Blown Elise

How do you make a Lotus Elise outperform a £750,000 McLaren F1 supercar?

This was the request of Adrian Newey, the chief designer for the McLaren Formula One Team, who had the cheeky notion that he could own a car that could embarrass his boss’s best efforts with their exotic F1 road car for a fraction of the cost.

So what was done to turn this mild little road car into an awesome McLaren eater? For a start, Adrian new of Ken’s ability with engines and his knowledge of forced induction going back many years. So having decided to let him loose on the project, here follows Ken’s story of the development and pain over a period of nearly 12 months

Ken’s diary

The solution was not exactly simple. In fact, you could say that it was impossible, knowing the fragility of the Elise/K-series engine. Even in normally aspirated form it is not without its problems. For a start, something radical was needed to be done to the block and crankcase as well as rods, crank and pistons. The cylinder pressures were about to be quadrupled – yes, times 4! The torque had to be phenomenal.

After inspecting the engine in pieces, there were not many routes that could be taken. The cylinder head was fitted with through bolts that were screwed to a cast aluminium plate underneath the main bearing ladder frame so that the cylinder block was sandwiched between the bottom plate and the cylinder head. The through studs are just 9mm diameter and of the stretch variety so that they can be tightened to a certain torque that allows their springiness to hold everything together, even during expansion and contraction of the aluminium block.

Anyone who has tried to knock a nail into a fence has experienced the hammer bouncing back in their face because the fence is so springy. You have to get someone to hold a brick behind the fence so that you have something solid to hit the nail against. In a similar way, the forces about to be created in the k-series engine would be unchecked because of the way it was designed with the springy through bolt idea and low mass at the bottom end. To eliminate this lack of rigidity, it was a simple case of making a substantial steel plate to replace the lightweight aluminium casting and so give the engine a good firm, solid base. The long tension bolts were increased from a nominal 9mm to 12mm and located through the bottom of the plate, attached by nuts underneath, and up through the block and cylinder head. In order to get the bigger studs through, all the through-holes of both the block and head had to be re-machined. Where the main bearing ladder frame attached to the base of the block, all the boltholes were jig bored to enable the studs to perfectly locate and keep the main bearing bores accurately in line. The studs were made with parallel sections to act as solid dowels in all their positions so that the main bearing assembly was totally rigid. Coupled with the mass of the steel girdle plate, immediately underneath, made this area of the engine very robust and virtually un-burstable.

To assist the strength of the liners and block, stronger cylinder liners were needed that also located in a better engineering manner than the production versions. We re-draw these and had them manufactured. The block had to be machined to fit them and they were positively located top & bottom. The pistons were next up; the originals would be useless so we had some forged pistons made with a small bowl in the crowns to lower the compression ratio to just under 9:1.

Once all this engineering work was finished, we assembled the bottom end and figured out the lubrication. As we intended to dry sump the engine and fit a turbo, we had to run a different feed via the new girdle plate and take extra pipes off the girdle plate to create cooling oil sprays directed at the undersides of the pistons to try to keep them below melting point during maximum boost situations! All this was carried out and it looked really neat. The original wet sump was altered, converting it to a dry sump, so that oil could be scavenged out and through the oil cooler and on to the tank. The tank had an outlet back to the engine’s own pressure pump, which was big enough to handle the extra flow required for the increase in power.

The Elise is a bit tight for space in the engine compartment and there was little room for all the extra tanks. However, we made a tall cylindrical tank and positioned it in front of the engine and to the left (kerbside) of the engine bay. The idea was to take the bottom of the dry sump tank below the level of the original sump so that the tank could self prime to the oil pump and the tank would not bleed all its contents back into the engine “overnight” as happens in most dry sump arrangements!

Mounting the turbo was another challenge. It was hopeless fitting a large T3 in front of the engine so we fitted it behind and fed the exhaust pipe under the engine to the turbine entry. From there it was a relatively simple task to make the rest of the exhaust system, fit a catalyst that we made using a special “Johnson Matthey” metallic substrate core (we needed to be street legal).

The next problem to overcome was intercooling – cooling the induction charge. The Elise has 2 air inlets to the engine compartment, one at each side of the rear bodywork. As only one side was required for the engine intake, we could use the other side to feed air to the intercooler – except that we didn’t have enough space in the rear for the intercooler and Adrian’s toothbrush! We made a closed loop intercooling system by building a smaller intercooler core into a tank and flowing coolant around it and off to a water cooler via a header and small electric pump with ducting fed from the air inlet duct to the cooler. This sounds cumbersome but turned out to be amazingly efficient and we were able to keep the inlet charge below 45C during boost on a hot summer’s day.

Next, fuel! The original tank pick-up and fuel pump would be hopelessly inadequate so we dissected the tank and built in a swirl-pot, fed from the surrounding fuel and then pumped from the tank with a much more powerful Bosch pump, sourced from their very comprehensive list of pumps.

Once the installation was complete, we could remove the engine and test the car’s set-up on the dynamometer. This was done and the new engine management system programmed. In all about 4 days on the dyno’. Then the engine was dismantled to check for any wear or problems, rebuilt and fitted to the car.

The dyno testing had been a tremendous success and we knew that if the car’s installation was correct, that we had a “McLaren beater” on our hands. We didn’t quite exceed the bhp/weight ratio but nearly equalled it. However, we beat the torque/weight ratio by nearly 15% so we knew we could out accelerate the McLaren from rest even if we couldn’t equal the 215 mph top speed (although that was more a question of aerodynamics and final drive ratios and would anyway have been largely academic).

To cope with this massive hike in power, as can be imagined, quite a bit of chassis work was required. For simplicity, the Lotus sports suspension pack was fitted which included some nice new dampers. Also, Adrian decided to fit carbon brakes in his concern that he would not be able to stop the car. As an interim, we fitted larger discs and better pads whilst these were being made. The wheels remained original because Adrian wanted to make the car look as though it had just left the Lotus showroom. However, they were shod in some super sticky tyres that Adrian managed to wheedle out of one of the manufacturers. – It’s sometimes handy to have the clout of a senior F1 designer!

Once complete, the car went out for road test, the first being EVO magazine

As a spin off from the development, many of those problems experienced with the normally aspirated racers were solved with our technology designed to cope with the ridiculously increased stresses of the turbo car. There are quite a few Caterhams, for instance, riding around with these amazing cylinder block modifications.

Another spin off was a simpler turbo conversion, developed for general performance use, which also became very popular.

Back to case studies home page