Ruling the roost

As the F1 rule-book becomes more complicated, so is the technology the teams use to improve their performance. And understanding that technology is more complicated still.

Throughout the history of grand prix racing, the only engine rev-limit was that defined by its design and the laws of physics. For the first time in 2007, that limit will be defined by a regulation. Thou shalt not rev beyond 19,000rpm – almost 2000rpm slower than the highest-revving ’06 motors. It’s a command that’s indicative of a sport with an identity crisis.

In addition to the rev limit, the specification of most of the engine is frozen. Block, camshafts and much of the internal design have to be just as they were when the motor was homologated last year. Regulations already define the number of cylinders, vee angle, number of valves, cylinder spacing; even centre of gravity height and maximum bore size were prescribed – not to mention the types of material permitted. 

No scope for engine development, you might think. But you’d be only partly right. Changes to electronic mapping, water and oil pumps, valve profiles and con-rod design are all permitted, for example. Hordes of clever people and vast computer resources are being deployed to optimise the areas in which they are still allowed to work. Computational fluid dynamics (CFD) programmes are becoming more powerful at an exponential rate, helping engine designers understand what is happening inside their creations. 

Mechanical efficiencies from savings in frictional, pumping and windage losses – including churning all that oil – will surely follow, as will better combustion from ever-more refined valve design. The efforts of the surface finish specialists and lubricant and fuel chemists are going to be more critical than ever too. The importance of more efficient pumps to scavenge the air – making the internals of the engine a near-vacuum, so reducing aerodynamic losses as the oil is circulated – will be enhanced.

In one sense, the restrictions have just made development easier by specifying the areas in which it can be conducted. Intelligence resources even greater than before – when there were an almost infinite number of decisions to define which areas of development might give the greatest payback – can now be unleashed in a more targeted way. 

Ever-developing computing power allied to enormous manufacturer R&D budgets make for an awesomely potent development environment. Last year, despite regulations that largely defined the format of the engines, teams found as much as 55bhp – power increases of up to 7.5 per cent – from the beginning of the season to the end.

To give those numbers some perspective, we can look back to the 1970s before the manufacturers had returned to the sport – and before computers gave them such a powerful tool. At the beginning of 1970 the Cosworth DFV 3-litre V8 was producing 435bhp at 10,000rpm. Its rival, the Ferrari flat-12, was making around 450bhp at 11,500. Eight years later, these were still the top two engines – with DFV now offering 495bhp at 10,800, the Ferrari 520bhp at 13,000. That’s an average power boost of 1.5 per cent per year for the Cosworth, 1.9 per cent for the Ferrari – and this was at a time when the regulations were virtually free. Had 2006 development rates been applied to those 1970 engines, by 1978 we would have had a 696bhp DFV and a 720bhp Ferrari flat-12. 

It can be appreciated just what a running battle the governing body has given itself in tying to contain engine performance. A similar battle will soon be waged on aerodynamics – with the FIA looking into homologated body packages. Longer term, there is even talk of banning the use of wind tunnels. It may already be too late on that one. CFD power is increasing so fast that the tunnel will one day be superfluous. The beauty of a CFD programme over a tunnel is that it shows not just the aero figures, but how they are derived. It explains why rather than just what. The limitation is in the number of cells that current computing power allows, meaning that the whole car cannot be modelled in the necessary detail to properly link up the influence of one change over the whole. But that limitation is dissolving fast. Already, BMW has chosen to invest in extra CFD computing power rather than build a second tunnel. Banning the use of CFD programmes would be literally impossible to police.

The genie has long-since escaped from the bottle and is now running riot. Regulation restriction, no matter how stringent, will merely slow development rates. At some stage F1 has got to decide if it’s the ultimate race-driving challenge, the ultimate shop window for selling cars or the ultimate automotive engineering discipline. With a world as clever as it is now, the solutions to any one of those three are deeply painful to the other two.

There will probably not be a resolution as long as the participants are making so much money from it the way it is. Lose most of the money, and most of the problems disappear.