The story behind the one-off project that comes close to being an ideal GT car

[From Automobile Connoisseur, volume 1]

THE COMPANIES within the GKN group, scattered through out the world, produce all manner of things from nuts and bolts through to pig troughs, in places as far apart as Birmingham and Australia. In the United Kingdom, however, they are principally known as manufacturers of automotive equipment, and there are very few British cars supplied which do not contain a very large proportion of component parts which have been made for the industry by one or another of the companies under the GKN umbrella. They make the nuts and bolts already referred to, fasteners, hinges, camshafts, crankshafts, con-rods, driveshafts, forged gears, stub axles, cylinder blocks, clutch and gearbox casings, differential housings, and as well as these, some of the parts which you can actually see on the complete car; wheels, bumper bars, radiator grilles, and so on. One of the companies within the group, Vandervell Products Ltd, built and raced the famous Vanwall Formula 1 Grand Prix cars with which Britain dominated the Grand Prix scene some ten years or so ago.

It may therefore come as something of a surprise to learn that this is the only complete motor car with which the GKN group have ever been associated in terms of complete manufacture. Few industrial concerns can contribute quite so many components to an industry in which they do not build a complete product, and it was no doubt with this in mind to some extent that the original project of the GKN 47D was conceived.

The idea was born at the London Motor Show of 1967, when Mr Claude Birch, a main board Director of GKN Ltd and Chairman of Vandervell Products, as well as other companies in the group, was attracted by the Lotus Elan while looking for a new personal car. The only disadvantage to the Elan, from his point of view, was in its engine capacity; he gave this a certain amount of thought and soon began to consider the possibility of installing a lightweight V8 engine.

And so, in due course, Lotus were approached with the idea. Was it possible to fit a V8 engine into the Elan? The answer was no. But the more Claude Birch thought about the idea the better he liked it and the project, as a first-class design exercise in which the maximum possible number of GKN components could be incorporated. Since it had proved to be impossible to build a suitable Elan, it then occured to him that it might be possible to carry out a similar project using as a basis the mid-engined Ford twin-cam powered Lotus 47, the competition version of the Renault-powered Lotus Europa, which was at that time available only on overseas markets. David Sankey, of GKN, and Fred Fox of Vandervell Products (the man who was in charge of the Vanwall racing car project) then went to Lotus with a suggestion. Lotus, in turn, agreed to consider such an idea and a meeting with Colin Chapman, the head man of Lotus, followed.

The Lotus 47 chassis could be used, but it was obvious at an early stage that the proposed installation of the V8 engine was going to be difficult. A variety of engines was considered. The first consideration was obviously the question of their availability as separate items. From those available in this way, a short list was eventually worked out, from which comparisons were made of weight and power. The 2 1/2-litre Daimler V8 produced 140 b.h.p. and weighed 149 lb.; the 4 1/2-litre Daimler produced 220 b.h.p., for 498 lb., while the Rover 3.5-litre light alloy V8 developed a net output of 165 b.h.p. and was surprisingly light at 321 lb. The Ford twin-cam 1 1/2-litre engine with which the Lotus 47 is normally equipped produces 105 b.h.p. in standard untuned form and weighs 260 lb. In fact, more power would be obtained for a similar weight from the Buick version of the Rover engine but reasons of national pride took a hand in the final selection of the Rover unit.

It was decided to use a German ZF five-speed gearbox, a type 5DS12-NR2, which they obtained from Lotus Who had used it in a Formula 1 Lotus-Ford car. Clearly the gearbox ratios were unsuitable, but initially measurements were taken of the overall dimensions of the engine and gearbox, with the clutch, and it was immediately obvious that the total length of the assembled unit would be too great to fit into the "Y"-sectioned Lotus chassis. The chassis would have to be lengthened, and the "Y"-section which would normally accommodate the power unit would have to be substantially modified.

A one-off chassis was built by Lotus experimentally, using a widened "Y"; this proved unsuccessful and was scrapped. A second experimental chassis was more satisfactory and this incorporated a "T"-piece at the rear, with a space frame attached to it which would provide the additional chassis length as well as widening the car and providing room for the engine and the rear suspension. Into this chassis the engine and gearbox were assembled; the flywheel hit the ground. This was merely the first of an enormous number of problems which make this car not only one of the most unusual, but possibly one of the most expensive ever built! A smaller flywheel was made by one of the companies under the GKN umbrella; obviously, it the needed a smaller clutch. A smaller clutch, naturally, would not easily cope with the power. Borg and Beck were consulted and co-operated to produce a special one-off clutch for GKN: a special 7 in. twin-plate job which, naturally, then needed a different bellhousing. This was fabricated from magnesium by Kent Alloys, another GKN subsidiary company.

With the transmission now safely accommodated, the question of gear ratios presented a further problem. As installed by Rovers in their production cars, the 3 1/2-litre and the 3005, the engine develops maximum power at 5,200 r.p.m., and is not expected to exceed this crankshaft speed by any appreciable margin. In the GKN car, however, it was expected to run to 7,000 r.p.m. and this argued not only a new crown wheel and pinion, as well as alternative ratios in the ZF gearbox, but also, subsequently, different road wheels. Engine lubrication also needed considerable thought; a new sump proved to be necessary. At one stage, the use of dry sump lubrication was considered, but this was eventually rejected on the grounds of complication. A new sump was therefore duly fabricated, again by Vandervell Products Ltd: and with the engine fully assembled, the whole thing was sent to Lotus to be installed in the chassis. The process of full assembly, so far, as the engine was concerned, involved a very complete re-build. The car was to be used as a test bed for GKN components, as well as being a prestige vehicle for the personal use of Claude Birch, and therefore the power unit was completely re-built to make absolutely certain of the use of GKN products in its assembly. Vandervell Products Ltd, in addition to the sump, provided lead bronze big-end bearing shells, main bearing shells, camshaft bearings, and the flywheel. GKN Forgings Ltd provided the connecting-rods while the Heath Street Division of GKN Screws and Fasteners Ltd provided the push-rods, and the rocker pad inserts. Con-rod bolts and nuts, cylinder head bolts and so on, were provided by GKN Bolts and Nuts Ltd, all the special titanium and very high duty bolts, screws, nuts and prevailing torque nuts were provided by T. J. Brookes (Leicester) Ltd of GKN Screws and Fasteners, and BKL Alloys Ltd provided a special alloy for the cylinder heads and block. Even the pressure relief valve filter in the oil pump was supplied by Birfield Filtration Ltd of the GKN Birfield Industries Ltd sub-group. GKN Birfield Transmissions provided a number of small alloy castings, including the spacer between the engine and the gearbox, and the clutch housing itself. GKN Forgings Ltd. made the clutch pressure plates and the engine was fitted with S.U. carburettors which, as a matter of detail, use damper pins made by the Screws and Fasteners Ltd sub-group!

Fitting the engine and transmission installation to the chassis meant that the rear suspension had to be revised; both the length and the position of the radius rods had to be altered, and with the engine around half a hundredweight heavier than the Ford twin-cam unit which is standard on the Lotus 47, quite a lot of experiment had to be carried out to achieve a satisfactory spring rate. It was eventually decided to use a combination of soft springs and stiff, though adjustable, Armstrong dampers.

The next problem was the fitting of the body shell. At that stage in the car's development, the racing Lotus 47 shell was bonded to the chassis, with non-adjustable seats. The GKN car was already intended to have a high quality of interior trim and in any case, while Lotus had already decided to fit a detachable body shell to the 47 for easier repair and maintenance, it was nevertheless obvious that the standard shell was now neither long enough nor wide enough to be fitted to the enlarged chassis. A further consideration was that the standard shell, since the 47 was intended as a competition vehicle, was in a very lightweight laminate, and it was decided to make the shell, which would have to be a special one-off in any case, of the heavier glassfibre laminate as used on the roadgoing Elan. Since the GKN project was intended to be a roadgoing machine, as well as a development vehicle, it was obviously desirable, in the interests of spare wheel accommodation, to fit the same sized wheels front and rear and it was eventually decided to fit 7 1/2 in. x 13 in. wheels, the same as the normal 47 front racing wheels. Some wheels were made up by Kent Alloys and tested on Sankey rigs, to pass the fairly rigorous tests with full marks. These wheels were specially made in the very ductile LM6 alloy, which has extremely good resistance to impact. On the disadvantage side, the material is less satisfactory when it comes to a highly polished finish, which does rather tend to offset the overall appearance of the finished product. It had originally been intended to use Dunlop SP68 tyres, but it was found at this stage that these would not fit within the wheel arch clearance, and to save changing the body shell once again, it was decided to fit the Goodyear 185-70HR x 13 low profile tyres all round. With the body shell in place, engine cooling proved to be a further difficulty. The body and chassis provides for no direct air cooling for the power unit, and the plumbing of the engine cooling system from a front-mounted radiator--in the final event, two front-mounted radiators, which provided a further snag by completely filling the boot and thus removing much of the luggage and spare wheel space--was further complicated because all piping and tubing had to pass through the centre backbone of the chassis, which was now of 1 2 gauge square section material. Eventually, the problems were ironed out but there still remains the problem of luggage space. Two boot spaces are in effect provided on the Lotus 47, or its road-going companion, the Europa; one is at the front, under the "bonnet", the other is in a special box at the rear. On the GKN car the fitting of the two front radiators and the spare wheels has in fact left no boot space available whatsoever.

The exact process of cooling the engine is quite intricate; the two radiators, of cross flow type, are connected in parallel and water is pumped through a "swirl pot" to separate the air from the water, which then passes along a tube under the chassis to the radiators. After cooling, the water returns along another tube under the chassis to the engine water pump and idling water temperature is controlled by electric cooling fans. These are operated by a thermostat and there is an overriding switch on the facia for driver control. The two 9-gallon light alloy fuel tanks, one of which incorporates a chamber below the cross pipe for a reserve fuel supply which is provided by a separate electric pump operated by a switch on the dashboard, are mounted behind firewall and connected by a cross pipe. Fuel is drawn from the left-hand tank by an electric pump feeding the two S.U. HS6 carburettors, and to prevent fuel vapourisation, the spill back is provided from the carburettors to the right-hand tank.

The main body shell unit is bolted to the chassis in six places, and the engine cover gives access to the spare wheel, the tool kit and the jack in a carpeted area; obviously, it also enables the engines service points to be reached. The front boot lid covers a carpetted area which also includes the battery box and servicing points for the brake and clutch fluid bottles and the windscreen washer reservoir, the amount luggage space, as we have said before, is minimal. The doors, also in glassfibre, are fitted with electric window lifts and all windows are made of tinted glass. Special Marchal headlights are fitted with quartz iodine bulbs. The interior is trimmed throughout in leather by Connolly, and the seats were specially made by a small coachtrimmer in Norwich. Insulation from engine heat and noise is provided by the firewall, which is constructed of two aluminium sheets enclosing a 1 in. asbestos filling, with the addition of a great deal of felt insulation material. Smiths instruments are fitted, including a 180 m.p.h. speedometer, which in itself posed a particular problem. Since the ZF gearbox used was originally designed for racing cars, there is no speedometer take-off point and so a mechanical drive had to be arranged from the front hub. A production drive unit was modified to fit the Lotus hubs, and Smiths themselves made up a special speedometer instrument. The other "clocks" are an impulse tachometer reading to 7,000 r.p.m., electrical oil pressure gauge and warning light, an ammeter, a water thermometer and fuel contents gauge. These are mounted on the facia and are supplemented by an electric clock which is fitted to the tunnel console.

Apart from a smaller flywheel and the different sump, which is of the same capacity as the original, the engine is tuned to produce 185 b.h.p., and to provide the maximum usable r.p.m. of 7,000. The exhaust system involved a great deal of thought and a special duplex system was devised by Servais, which went through the suspension. It employs a manifold to each bank of cylinders, feeding through special large capacity absorption type silencers and finally discharging its "noxious" gases through heavily chromed tailpipes. Iskenderian hydraulic tappets are fined to enable the safe use of the high engine speed, and the camshaft is the hottest road profile made by Iskenderian. In search for, still higher performance, carburation is still the object of experiment and it is proposed to try Webers, a Rochester carburettor, fuel injection and twin turbochargers as well as the existing S.U. instruments.

The performance, as we have said, is remarkable. The combination of gear ratios and wheel and tyre size with the final drive of 14:39 gives road speeds of 10, 14,18, 22 and 25 m.p.h. at 1,000 r.p.m. in each gear. Assuming that the engine is run to its maximum speed of 7,000 r.p.m., this can be translated into speeds of 70, 98, 126, and 154 m.p.h. in first, second, third and fourth gears, with, evidently, substantially more available in fifth! This sort of performance argues the necessity for outstandingly good brakes. Light alloy calipers are fitted as standard on the 47, which is, after all, the racing machine, but these were nevertheless replaced by Aston Martin iron calipers, used in conjunction with standard Lotus 47 discs.

The total weight of the car is approximately 2,000 lb., and the distribution is 45% to the front and 55% to the rear. It has a wheelbase of 94 in., a front track of 53 in,, and a rear track of 51 1/4 in. The overall length of the car is 13 ft. 4 1/2 in., its overall height 40 3/4 in, and its total width 64 1/2 in., which compares with the 13 ft. 1 1/4 in. length, the 43 in. height and the 64 1/2 in. total width of the standard Lotus Europa.

Although the car has come to be known as the GKN 47D, this is merely because this is the registration number which was eventually found for the car. Clearly, it had been hoped to register the car as GKN 1 but no such vehicle existed with this registration number. There is unfortunately no space in which to record the full degree of detective work which went into the search for a suitable number once this information had been established, but eventually the number GKN 47 was tracked down to a tractor registered by a Sussex timber firm in 1942. Still more detective work established that the vehicle had now been broken up, which meant that the registration mark was no longer capable of being issued; a further complication, just in case this proved to be insufficient was that it is apparently not possible to transfer a registration from or to a goods vehicle!

There followed still more investigations. Eventually, the number GKN 47D was traced to a moped and this was ultimately exchanged and the car was not only registered, but christened as well!

The future of the car is uncertain as a production vehicle. Quite plainly, its limitations with regard to luggage space and its enormously high cost (which, undoubtedly. could be reduced by volume production) naturally make it difficult to assess as a practical road going car for the private motorist. A short test run on a racing circuit proved it to have outstandingly high qualities of road holding, acceleration; and an obvious ability to cruise at very high speed in comfort, with a very low level of exterior and mechanical noise, and with excellent ventilation. GKN are adamant that it is not intend at any time to produce the car in series and that it is merely to be used as a development vehicle for their own products.

Meanwhile all we can do is hope


Return to the 47D page
Return to the main car page
Return to the main photo page
Return to click Ben's Home page

Please send any comments, suggestions, complaints to seven+47@47d.org