross-members because the Series 2 model used a
rubber mounted engine. However, if the power unit fitted was actually a supercharged
T57G power unit, then for lightness, the engine could have been solidly mounted in the
frame just as the Series 1 type was. As a result, and because of much lighter bodywork,
many of the 1939 specification chassis bracing components could have been removed.
That is to say, the chassis frame might very well have been retro-built to the earlier T57GR
(Grand Raid) specification, especially as the carʼs much lower bonnet line and steering
wheel position suggests this modification anyway.
The Tank bodywork was completely different from the 1937 version, where lessons had
been learned. The original 1936/37 car had suffered many overheating problems due to
the enclosed bodywork which became especially apparent during the hot summer day of
the 1936 ACF. Issues included heat-seized wheel locking nuts, as well as underbonnet
overheating leading to vapour locks in fuel lines, making engine re-starts almost
impossible. Water had to be slopped around the right-hand side of the engine, carburettor
and fuel lines to cool them before re-starting after pit stops. The 1936/37 version had
several component and body modifications to counter this, such as moving the radiator to
the left to improve cool air to the problematic side of the engine, as well as added louvres,
and permanently removing the rear underbody panel behind the rear axle.
The 1939 bodywork however, utilized massive air outlets either side of the scuttle, as well
as the extensive use of the latest ram-air aeronautical technology to cool the tyres, brakes
and magneto. Streamlining was greatly improved with an absence of louvres, using
recessed fuel and oil filler caps, and by stowing the spare wheel in the tail completely
hidden under a panel. Even the headlights had their own streamlined cones. Mushroomheaded
rear jacking points ensured the car would not slip off its jack during wheel
changes. A mysterious hole in the left-hand lower bodywork, just ahead of the scuttle,
seems to be an access tube for an electric external starter mechanism. In this instance,
the side starter drive engaged with the clutch-shaft through bevel gears in the bellhousing.
For starting, the gearbox had to be in neutral and the clutch pedal free. The
Grand Prix T50B with a T57S45 gearbox used a similar application. The dashboard of the
car is still a bit of a mystery, so it is unknown whether the car used twin tachometers or
not. A photo published on page 159 of Barrie Priceʼs superb book, ʼ57 - The Last French
Bugattiʼ, was thought to be that of the 1939 Tank, but on closer examination may be seen
to be that of the October - November 1936 World Record car with its specially streamlined
cowl. Bonnet louvres may be observed which were clearly for the 1936/37 car.
A photograph exists of the 1937 car leading ahead of the 1939 version at Le Mans, where
some people have suggested the 1937 version was a stand-by car. The 1937 version
bears a race number 1 like the 1939 car, where in 1937 it was race number 2. However, it
may be noted that all cars in the 1939 event used a black number on a white roundel,
whereas in 1937, they all used a white number on a black roundel. Consequently, it is
clear that the 1937 car was only there for publicity purposes, having won in 1937 and
unbeaten since, as well as an exhibit for the pageant. Some of its lightweight and GP
components may very well have been ʻborrowedʼ for the 1939 car. Parts were noted
missing when stored in Bordeaux during WW2. It still bore the number 1 race roundel long
afterwards even when it was acquired by Uwe Hucke in the early 70ʼs.
In Part 3, I shall discuss the T57S45 Tanks.

[fivestar]

This is brilliant. many thanks.
I look forward to part 3. I believe one of these cars became the car in which Wimille won the Luxembourg Grand Prix in June 1939. the other?

Edited by fivestar, 24 April 2012 - 03:17.

[Tim Murray]

Thank you - this is fascinating stuff.

[MTAnorak]

Type 57S45 Tank 1937
This is one of the least known about racing cars from the period. In late 1936, Bugatti
announced that six of these 4.7 litre streamliners were planned for the following year
although only one car was ever exhibited in public as such. Displaying race number 16
and registered 1202-W5 (W5 numbers were temporary, like trade pates), it stunned
everyone who saw it briefly perform in practice prior to the 1937 ACF at Montlhery. It
appears that Benoist took the car around the Autodrome without the formal permission of
track officials, so was promptly disqualified from the race. Effectively a world champion of
his day, Benoist was an exceptionally experienced and successful racing driver who
should never have made such a serious error of judgement; assuming of course that it was
his own decision. A second T57S45 noted at the Works with race number 14 and
registered 2103-W5, was reportedly damaged during transit from Molsheim to Montlhery
that same day and so returned to the factory. Until relatively recently, the history of the two
ultra-streamlined T57S45 Tank bodied racing cars was minimal, frustrating and confused.
The very limited information available about these cars was primarily sourced from news
reports and just a few photographs. The cars were reportedly destined for Benoist and
Wimille to campaign in the 1937 Le Mans 24 Hour Race, and two weeks after that, in the
ACF at Montlhery. However, they never materialised except for the brief appearance of
one car in the latter event for practice before seeming to vanish forever! Accordingly, the
details and history of these cars have been vague ever since.
Several Bugatti books suggest that the T57S45 appeared in practice at Le Mans, although
this was a mistake which seems to have originated from a contemporary news report
which led, in 1954, to it being innocently published in “The Bugatti Bookʼ by Barry
Eaglesfield. Unfortunately, this error was thereafter often repeated as ʻhistory”, even
though doubt continued amongst enthusiasts. Modern research has confirmed that the
official Le Mans organizers, Automobile Club de lʼOest, have no records whatsoever
regarding the T57S45 Tanks. Studies suggest that some of the perplexing specification
and history of the T57S45 originates from formal publicity material issued during the
October 1936 Paris Salon, along with a revised leaflet shortly afterwards in December.
These illustrated the proposed T57S40 (3987cc) and the T57S45 (4745cc) road-going
sports cars. From the outline diagrams published, these two were quite clearly arranged
around a T57S ʻsurbaisseʼ chassis frame on a 2.98 metre wheelbase. As a result, it might
be reasonably taken for granted by historians that the racing Tank version would not only
have comparable chassis dimensions, but logically, was the next maturation phase of the
earlier, T57S surbaisse-framed T57G Tanks. For those outside the ʻWorksʼ inner circle,
such a conclusion was encouraged by the fact that one of the earlier streamliners had
apparently vanished from active competition not long after the Paris Salon, as too the
persistent rumour of the existence of a ʻfourthʼ T57G Tank. As a result, it was entirely
logical for reporters to assume that the two missing T57G cars had been stripped and rebodied,
to be powered by the much-vaunted 4.7 litre T50B engines.
Additionally, bolstered by the erroneous 1937 Le Mans practice reports, a number of
Bugatti historians generally upheld this view until the mid-Nineteen Eighties. There was
however, another train of thought. A number of enthusiasts had difficulty reconciling the
ʻofficialʼ T57S45 dimensions with the well-known photos of car number 16 at Montlhery.
The car just did not seem to have the right proportions for a T57S chassis. There was
opinion that their dimensions were closer to those of a T59, where this contention was
hampered by the fact that Factory records were not then readily accessible, and very few
photos of the cars had ever been published. Nevertheless, with some encouragement for
new research and the committed studies of Hugh Conway (Senior), by 1987 and the 4th
Edition of his book, ʻle pur sangʼ, Hugh formally confirmed that the build sheets specified
Type 59 chassis side frames as long suspected by some. With the establishment of ʻThe
Bugatti Trustʼ at Prescott, factory records are now more accessible than they were,
although unfortunately, the Trust archive still does not have either a complete set of
engineering drawings of the T57S45, or adequate photographic records which show the
precise layout and composition of components. However, they have successfully amassed
a good collection of external photos of these cars from most angles. Moreover, with the
research of Bugatti Trust photographer David Morys, along with period material compiled
by the late Anthony Blight, the well known American Bugatti specialist Ray Jones has
obtained enough information to construct a replica of the super-streamlined T57S45 Tank.
The construction and sudden disappearance of the two Tanks has long been a mystery; for
what happened to them? My personal opinion is that these cars were a genuine Bugatti
venture, but was one plagued with problems and a lack of progress with the T50B engine
project. As a result, there may very well have been elements of ʻspin doctoringʼ or
propaganda to present a more positive picture of development than there actually existed.
It is known that Bugatti received 100,000 francs (£11,600) of funding from the ʻFonds des
Courseʼ in late 1935 to help develop a racing engine which could compete with German
racing machinery. Later in 1937, Bugatti was offered funding of 6.9 million francs
(£800,000), in staged payments, towards the development of engines and contra-rotating
blade gearboxes for Louis de Mongeʼs advanced wood composite aircraft designs, the
P100 and P110; a portion being received in August 1938. Additionally, the French
Government offered further funds towards the development of engines for a high-speed
torpedo boat. Records suggest that all of these engines were essentially light alloy T50B
types. We may see therefore, that the various bodies who donated funds would, from time
to time, have expected to see some sort of progress for their investment. Additionally,
there was almost certainly an international political aspect, in an attempt to demonstrate
Franceʼs technological ʻcatch-upʼ with Germany. Modern studies indicate that Bugattiʼs
T50B engine development went through many stages of success and failure, although a
reliable and really powerful T50B power unit did not appear until 1939. As a result, in the
summer of 1937 it is clear the T50B engine was still not really race-reliable, and that
perhaps the disqualification of Benoistʼs T57S45 from the ACF was extremely convenient.
Given that T59 chassis frames formed the basis of the T57S45 Tanks, is it possible to
identify them? There are several possible scenarios, although mine is as follows. As
discussed in Part 1, four of the nine T59ʼs built were sold off in 1935, leaving Bugatti with
five redundant cars. One of them (later two) was being developed as a T50B single-seater.
One T59 seems to have remained essentially a standard version and one crashed badly in
July 1936. However, two T59-like cars ran as T59/57G in the 1936 Comminges race for
unblown cars the following month. To the consternation and outrage of rival competitors,
these two racers were presented as Type 57. Nevertheless, as previously discussed, this
is effectively what they had become, and so the track officials permitted them to race
despite protest. A photo on page 145 of Robert Jarraudʼs book, ʻBugatti doubles arbresʼ
shows that car, race number 16, has a central ʻGʼ type magneto and T57 fuel pump
system, which clearly suggests a T57G engine in a T59 crankcase. It follows that its ʻsisterʼ
car must have been essentially identical in order to pass scrutineering. Interestingly, other
photos of this event show that of the two cars, one of them had a riveted spine ʻAtlanticʼ tail
and the other a smooth fabricated version. Race number 14, probably 57248, had a
riveted spine tail and was driven by Wimille. Car, race number 16, was driven by Benoist.
During the winter of 1936/37, both cars acquired all-synchromesh T57S45 gear-boxes,
where this modification is characterized by the addition of long rear suspension radius
arms. This major winter overhaul taking several months would have been an ideal
opportunity to prepare bucks for the new T57S45 streamlined bodies. I believe these two
cars were completely stripped and modified in preparation to become T57S45 Tank racing
sports cars for the 1937 season. The right-hand chassis frames were specially modified to
clear the long-awaited T50B engineʼs supercharger, and rebuilt with large dural crossmembers.
The T57S45 gearbox was mounted further back in the frame to allow for the
longer engine. However, as discussed, the T50B engines remained problematic, so as a
consequence, the two cars were then refitted with T59/57G engines; numbers 4 and 5.
Chassis 57248 was re-bodied with skimpy bodywork and cycle wings, to become the ʻT57
Sportʼ, in time for Wimille to race and win at Pau in February 1937. I believe the other (ex-
Benoist) chassis was rebuilt with its original T59 bodywork with fabricated tail. Now with a
supercharged version of the T59/57G engine, this car was driven by Wimille to win the
400,000f prize at Montlhery in April 1937. Following this event, I believe the car was rebuilt
as the T57S45 Tank, race number 16, which was seen at Montlhery for practice. I
think it is likely that it was the only one of the two to be fitted with a running T50B engine.
After the winning the Pau GP, the T57 Sport (57248), ran in Tunis, and won at Bone. The
following month, the car won the first heat at Miramas, but ran a big-end in the third. Since
it had already been prepared for its T57S45 body during the winter, the car was quickly
rebuilt with its super-streamlined coachwork and the supercharged T59/57G engine from
Benoistʼs car. Under-bonnet photos of the T57S45 Tanks have not yet been found, and
curiously, those taken of car number 14 with its bonnet up, have been taken at such a low
angle it is not possible to see the engine at all. I suggest this because although 57248 has
many of the essential modifications to become the basis of a T57S45 Tank, and whilst its
right-hand chassis frame has been modified to clear a T50B supercharger housing, the
frame itself has never been drilled to accept this power unit. Fortunately, the T57 Sport,
chassis 57248, has never been significantly modified and is conserved largely as it was in
1938. Modern scrutiny of the car confirms the chassis details mentioned, including its
shorter propellor shaft. If it was indeed built as a T57S45 Tank with a supercharged 3.3
litre T59/57G engine, then taken as a whole, the episode at the 1937 ACF has something
of the appearance of being a public relations exercise to bolster Bugattiʼs promotion of the
T50B whilst it remained unreliable. The disappearance of the T57S45 Tanks may have
been due simply to their relative weight in comparison to overall performance, tempered
with a change of strategy for the 1938/39 season away from sports car grand prix.
The T57 Sport, 57248, back with its skimpy bodywork and unsupercharged engine, raced
and won the Marne Grand Prix on 18 July, just two weeks after the ACF. Whilst this does
not seem to provide much time, its original engine had been completely overhauled since
retiring at Miramas in June, with plenty of opportunity to fully test and tune it on the Froude
dynamometer. Other significant components in the car remained unused, and the refitting
of its original bodywork would have been as straight-forward, as it would be during any
major service.
The T57S45 Tank which appeared at Montlhery in 1937 had a very distinctive fuel filler
position and cap, where many had noticed the similarity with that of the skimpy bodied,
bulbous-winged T59/50B which won the Luxembourg Grand Prix in June 1939. It seems
very likely that this was essentially the same chassis, having earlier been the other
T59/57G in 1936. The T59/50B ran again at St Gaudens for the Comminges Grand Prix in
August 1939, and now resides at the National Museum in Mulhouse. The car is currently
fitted with the special twin-wheeled axle originally seen on the T59/50B single-seater
(chassis 50180) at the Prescott Hill Climb in May 1939. The axles were probably swapped
over post-war when this car was entered for the Coupe de Prisonniers in September 1945.
The bulbous winged T59/50B currently has a new, undrilled, chassis frame which does not
appear to bear a chassis number. Several people seem to think the frame was replaced
during its restoration by the Schlumpf brothers. Further research is necessary.
In Part 4 I can express some opinions about the development of T50B engines and
perhaps reconcile some of the anomalies about T59 chassis numbering.

[Cris]

Great reading; thanks, MTA.

Cris

[figoni]

Yep absolutely fascinating stuff, keep it coming, interesting that after all of this time nobody has been able to pin down the records for some of these cars, chassis numbers et al. I guess we may never know.

[fivestar]

A couple of photos showing the fuel filler position of the T57/45s Tank and the T59/50B.

[MTAnorak]

Part 4
Type 59 chassis numbering
Anomalies in T59 chassis numbering have long been a mystery as well as a source of
confusion where they range 441352, 50180, 54208, 54213, 57248 including the expected
59 variants. Here I proffer a possible explanation after years of consideration.
Firstly, we have to recall the history of the development of the Type 59 racing car. While
Ettore Bugatti was immersed in the Railcar project at his Paris office in 1932, his son Jean
had effectively taken over the responsibility for the factory at Molsheim, including the
development of the Type 57 touring car. It is apparent that for almost a year, Jean Bugatti
designed and constructed the T59 racing car without the consent of his father, or for that
matter, having ever informed his accountant! This is known about from a well-known story
from the period, when Ettore Bugatti discovered cars and components during an
unannounced visit to the Molsheim factory; possibly around October 1932. Absolutely
incandescent with rage, he immediately ordered the Type 59sʼ destruction. Yet shortly
afterwards he was charmed around by Jean and famously contributed to the project by
designing its beautiful and unique radial spoke wire wheels; the earliest sketch being
dated November 1932! Since Jean had neither the mandate nor budget to create these
racing cars, especially during the 1930ʼs Depression when all costs had to be meticulously
considered, his fatherʼs fury is quite understandable. It also provides a cogent explanation
as to why three of the T59ʼs mysteriously have inappropriate Type chassis numbers.
These being for a Type 44 and two Type 54s!
Frame numbers are stamped on rear spring hanger cross-beams, ranging 1 to 8, and
engine numbers range similarly, but do not appear to relate directly to chassis numbers. It
is likely that in practice, engines may have been swapped around from time to time.
However, having explored several possible grounds for the atypical chassis numbering
over a number of years, I have had to conclude that Ettoreʼs exuberant and headstrong
son Jean, deliberately camouflaged the construction of the first three Type 59 chassis
assemblies by billing costs to existing and otherwise authorised 1932 projects! Hence the
chassis numbers: 441352; 54208 (later numbered 59123); and 54213 (later numbered
59124), none of which for 1932 are otherwise accounted for in existence. Some pretty
imaginative and audacious book-keeping perhaps? If this deduction is correct, it would
suggest that these three cars were the very first Type 59s to have been constructed, and
that 441352 might very well be the first! Jean may have been able to hoodwink Henri
Pracht, works secretary cum accountant, deeply preoccupied with office work, but he could
never dupe Le Patron!
Chassis numbers 57248 and 50180 are both assigned to T59 chassis frames although in
this case they truly relate to their respective engines. Both of these numbers appear to
have been allocated between the end of 1934 and early 1935 at the time Bugatti
disbanded the racing department and formally proposed the development of a super
sports car for 1936. At the beginning of 1935 Jean Bugatti announced that the company
would be developing the ultimate super sports car which would replace the outmoded Type
55, and be of similar dimensions. Most historians agree this project culminated in the T57S
on a 2.98m wheelbase. However, consider that the Type 55 had been based upon the
same redundant T47 ʻCompetitionʼ chassis frame as the T54 Grand Prix racing car. Given
this mindset, the initial concept may easily have been to dispose of redundant T59 frames
and components through their evolution into sports cars. The entry date of the chassis
number is not recorded for 57248, but it falls sequentially in the Factory Register between
22 December 1934 and 18 May 1935. This is precisely the period between the
disbandment of the racing department and the decision to develop a super sports car.
Another Type 59 chassis frame was assigned the serial number 50180, where as before,
the date is not recorded, but it follows in sequence after an entry on 5 December 1934. I
believe this notable chassis numbering expressed Jean Bugattiʼs commitment to two
sports car prototypes, where according to usual practice, suggests cars with engine types
57 and 50 respectively; the embryonic T57S and T57S45. Whilst sports car projects seem
to have been the intention in 1935, following ʻFonds des Courseʼ funding, by 1939, 50180
had evolved into the supercharged T59/50B, single-seater Grand Prix car seen today in
the National Museum at Mulhouse.
Some comments about the T50B Project
The evolution of the T50B engine is still subject to a lot of study and differing opinions.
Referring back to Part 1 in which I suggested that Type numbers should be viewed as
project numbers, this approach might explain several things. The first record of the T50B is
as early as 1934, which is the period during which Bugatti was still developing his Type
50S super sports car project. However, as a result of the burgeoning world financial slump
and the need to rationalize spending at the works, by the beginning of 1935 the extremely
expensive T50S super sports car project had been dropped. Interestingly, the T50S
appears to have had many of the attributes later seen in the T57S including a ʻsurbaisseʼ
chassis and a ʻveeʼ radiator. Naturally in 1934, the T50 iron engine would have been
subject to development as a suitable power unit for the T50S; hence the embryonic T50B
project. This engine had long-utilized a 2-4-2 crank array, providing it with notable
smoothness as well as a potential for higher rotational speeds.
In June 1935, Benoist tested a T59 in the ACF (thought to be chassis 441352) fitted with a
special engine. The rather suspicious circumstances of the carʼs night-time official
examination has been previously discussed and joked about. Additionally, it has been
memorably noted that during the race itʼs ill-fitting bonnet flew off, where Benoist managed
to catch it without injury! Some historians suggest the power unit was the highly-tuned 4.9-
litre T50 iron engine from Count Czaykowskiʼs fatally wrecked T54. However, I believe it is
likely that it was an early highly-tuned T50B ʻironʼ engine with a dry-sump lubrication
system, although perhaps in reality both were essentially one and the same in
specification. Bear in mind too, this is the period when Bugatti had stridently abandoned
Grand Prix racing because of state sponsored German and Italian competition.
Consequently, in collusion with like-minded ACF officials, there seems to be an element of
subtle tactics going on here, perhaps insinuating to the newly-formed ʻFonds des Courseʼ,
the need to provide Bugatti with financial assistance to further develop this ʻnewʼ engine
and running gear. It is obvious that the car was not compatible with the T50B iron unit,
where in pre-T57S45 all-synchromesh gearbox reality, the standard T59 version was
completely incompetent to transmit all the torque that this engine could put out.
Nevertheless, as mentioned in Part 3, some months later, the ʻFonds des Courseʼ did in
fact award Bugatti 100,000f of development finance for racing.
Complex aluminium and magnesium lightweight alloys were relatively new materials to
Bugatti in 1935, in which their practical design parameters and limitations had not really
been established within the drawing office. As a result, Bugatti appears to have arrived at
some rather over ambitious notions for the proposed power to weight aspects of his yet to
be developed lightweight version of his T50B engine. As a result, this project was hindered
for years by many problems and delays.
For many years I have tried to unravel the sometimes undocumented and certainly
complex history of Bugattiʼs cars from this period, in an attempt to make sense of the
known facts and rumors which surround them. I trust these essays will prompt enthusiasts
to continue studying these fascinating cars, and I look forward to any counter-arguments to
those proposed, or anything which might bolster them. During the next couple of months I
am expecting to be moving home, so please bear with me over any delays in responding.
Vive la Marque!

[MTAnorak]

Part 4
Type 59 chassis numbering
Anomalies in T59 chassis numbering have long been a mystery as well as a source of
confusion where they range 441352, 50180, 54208, 54213, 57248 including the expected
59 variants. Here I proffer a possible explanation after years of consideration.
Firstly, we have to recall the history of the development of the Type 59 racing car. While
Ettore Bugatti was immersed in the Railcar project at his Paris office in 1932, his son Jean
had effectively taken over the responsibility for the factory at Molsheim, including the
development of the Type 57 touring car. It is apparent that for almost a year, Jean Bugatti
designed and constructed the T59 racing car without the consent of his father, or for that
matter, having ever informed his accountant! This is known about from a well-known story
from the period, when Ettore Bugatti discovered cars and components during an
unannounced visit to the Molsheim factory; possibly around October 1932. Absolutely
incandescent with rage, he immediately ordered the Type 59sʼ destruction. Yet shortly
afterwards he was charmed around by Jean and famously contributed to the project by
designing its beautiful and unique radial spoke wire wheels; the earliest sketch being
dated November 1932! Since Jean had neither the mandate nor budget to create these
racing cars, especially during the 1930ʼs Depression when all costs had to be meticulously
considered, his fatherʼs fury is quite understandable. It also provides a cogent explanation
as to why three of the T59ʼs mysteriously have inappropriate Type chassis numbers.
These being for a Type 44 and two Type 54s!
Frame numbers are stamped on rear spring hanger cross-beams, ranging 1 to 8, and
engine numbers range similarly, but do not appear to relate directly to chassis numbers. It
is likely that in practice, engines may have been swapped around from time to time.
However, having explored several possible grounds for the atypical chassis numbering
over a number of years, I have had to conclude that Ettoreʼs exuberant and headstrong
son Jean, deliberately camouflaged the construction of the first three Type 59 chassis
assemblies by billing costs to existing and otherwise authorised 1932 projects! Hence the
chassis numbers: 441352; 54208 (later numbered 59123); and 54213 (later numbered
59124), none of which for 1932 are otherwise accounted for in existence. Some pretty
imaginative and audacious book-keeping perhaps? If this deduction is correct, it would
suggest that these three cars were the very first Type 59s to have been constructed, and
that 441352 might very well be the first! Jean may have been able to hoodwink Henri
Pracht, works secretary cum accountant, deeply preoccupied with office work, but he could
never dupe Le Patron!
Chassis numbers 57248 and 50180 are both assigned to T59 chassis frames although in
this case they truly relate to their respective engines. Both of these numbers appear to
have been allocated between the end of 1934 and early 1935 at the time Bugatti
disbanded the racing department and formally proposed the development of a super
sports car for 1936. At the beginning of 1935 Jean Bugatti announced that the company
would be developing the ultimate super sports car which would replace the outmoded Type
55, and be of similar dimensions. Most historians agree this project culminated in the T57S
on a 2.98m wheelbase. However, consider that the Type 55 had been based upon the
same redundant T47 ʻCompetitionʼ chassis frame as the T54 Grand Prix racing car. Given
this mindset, the initial concept may easily have been to dispose of redundant T59 frames
and components through their evolution into sports cars. The entry date of the chassis
number is not recorded for 57248, but it falls sequentially in the Factory Register between
22 December 1934 and 18 May 1935. This is precisely the period between the
disbandment of the racing department and the decision to develop a super sports car.
Another Type 59 chassis frame was assigned the serial number 50180, where as before,
the date is not recorded, but it follows in sequence after an entry on 5 December 1934. I
believe this notable chassis numbering expressed Jean Bugattiʼs commitment to two
sports car prototypes, where according to usual practice, suggests cars with engine types
57 and 50 respectively; the embryonic T57S and T57S45. Whilst sports car projects seem
to have been the intention in 1935, following ʻFonds des Courseʼ funding, by 1939, 50180
had evolved into the supercharged T59/50B, single-seater Grand Prix car seen today in
the National Museum at Mulhouse.
Some comments about the T50B Project
The evolution of the T50B engine is still subject to a lot of study and differing opinions.
Referring back to Part 1 in which I suggested that Type numbers should be viewed as
project numbers, this approach might explain several things. The first record of the T50B is
as early as 1934, which is the period during which Bugatti was still developing his Type
50S super sports car project. However, as a result of the burgeoning world financial slump
and the need to rationalize spending at the works, by the beginning of 1935 the extremely
expensive T50S super sports car project had been dropped. Interestingly, the T50S
appears to have had many of the attributes later seen in the T57S including a ʻsurbaisseʼ
chassis and a ʻveeʼ radiator. Naturally in 1934, the T50 iron engine would have been
subject to development as a suitable power unit for the T50S; hence the embryonic T50B
project. This engine had long-utilized a 2-4-2 crank array, providing it with notable
smoothness as well as a potential for higher rotational speeds.
In June 1935, Benoist tested a T59 in the ACF (thought to be chassis 441352) fitted with a
special engine. The rather suspicious circumstances of the carʼs night-time official
examination has been previously discussed and joked about. Additionally, it has been
memorably noted that during the race itʼs ill-fitting bonnet flew off, where Benoist managed
to catch it without injury! Some historians suggest the power unit was the highly-tuned 4.9-
litre T50 iron engine from Count Czaykowskiʼs fatally wrecked T54. However, I believe it is
likely that it was an early highly-tuned T50B ʻironʼ engine with a dry-sump lubrication
system, although perhaps in reality both were essentially one and the same in
specification. Bear in mind too, this is the period when Bugatti had stridently abandoned
Grand Prix racing because of state sponsored German and Italian competition.
Consequently, in collusion with like-minded ACF officials, there seems to be an element of
subtle tactics going on here, perhaps insinuating to the newly-formed ʻFonds des Courseʼ,
the need to provide Bugatti with financial assistance to further develop this ʻnewʼ engine
and running gear. It is obvious that the car was not compatible with the T50B iron unit,
where in pre-T57S45 all-synchromesh gearbox reality, the standard T59 version was
completely incompetent to transmit all the torque that this engine could put out.
Nevertheless, as mentioned in Part 3, some months later, the ʻFonds des Courseʼ did in
fact award Bugatti 100,000f of development finance for racing.
Complex aluminium and magnesium lightweight alloys were relatively new materials to
Bugatti in 1935, in which their practical design parameters and limitations had not really
been established within the drawing office. As a result, Bugatti appears to have arrived at
some rather over ambitious notions for the proposed power to weight aspects of his yet to
be developed lightweight version of his T50B engine. As a result, this project was hindered
for years by many problems and delays.
For many years I have tried to unravel the sometimes undocumented and certainly
complex history of Bugattiʼs cars from this period, in an attempt to make sense of the
known facts and rumors which surround them. I trust these essays will prompt enthusiasts
to continue studying these fascinating cars, and I look forward to any counter-arguments to
those proposed, or anything which might bolster them. During the next couple of months I
am expecting to be moving home, so please bear with me over any delays in responding.
Vive la Marque!