An engine that looks the same as it did 50 years ago can now breathe more easily, form a better fuel/air mixture and burn it more efficiently

Engines aren’t just about motive power. They breathe life and soul into a car. The thrill of acceleration and the satisfying response to your command are complemented by glorious sounds, tingling vibrations and – rather like a fine wine – a distinctive smell. But, like fine wine, some engines have improved with age. In historic racing, those engines are producing significantly more power now than when they were brand new. And that’s more than just a little intriguing.

It’s not a small increase either. Alfa Romeo’s twin-cam is undoubtedly a design classic, and not only provides prodigious power for its size but sounds and looks wonderful too. When Alfa campaigned in touring cars back in the 1960s, the 2-litre was tweaked up to about 180bhp, but a similar-spec engine today can squeeze out a positively exhilarating 210bhp on modern fuel with suitable cams and the discreet addition of mapped ignition.Formula Junior cars of 1000cc were producing a respectable 100bhp back in 1968 when the formula ceased, but the same racers battling it out in historic formulae today are producing nearly 30% more, which is quite an achievement without any drastic changes to the engine.
 
It’s not just the small engines: classic Cosworth DFV V8s, Ferraris, Bugattis and Jaguars are all finding new strength in their retirement. The much-loved and iconic Ford Mustang won the Indianapolis 500 in 1964 less than a month after its launch while, the following year, Shelby GT350Rs won no fewer than five of the six Sports Car Club of America divisions, all of which would indicate that it was fairly well tuned at the time. But at a recent auction a near-identical car boasted a dyno sheet showing 380bhp and 335lb ft from its 4.7 litres, a 9% improvement with age over the ’60s race-winning 350bhp.

Now, having spent a couple of decades in the car industry, I am only too aware that there will always be some tuners whose dyno test results are on the optimistic side, and any tuner who is making genuinely fast cars will be a little reticent to tell you their best-kept secrets. Yet the sheer volume of cars posting faster lap times and clutching plausible dyno sheets implies that something is different. You’d be forgiven for wondering what’s going on. And, in short, there’s quite a lot going on.

By far the most influential change since the ’50s is our understanding of the natural world – from chaos theory to quantum physics – so much of our picture of how things work has marched on to previously unimaginable realms. Most relevant to this subject are the revelations about gas flow and how to make an engine breathe effectively. Many of the tuning theories popular back in the good old days, when mechanics wore flat caps and suitable ties, were based on minimal data and spirited conjecture that, to be fair, was the best method available. Fast-forward to the present day and we find that every aspect of engine design is calculated to within an inch of its life: engines are flow-tested and run on a computer well before any metal is cast. It is remarkable to note, however, that even with complex computer-based modelling systems and a massively improved understanding of combustion dynamics, there is still a small percentage of the fierce turmoil in the combustion chamber that cannot be calculated. It’s just a little too chaotic.

Anyway, back to the plot. You will probably have heard of ‘gas flowed’ heads, but do you really know what it actually means? Some cheaper cylinder heads might only have had the ports sanded down and smoothed over, while the more reputable heads may have wider ports with a subtly different shape in order to influence the way the air is presented into the cylinder. Half a decade ago the top teams were using flow benches where, to put it simply, a big fan sucks air through the cylinder head being tested – the higher the flow, the more power could be had from the engine. Or so they thought. But it turns out that life is not quite that simple.

The complication arises because the flow through a port stops and starts very rapidly as the valves open and shut. Half a century ago, engineers had no means of accurately measuring what was going on in detail between each cylinder firing; they could only measure the average flow over several cycles.
 
It’s different now. Picture the scene with an open inlet valve: air is rushing into the cylinder, then the valve shuts but the air column is still moving due to its momentum and slams against the back of the valve. We now know that if you get the timing just right then when the valve opens again there will be a nice head of pressure there to force a little more air in.

This works because air is surprisingly heavy stuff. Each cubic metre weighs about 1.2kg – the air inside an empty shipping container weighs more than an average fully grown man. But, I hear you say, surely there is not that much air in an engine for its momentum to make a difference? Well, for every 100bhp you get through approximately 76 litres of air per second – that’s like drinking 133 pints of beer in one second.

Going back to the inlet port, the air’s momentum depends on its speed and, if you open the port out too much, the air speed drops, and with less momentum to give that helping push you can end up with less air going into the cylinder on the real-life running engine, despite having good flow bench results. This, with a little imagination, gives us a clue about how the efficiency of the inlet system, exhaust system, port design and valves is heavily dependent on camshaft design and timing.
 
In a further twist to the old logic, it turns out that a little bit of rough surfacing on the inlet port can be a good thing, particularly on carburettor-fed engines where fuel has a nasty habit of falling out of the air and dribbling along the manifold walls. Then a bit of rough surface encourages the fuel to vacate the metal and rejoin the air flow. So the cheapest of the ‘flowed’ heads would actually make the problem worse.

Applying this knowledge to rebuilding a historic race engine means that, even within tight regulations, valve seat angles, cam profiles and valve spring materials can be optimised to get a bit more wind through the engine. John Crabb from Piper Cams says: ‘The whole intake/valvetrain and exhaust system is considered to be one mechanism, where all the parts are designed to work in harmony. Valve seats, valve sizes, cams and gas-flowing the intake and exhaust are all designed to improve the flow path through the whole engine system.’
 
As well as improving flow, modern valve materials can be lighter, which allows them to move faster. This means that the valve spends less time tied up in the tedious process of opening and closing, usefully spending more time being fully open or closed. But just as importantly it allows the engine to rev higher before the valves start ‘floating’. Valve float happens when the valve is opened very fast and momentum keeps it lifting, losing contact with the cam lobe: in severe cases it can be flung into the piston face, which is potentially catastrophic. Lighter valves stay under control at higher speeds, and higher revs can mean more power, which is generally appreciated.

You might think that changing the valve material to something more modern is cheating, but actually the material can still be steel, just a lot purer, so the thickness of the valve head can be reduced without compromising strength or heat dissipation. Technology’s wonderful.

We all like to admire a nice set of pipes, and similar leaps of understanding have helped to improve intake and exhaust manifolds as well as silencers and air filter housings. A recent historic Porsche 911 race car had its power raised from 204 to 230bhp largely due to improved tubework that better complemented the way the engine worked. Matching exhaust headers and intake manifolds to the port ensures that the momentum is maintained, both for generating a pressure pulse at the inlet and a slight depression at the exhaust valve. Also, where runners join together, the pulses can be combined, either to amplify the effect (high peak power but narrow useable rev band) or to dampen the effect (lower peak power but wider usable rev band). Sometimes a slightly lower peak power figure can make a car faster on the track when the spread of power is more liberal, giving greater average acceleration in each gear and reducing the number of gearchanges needed per lap.

Sometimes a power gain can be achieved by a new tank of petrol. With all the adverts everywhere these days you may be aware that fuel technology has marched ahead quite dramatically. It is remarkable that, even with the same energy content and octane rating as older fuel, some modern ones manage to produce more power. This near-magical feat comes from a combined attack on many fronts: by modifying the way the fuel mixes with air, the accurately engineered blend of around 200 different substances can be tailored to ensure less fuel is wasted as unburnt hydrocarbons, and if more of the fuel burns in the combustion chamber rather than simply falling out of the exhaust port then you get more performance.