Formula 1 Analysis by Rosehillpilot

"As soon as we got to the top of the hill in 1966, the rain started. I was behind Mike Spence. He hit the torrential downpour and went sideways into a barn. Water was running across the road and I also left the circuit and hit a small tree, finishing upside down. I had gas from my BRM going down the back of my neck. The car was lifted off me, and I got out and ran across the track to where Jackie Stewart had crashed. Graham Hill pulled over, so he and I removed Jackie’s steering wheel and helped to get him free. His tank was split and he was covered in fuel. Then we heard a helicopter overhead and thought it was a medical one, but it was John Frankenheimer’s cameraman filming the whole thing." Bob Bondurant

What a vivid and chilling memory of the old circuit at Spa. I have taken it from yesterday's issue of The Red Bulletin. This beautifully produced magazine is published on each of the four days at every grand prix. It is distributed free to everyone in the paddock so it is a bit of a status symbol to be seen in possession of the publication. Unless you work for McLaren that is, because Ron Dennis repotedly banned the bulletin from McLaren premises. It is rather jokey in a way that failed to appeal to Ron's well-known sense of humour last year. Those of us who cannot gain entry to the paddock at Spa this weekend can enjoy the magazine on the Red Bulletin F1 website.

Meanwhile back to the race. Has Vettell taken as much fuel as his Q3 time suggests? Will this be a master-stroke in what may be a chaotic race. Should Hamilton have taken a bit more than it seems he did? Will anyone apart from the McLaren drivers manage to warm up the Bridgestone's today?

As Robert Kubica said to his engineer after a run in the wet and the cold on Friday: "The car is nowhere....it just doesn't work!" This will be a difficult race for everyone.

Very technical stuff: more so than I first thought
I slept badly last night and some of this unrest was down to the mysterious inerter that is currently a McLaren team exclusive. Further thought has clarified my thinking about one aspect of the gadget at least. The rotary inertial weights must spin one way as the device closes and the other way as it opens again. The gearing will be fixed and there will be no freewheeling. The stored energy will be fully dissipated as the motion of the inerter stops. The weights will then be spun up again as the suspension returns towards ride height after a bump. I now believe I was wrong to say in my last posting that the inerter will add the effect of extra mass to the sprung element of the vehicle. The device adds what we might call virtual-mass to the unsprung elements of the car: the hub, wheel, tyre, and outboard suspension elements. But the inertial effects of mass are added without actually adding much real mass to the car, and none to the hub itself. The inerter thus makes each unsprung corner of the car to which it is fitted behave as if it was heavier than it really is, while adding only a small amount of real weight to the chassis. In effect it is like bolting a ballast weight to the hub, but the cost in actual weight is well within the margin to which F1 cars have to be ballasted to reach the minimum overall limit. Thus the only cost is the loss of an equivalent weight from this moveable ballast.

Now why would this be a benefit? Conventional wisdom suggests that the unsprung weight (by which I actually mean mass, but let’s not be too pedantic) should be as low as possible. The ratio between the sprung weight (the chassis, engine and driver etc) and the unsprung weight (basically the wheels and suspension) should be as high as possible. In this way the sprung element can be regarded as a stable platform around which the wheels move up and down in response to bumps in the road. The problem with the modern cars is that they have had to become very stiffly sprung to cope with the vertical aerodynamic loads, the huge cornering forces in fast curves, and the necessity to avoid much variation in ride height because it would produce huge changes in underbody downforce. The front suspension movement is now usually provided simply by the flexibility of the carbon wishbones themselves. Adjustment is by ‘packers’ placed in the dampers that would feel in your hand like completely solid washers. The immensity of the forces involved and the tiny range of movement is astonishing.

The softest suspension springs that the cars now sit upon are the tyres themselves. The vertical spring rate of the tyres is much softer than the spring rate of the suspension and more vertical movement occurs in the tyre than in the suspension. The spring rate of the tyre varies with tyre pressure. I have discussed earlier how this can produce very problematic handling while the tyres are below optimum temperature. There was a time in the past when suspension was relatively solid, though not by today’s standards. Tyres were tall and narrow and tyre pressure changes were used to fine-tune the balance of the car. Sir Stirling Moss has said that he was sensitive to changes of less than one psi. But now you cannot do this deliberately as the safe range of pressure when the tyre is fully warm is very small. Over or under pressure will cause localised tread overheating as well as risk damage to the tyre structure and failure of the carcass.

So, to put it simply, the tyre is the suspension medium primarily of the wheel and hub. The rest of the car can be regarded as a secondary chassis (shades of the notorious swan-song Colin Chapman Lotus!). The bounce frequency of the tyre is the real measure of its effect as a spring upon the car. So in theory this could be adjusted by manipulating tyre pressures, but I have argued that this is now very limited by the nature of the tyre itself. It can also be altered by adding weight (remember it is mass) to the wheel hub. The more weight the lower the tyre frequency and the ‘softer’ the springing through the wheel.

You certainly wouldn’t want to add big weights to the hubs. But the inerter produces the same effect without forcing you to do such a crude thing. It produces the inertial effects of added mass and thus changes the frequency. Inerters with different gearing could be used to adjust the amount of virtual-mass that is added. The end result might be that the unsprung mass of the tyre itself, which is the contact patch, may patter less on the road surface and thus optimise grip. At the same time the handling balance of the car could be adjusted by changing the spring-rate of the tyres. An increased inerter effect at the front would lower the frequency of the front tyres and could reduce understeer. The springs themselves, the dampers, and the roll bars, can remain massively stiff to control chassis movement when pulling huge lateral loads. But the McLaren race engineer may be able to dial out the troublesome low-speed corner understeer by adjusting the inerters. Nobody else on the grid has this option.

The inventor of the inerter claims in technical papers that its use separates roadholding effects from ride effects. I presume he believes that high frequency oscillations generally affect the tyre's contact with the road, while lower frequencies affect the human occupant. In racing we tend to call ride effects "handling", as it is the secondary effects of ride that the driver feels as the balance of the car. So I am disagreeing with the sage himself when I suggest that this device, like almost every other suspension component of a racing car, actually affects both grip and handling. And this is ignoring the fact that the real inerter will have both stiction and friction which will introduce a damping as well as an 'inerting' effect.

This is all speculation from the armchair, but it could explain the noticeable strengths of the McLaren. And I think that the inerter might even be helping to increase the rate of front tyre warmup, which is both a blessing and a curse for the team depending on the weather and the available tyre compounds…..

Two quick answers on qualifying issues
Hasan asks in a comment yesterday whether I can use my method to calculate hypothetical fuel-corrected practice times for Monaco and Montreal. Alas this is impossible. Both Lewis Hamilton and Kimi Raikkonen had to make early pit stops at Monaco because of damage. Fuel was added to both cars I believe. Thus we cannot tell how many laps of fuel they carried in Q3. There are then too few pairs of cars in Q3 from which to estimate the time penalty for a single lap. In Canada the whole race was interrupted by the safety car and the leading runners all took that opportunity to pit. Again this removed the possibility to deduce the number of laps of fuel they carried during Q3 and the calculation cannot be made.

You could say that Monaco is probably a lower energy track than Hungary and that Canada may lie somewhere between Hungary and Valencia in this respect. I have calculated the percentage change of total laptime suffered for an extra lap of fuel and my figures show a lap time penalty of 0.202% at Hungary and 0.26% for Valencia. You might then simply guess it’s 0.98% at Monaco and 0.23% in Canada but you would still need to know the fuel loads in Q3. And it would still be guesswork and not any sort of reality-based calculation.

The teams themselves have the exact figures. Whatever the time penalty on their own car for one lap of fuel, it will be almost exactly the same for everyone else. As I have pointed out before they use the Doppler-shift of the engine note to calculate the rate of acceleration of the other cars on a straight where the cars are power limited and there are no traction issues. It is easy for them to then calculate the all-up weight and thus the weight of fuel on board. The only guesswork they will have to use in the calculation is the assumed engine power of their rivals. I have a foot in the door of one team and a toe in a couple of others, but I am certain that nobody would leak this sensitive data to me.

Next I noticed a comment on a forum somewhere asking why Kimi Raikkonen could not simply do two warm-up laps before his Q3 runs to heat up his tyres. The rules make this a high-jeopardy strategy. I think that even with the engine turned down the two out-laps (which have to be fairly vigorous) might burn about the same fuel as a single hot lap. The in-laps will use only a small amount of fuel but it all adds up. Even when you only do a single out lap for each run the car should be faster on its second attempt simply because it is carrying less fuel, even if no adjustments are made to it. If you begin to increase the fuel differential between the first and last run of Q3 the first run will become even slower. Thus the first run will be less useful in terms of protecting you against driver-error on the second run, or any other problem. The team would also have to monitor the potential traffic in two laps time. Until the advanced crystal-ball system is introduced it is easier to guess when to go out if you are going straight into your fast run immediately after the out-lap.

I am aware that this has been a very dry posting but I am sure there will be plenty of colourful and controversial things to write about after the race in Belgium. Especially if the weather is as bad as the forecast is suggesting. I just hope that the inerter gadget will stop giving me sleepless nights. At least I can take a break from worrying whether it is correct to use 'affect' or 'effect' in every other sentence......

Ciao

The historic circuit at Spa-Francorchamps
There are only a couple of days to wait before the cars are released onto the circuit at Spa. This historic racetrack was once made up entirely of public roads and included many ultra-fast corners as well as all the roadside hazards that you might expect if you were driving through Belgium in the middle of the last century. Some of the best images of the old circuit can be seen in the classic John Frankenheimer movie ‘Grand Prix’. The film-makers shot the opening lap of the Belgian Grand Prix from an Intercontinental Formula chase car. F1 had dropped to 1.5 litres, so the 2.5 litre camera car was easily able to keep up with the field (by all means correct me if I have these details wrong). I believe that the car they used was actually the first McLaren F1 prototype with a monocoque constructed of balsa-wood and aluminium sandwich material. Perhaps it was driven by Frank Gardner? This was the 1966 race and there was a thunderstorm on the opening lap. Many cars crashed and the brief sequence used in the film should be watched in slow motion if you have it on DVD. There are crashed cars all over the place. Jackie Stewart was trapped in his BRM and had to be rescued by his team-mate Graham Hill. It was a miracle nobody was killed. In 1960 Mike Taylor was seriously injured when his Lotus crashed into the forest after losing its steering, and Stirling Moss was badly hurt in another crash in practice. Chris Bristow and Alan Stacey were both killed in the same race. The stakes were too high for common-sense, though the circuit was awesome and must have given the drivers a real sense of going for a long fast drive through the countryside. The old track was almost 9 miles long, and although I have never raced at Spa I am sure that as an adventure it was was second only the old Nurburgring (the Nordschleife). The new shortened circuit retains only a few of the old classic corners but it still takes a dramatic path through the Belgian forest and farmland.

In view of the long series of terrible accidents in all levels of racing, and the impossibility of moving farmhouses and the like, the circuit had to be changed eventually and the shortened modern version is the latest iteration. It is still fast and flowing and a real test of both machine and driver. There are a couple of potential overtaking spots but generally it has to be said that the dauntingly fast corners prevent today’s F1 machines from following closely behind each other. This is a very high-energy circuit and Bridgestone have decided that the tyres they took to Spa last year were marginal. This year they are taking harder compounds and this will help McLaren and hinder Ferrari as we have seen all season. However, the very long out-lap should reduce the relative weakness of the Ferrari’s in qualifying and the dry race speed of the cars should also be comfortably faster than McLaren.

But the weather forecast is grim. The continuing and unforeseen sub-normal temperatures that have prevailed right through the European summer will continue this weekend, and the temperature is forecast to be colder on Sunday than on Saturday. Overcast skies will also reduce the track temperature. All this is bad news for Ferrari and perhaps for BMW too. But the forecast for Sunday also includes rain. This will make the result much more of a lottery. Once the cars are on wet tyres the Ferrari’s should be competitive, but if there are many laps run in light rain on the soft or medium grooved rubber the Scuderia drivers may suffer from a desperate loss of grip just as they did at Silverstone.

Traction-control revisited
The tremendous advantage enjoyed by McLaren and Lewis Hamilton in particular at the wet race in the UK gave rise to much speculation about the use of variable engine mapping to make the power delivery more docile. It was suggested that some clever form of steering wheel selection was being used by McLaren to switch to an appropriate map at each gearshift or perhaps at each corner. The team also seemed to enjoy a huge advantage at Hockenheim in cold and slippery conditions. But since then Ferrari has done well (engine failures aside) to lead the McLarens at two circuits that were thought to suit the Woking team. Has Maranello upped its own game in this respect or has Charlie Whiting had a close look at the McLaren system and asked for it to be removed?

My money is on the former scenario. Recent in-car shots have shown a very high workload for the Ferrari drivers as they make constant rotary-switch selections on their steering wheels all around the circuit. Astonishingly Felipe Massa was making adjustments as he pulled alongside Lewis Hamilton after the start in Hungary. What a moment to be fiddling with something inside the cockpit. We cannot know whether Felipe was selecting the next engine-map to optimise traction out of turn one, or whether he was adjusting the brake balance to allow him to brake really late into the corner on cold tyres, but he was certainly busy at what was a very intense moment. Shots from inside the cars at Valencia have shown a similar flurry of activity. So everyone seems to be doing whatever it is, even if McLaren may have developed a more driver-friendly interface with which to make these selections.

McLaren’s real secret-weapon?
It is now public knowledge that the rumoured J-damper that Renault was accused of learning about from stolen information by McLaren last year (in a very exaggerated PR exercise mounted by Woking to obscure other issues) is in fact an ‘inerter’. What is an inerter you may ask, and I have been struggling to get a grip on this very secret device myself. Firstly it is quite different from the ‘mass damper’ run by Renault two seasons ago, and which they were eventually required to remove. Renault’s system was based on the innovative device that was used in the early Citroen 2CV. Yes, you did read that correctly: the wonderful old 2CV.

In much the same way that Nelson Piquet surprised the press years ago by saying that the sophisticated active-ride system of his Williams was much the same as the hydro-pneumatic suspension pioneered by Citroen on their DS road cars years earlier, I am saying much the same of the secret mass-damper in the Renault nose.

Citroen controlled wheel patter by fitting each hub (or perhaps only the rear hubs, I’m not certain) with an inertia damper consisting of a spring-mounted 3.5 kg iron weight inside a vertical cylinder. Renault used the same concept by mounting a sprung ballast-plate in the nose to damp out the bouncing that makes the cars tricky for the driver after mounting a kerb, for example. Tall buildings in earthquake zones or where there are high winds now have huge mass dampers to prevent a harmonic oscillation building to a crescendo and bringing down the whole structure.

The Cambridge University patented system used by McLaren is different. The Penske shock-absorber people have been working with the team since 2003 to develop the device. It is mounted like a damper and probably looks like a rather fat version of a damper. But instead of simply resisting closing and opening like a damper, which is tuned to prevent harmonic frequencies of both chassis bounce and wheel patter, the inerter spins up a flywheel as it is closed. Think of it as being like one of the toy spinning tops that is spun up by pushing down on a vertical plunger. Thus it will resist acceleration towards the closed position but once moving will not resist a constant rate closure. What happens on rebound is unclear. Either the spinning flywheel will freewheel, or it will strongly resist rebound, or perhaps the stored inertial energy is recovered as the inerter opens again, actually accelerating the recovery to ride height. The flywheel could thus reduce the time taken to return to ride height and may also help to reduce porpoise-oscillations. With an inerter it may be possible to run lower spring rates. Another way of looking at things is that the inerter may make the car ride as if it had a much heavier sprung mass than it has in reality. But we need a little more mechanical detail to get any real picture of how this works.

The inerter certainly does not behave like a damper and dampers will have to be fitted in addition to the inerter. This is a new device that will have a complex affect on ride, roadholding, and handling. Malcolm Smith, the co-inventor of the inerter draws a complex analogy will electrical circuitry. He argues that while the damper is analogous to a resistor, the inerter is analogous to an electrical capacitor. This certainly seems to say that the energy stored as the inerter is compressed is then returned when it extends again. He develops this idea with some complex mathamatics (well, complex for me) and a bold development of what is finally an argument by analogy. It was not helpful for me and I struggle to understand the operation of the inerter in more common-sense race engineering terms. But however it works it may be having a visible affect on the McLaren.

Most F1 cars struggle around slow corners with levels of understeer that would be unacceptable to the driver in many lesser categories. It is the price that is paid for good balance in the faster corners when the downforce builds up. The centre of gravity of these vehicles is in a compromise position with regard to mechanical balance as everything has to be optimised for the all important high-downforce situation. But I have been tremendously struck by the agility of the McLarens in slow corners and when kerb-hopping. I commented on the perfect balance that the cars seemed to enjoy after last year’s Monaco grand prix. They seemed to have an optimum neutral to oversteer balance in both fast and slow corners and this is very difficult to achieve. In Valencia the in-car shots from Hamilton’s car were very impressive. There was no shortage of front end bite into the slow corners. In fact it looked as if it might have been better to have a touch less oversteer in places. I have to add that Lewis was driving beautifully in a car that was matched to his style.

Now that the story of the inerter has been made public it is impossible not to wonder whether it is the advantage gained from this device that allows the McLarens to behave in this way. Finally, tyre temperature issues have a greater affect. Although the McLarens were spectacularly driveable in qualifying, during the race in Valencia they were running the compounds too hot and the in-car shots then revealed understeer deadening the responses of the cars all around the circuit. But even the Ferraris on optimum-temperature tyres could not match the poise of the McLarens when they were running with perfectly heated rubber themselves.

A few miscellaneous comments
Interesting that the Red-Bull drivers should be talking down their Renault engines. Toro Roso has certainly been faster than the Bulls but the team itself, and both of its drivers, are probably contributing to their pace. The chassis is by no means identical to the Red Bull as the packaging of the Ferrari engine inevitably produces different constraints around the rear of the car and the diffuser. Different ancillaries will also produce different weight distribution possibilities.

The Mercedes engine is probably the most powerful on the grid at the moment. But it is only a few years ago that Rory Byrne was warning the Ferrari team that Renault was headed towards “all the nines”. 900 horsepower, 90 kilos weight, and 19,000 rpm. It seemed an almost unbelievable combination at the time.

The latest warning Rory will have given his colleagues in the Scuderia is that Ross Brawn must have a shopping list of engineers and aero specialists at Maranello. And Honda may be happy to stump up whatever it takes to get them to defect. Unless, that is, the FIA successfully introduce genuine budget caps to control and equalise all team spending. As the whole world sinks towards recession it has to make sense….

A last thought on Valencia. Ferrari may have gambled and won with a bold strategy to sidestep their lack of pace in cool qualifying conditions, but I am prepared to bet that they actually wanted to get Massa out in front of Kovalainen after his first stop. They just failed to do this and as a result it was Kimi Raikkonen who was slowed when Massa came out of the pits. Just another failure by the team to hit the centre of the sweet spot in Spain. But they still won a race that everyone expected to go to McLaren.

You have to compare and contrast the absurdity of the TV commentators’ bias as they bayed for a penalty to be imposed on Massa for not actually impeding Adrian Sutil. In stark contrast was their failure to accept that there might be any penalty imposed on Lewis Hamilton when he tail-ended Raikkonen in Canada. I have to say that Adrian almost made matters worse in Valencia by braking. Mind you, until Felipe turned away from his path it looked as if contact was inevitable and Sutil’s action was instinctive. But he took a moment or two to get back on the power and for a second it looked as if the two cars would end up wedged side by side as they funnelled into the narrow pit-lane exit. While on the subject of the TV commentary team I have to say that I will Miss Mark Blundell next season. He is honest and calm. His colloquial style of speaking may confuse non native-English speakers, but it is a very vivid vernacular. Miles better than anything from the superficially more articulate and very much more political Martin Brundle.

That’s enough until after the race. The little town of Spa is a bit dull to be honest. There isn’t even a decent pub as the British would understand it, and pavement cafes may be a little waterlogged this year. But you can celebrate the victory of your favourite team with many glasses of the excellent Leffe, of drown your sorrows very quickly if things go wrong with one of the much stronger ecclesiastical beers that make Belgium famous. And if things go really badly there is always the consolation of chocolate. Enjoy yourself if you are there….

Ciao

I certainly caused a lot of excitement when I published my attempts at correcting the Q3 times for fuel load. I welcome and appreciate comments of course, but I do not want the comments to become a forum discussion. I have been in the habit of occasionally replying to comments by adding my own replies. I will try not to do this in future, but I will respond to the issues raised here in the blog itself, or in private emails.

Firstly, the main purpose of the attempt to fuel-correct the times was simply to highlight four things.

1. The Q3 times do not reflect the ultimate pace of the cars and drivers, but are influenced by fuel load to an extent that far outweighs the relatively small differences between drivers over a single qualifying lap.
2. Ferrari has a huge problem in qualifying if track temperature is colder than expected.
3. Ferrari has made some radical strategy decisions recently to get their cars higher on the grid than their real pace in qualifying conditions would allow if they took the same fuel as most of their rivals.
   
4. In Valencia it is clear that Ferrari were trying to get Kimi Raikkonen through to the finish by running his engine gently in the race, and there would have been no benefit at all if he had taken a light fuel load to start higher up the grid in this particular race.

How I correct the Q3 times for variation in fuel load
There has been a lot of speculation within some forums about my method, and a lot of heated debate as to whether I have any idea what I am talking about and whether the numbers mean anything at all. It is absolutely true to say that if you know the fuel burn of the cars over one lap, and also know how much the car is slowed by every kilo of added weight, you can then deduce the correction that should be applied. The teams do this in their pre-race computer simulations, and further refine their figures once they are actually running the cars in practice. By Sunday they will have a pretty accurate idea. Throughout practice, and particularly when testing between races, they all mask the relative performance of their cars. The top teams almost always run fairly heavy to keep their potential hidden, while the tail-end teams sometimes run very light to please their sponsors and raise their morale.

This is a very secretive sport. If we had access to data from one of the top teams we could correct all the Q3 times for fuel very easily. Believe me, this is not data that you will be given by anyone, and certainly not information that you will find by searching the internet for anonymous estimates. My method is based entirely on the observable performance of the cars in Q3 and in the race. You can make the same calculation yourself. Here is the method:

1. Get the official Q3 times and the lap-chart of the race from the FIA website. It is important not to use the indicated laps-to-go information on the TV to record the lap on which cars stopped for fuel. The display shows which lap the leader is on, and this can make it seem that cars have completed an extra lap before their stops. Avoid this error by using the official lap chart.
2. Calculate the time difference between the two drivers in each of the teams that got both cars through to Q3.
3. Compare the number of laps each of these drivers completed before they stopped for fuel.
4. Divide the time difference by the number of extra laps run by the heavier car. This gives a raw time difference per lap of fuel. But obviously one driver may have had a better Q3 lap than the other so we must continue to manipulate the figures.
5. Add up all the figures for one lap of fuel that you have for each team and take the average. We will have to make the assumption that some drivers on heavy fuel will have a better qualifying lap than their lighter team mate and that the opposite will happen within other teams. The more pairs get through to Q3 the greater the statistical probability of this being true, and the closer this average lap-time difference will be to the theoretically correct figure.
6. Apply this average correction per lap of fuel to all the times recorded in Q3 according to however many laps each driver ran longer than the first to stop. If people have accidents and have to pit early for repairs, or come in for wets because it has rained, you cannot get valid data on fuel load. So this calculation cannot be completed at every race.

A discussion of these results
The first objection I noticed was that I must be doing something “unscientific”. Well, as far as that goes this is an appropriately scientific method in that there is a primary hypothesis that cars go slower in Q3 in a way that will be proportional to the number of laps of extra fuel they carry compared with their lighter rivals. The truth of that hypothesis is already well established and the figures I derive from my method bear it out just as you would expect. The big problem with my calculation is that it is statistically weak. There are few teams that get both drivers into Q3, so we are dealing with a very small set of results. Therefore the assumption that the specific contribution made by the drivers will average out over these few teams is rather optimistic. There is quite a high chance that all the heavy fuel drivers might happen to have better laps than their lighter team-mates, thus giving a result that will give too low a correction, or that the light-fuel drivers will all have ‘better’ laps and the result will give too high a correction. The final fuel-corrected laptimes are therefore not ‘statistically significant’ and cannot be seen as solid truth. But I am fairly confident that the correction is within the same order of magnitude as the top-secret calculations that the teams themselves use.

In their press releases after the Hockenheim race Ferrari announced that the penalty per lap of fuel was 0.35 of a second. This is actually a much larger correction (in terms of percentage lap-time) than I applied in my calculations for Hungary and Valencia. This is probably partly because of the high-downforce nature of the Hockenheimring, but obviously also because Ferrari would not have made public their exact figures. There was an agenda behind the announcement and it was not an attempt to give their secrets away to the whole world. As an order-of-magnitude indication I think that it does support my own results.

Next I note Martin Whitmarsh’s surprise statement on TV that Heikki Kovalainen was "out-qualifying Lewis [Hamilton], when you correct for fuel". I presume he meant sometimes and not always. This is supported by my own calculations, which suggest that the two McLaren drivers are very closely matched.

The conventional rule of thumb that is sometimes stated publicly is that the cars are slowed by about 0.04 of a second for every kilo of weight that is added and burn around 0.67 litres per kilometre. These are gross approximations, but again they do seem to support the numbers that I derive from Q3. In fact the penalty depends on how much aerodynamic load is being used in the fast corners. Frictional grip is independent of mass as added mass increases weight in exact proportion. Thus back in the days before high downforce the cars were only slowed on the straights by added weight: a fixed amount of power was having to accelerate a higher mass. Braking and cornering were friction limited and added weight had only secondary affects on tyre-temperature. Now that the cars generate huge downforce, their grip is determined largely by that rather than by mechanical friction. Adding weight will slow the car in fast corners as, in effect, the power from the aerodynamics will have to accelerate a greater mass around the corner. Low energy tracks will have less penalty per kilo than high energy tracks, where the affect of weight will be larger. Basically the affect of weight on laptime is proportional to the downforce level required by the circuit. It is circuit dependant rather than chassis dependant. Every car will be slowed by about the same amount.

The issue of safety cars and extra fuel to protect from their deployment has also been raised. This really is a red herring. All of the cars will have to make their stops with enough fuel remaining to complete a couple of slow laps just in case the pit-lane is suddenly closed by a safety-car period at the moment they planned to stop. To run heavy but stop early for more fuel is not a logical system. It will guarantee a slower total race time. Nor will it offer any protection from disruptions that are random events. If you make your stops with several laps of fuel left in your tank you are adding seconds to your race time. The only way to plan strategy for a safety period is to know in advance when it will occur. So far nobody has developed a suitable crystal-ball. This whole issue entered the folklore of the sport when Ron Dennis began to advance concern over safety car disruptions to deflect probing questions about his choice of fuel strategy between drivers. And that all started when he made the mistake of saying that his team controlled his drivers “through strategy rather than team orders” after Monaco last year. Absolutely true, but Ron should never have said it.

An interesting objection to my figures was the observation that if you apply my correction for Valencia (which is 0.258 seconds per lap of fuel) you arrive at a theoretical laptime for almost dry tanks that is unbelievably fast. Well this does appear to be the case, and is even more true of some of the suggested “rule of thumb” corrections. I would suggest that with the current qualifying rules we may have lost sight of the ultimate potential of a top F1 car on fresh tyres and empty tanks but there must be something more going on. The lap-time correction is possibly non-linear with regard to total weight. If you work it to correct fuel loads that vary by a few laps all may be well, but if you try to apply it over 15 or more laps there seem to be gross inaccuracies.

Finally, I was trying to illustrate a point in the blog. It is certain that most people greatly underestimate the affect of fuel load in Q3. It is the most important part of the final qualifying session that I have likened in the past to a game of poker. Ferrari finally revealed their hand in Valencia when Felipe Massa stopped so early, and Martin Whitmarsh was left looking absolutely stunned on the pit-wall. The one thing that my calculations certainly demonstrate is which drivers are the least affected by extra fuel, and which are more affected. Kimi Raikkonen loses less time per lap of added fuel when compared to his team-mate than the average heavy runner compared to their team-mates. This certainly does seem to indicate the same speed differential as is reflected in fastest race-lap comparisons. Whatever his problems on cold tyres Kimi is undeniably quick.

Incidentally, the fuel-stop was re-introduced to modern F1 when the rising levels of downforce began to increase the gain in speed enjoyed by running with a low fuel load. The first man to recognise this was Dave North, who was Gordon Murray’s assistant at Brabham at the time. Dave is a very quiet figure deep within F1 and later went on to work with Ron Tauranac for a year before rejoining Murray at McLaren. He finally went to Renault.

There is a wonderful story to be told about the way in which the Brabham team learned exactly how much downforce was being generated by the first Lotus venturi ground-effect cars. The venturis had been regarded as a pointless gimmick at first. But this is a bit too secret to reveal in public, even though it was accidental and involved no subterfuge.

Anyway I hope that my fuel-corrected times have provoked some thought. The comments and criticisms have made me think too. I remember the title sequence at the start of the old TV ‘Kung-Fu’ series:

Q: “But now you have taught me everything you know why must I stay with you, Master?”

A: “What makes you think that I have taught you everything I know, Greenhopper?”

Ciao

PS: ref the comment from Hasan posted below this entry, I think that Ted Kravitz was simply repeating something from a post-race McLaren press briefiing. There were two tenths between Lewis and Felipe in qualifying and their fuel load differed by two laps. Maybe this led to the thought that each lap of fuel generated a penalty of 0.1 seconds a lap. I am certain that the true figure for Valencia is at least 0.2 of a second. So the story of the race is that (on fuel-corrected time) Ferrari was at least 0.2 seconds slower than McLaren in the cool conditions of Saturday qualifying. In the race Felipe was almost 0.2 seconds quicker than Lewis. So there was a 0.4 second difference between the relative pace of the two cars in warm and cool conditions. Ferrari had to run a very bold low-fuel gambit to get to the front of the grid and exploit their superior race-pace. I think that the same thing was attempted in Hungary but the car was not run quite light enough to capture the pole. Massa did manage to jump Lewis at the start in spite of this, and could then pull away.

Perhaps I got a little too excited at the sheer novelty of the new circuit in Valencia. As it turned out this new track was still pretty difficult as far as overtaking is concerned. The high speed curves leading into so many of the braking areas removed the possibility of overtaking from many of the slow corners. If the Hungaroring is a ‘point and squirt’ circuit I suppose that this one is a ‘point and flat-out blast’ sort of a place. Nonetheless this must be a very exciting ride for the drivers and the last run through the ultra high speed left and right flicks towards the final slow entry onto the pit straight must feel like an amazing trip. Okay, so it is all easily flat, but perhaps not next year with the reduced downforce. And we can only guess what it might be like in the wet.

The safety issues of a circuit like this are interesting. The lack of run off areas alongside some of these very fast corners is not a worry in itself. The cars would hit the wall at a very shallow angle. The impact would be easily survived by both the safety structure of the cockpit and the driver it contained. More worrying might be the way the car, now transformed into a high speed sledge with no steering, would then be funnelled down the track all the way to the next slow corner. This could be a rather high-stakes game of pinball as the runaway wreck skittered through the cars braking and turning into the next tight turn. In the event it did not happen this year. Valencia is an exciting place but the shortened version of the Hockenheimring still has my vote as the best of the new-age circuits, and it is certainly the one that allows even the present F1 cars to race each other properly.

Valencia is very like some of the temporary North American circuits that the indycars used to race on. Many of them were constructed on derelict industrial land. The facility at Valencia has been rushed into existence for this year’s race, but it is a bold project that will look better and better as time passes and the details are completed. The restoration and conversion of the old fish-market building into a spacious garage area for the teams with VIP spectator areas above has created a beautiful and unique focal point.

David Coulthard observed, during a rather choppy ride on the Red Bull sponsored yacht that unsettled the stomachs of some of his fellow pilots, “You can tell it isn’t America because people aren’t eating hamburgers”. No indeed: the fish market may have moved to a utilitarian modern shed-like structure, but the cuisine of the city will be full of exotic sea-food never seen by the British. And parts of fish that we might never imagine eating are essential ingredients of many local dishes. To visit a Spanish fishmonger is to step into a breathtaking display, only rivalled by the markets in Seattle, Sydney, and of course the daddy of them all in Tokyo. To find somewhere to actually sample all this wonderful food is a bit more difficult. The average Spanish restaurant caters to tourist tastes by simply frying the usual suspects, or by producing one of the ‘kitchen-sink’ paellas that visitors expect. Local knowledge is needed to sample authentic local cooking.

Cool overcast weather hit the circuit on Saturday. This was totally unexpected as the city is usually sweltering in relentless sunshine and temperatures in the mid thirties at this time of year. The sun managed to return on Sunday and temperatures rose enough to get the Ferrari tyres working again, but qualifying was an unexpected and unpredictable challenge for everyone. Expectations were confounded and it is lucky that Bridgestone brought such soft compounds to the race. If they had supplied the harder tyres that most teams thought would be needed it would have resulted in utter chaos. Perhaps the Japanese have developed a really good crystal ball?

Qualifying times corrected for fuel load: the real story of speed differentials
The result was that this really was a race of two halves with unexpected difficulties complicating qualifying and race-day finally bringing a somewhat more expected pattern. Once again I have corrected all the Q3 times for fuel load. The method I use gives the best possible result without access to the Doppler-shift engine sound analysis that the teams use to finesse the same calculations. Admittedly they are trying to guess fuel load before the race. I know the relative fuel loads as I have the easier task of working things out after the event. So it may just be that this set of fuel-corrected times is more accurate than anyone in the paddock had access to before the race. What a very pleasing thought for an analyst on a shoestring budget. Here is my best estimate of the times that each driver would have set in Q3 if they had all had the same fuel on board as the lightest car. Which was Felipe Massa’s Ferrari:

Robert Kubica 1’ 38.619
Heikki Kovalainen 1’ 38.649
Lewis Hamilton 1’ 38.684
Kimi Raikkonen 1’ 38.715
Nick Heidfeld 1’ 38.828
Felipe Massa 1’ 38.989
Jarno Trulli 1’ 39.279
Nico Rosberg 1’ 39.433
Sebastian Vettel 1’ 39.627
Sebastien Bourdais 1’ 39.977

This is a very intriguing set of hypothetical times. I also think that is an utterly plausible set of data. Remember that the teams themselves have a good grasp of the relative equal-weight potential of all the cars. They spend a lot of time and money gathering this information and processing it through their computers, whereas I have spent some time but no money at all. The unexpected weather on Saturday would have been a wild card, but the principle remains the same: tactical fuel decisions would have been taken on a fairly sound guesstimate of these relative potential Q3 times. Everyone wants to be high up the grid, but a low fuel start with a car that was not going to have great race-pace in the warmer weather conditions on Sunday would condemn you to simply fade away down the order in the race itself.

First let’s look closely at these corrected times in which less than one tenth of a second seperates the fastest four drivers. A great Q3 laptime from Kubica. Heidfeld is a couple of tenths slower, possibly because of his greater dislike of the wayward feel of the cars on cold tyres. McLaren not only very strong in the cool weather but enjoying very equally matched drivers. The team is putting everything behind Lewis. As the equipment is equal, and once again Heikki is a shade faster (by three and a half hundredths!), the team had to achieve the desired outcome through strategy. Heikki was given a hefty three laps of fuel more than Lewis in order to drop him behind his team-mate on the grid. Vettel shows the expected comfortable three tenths margin over Bourdais. The fuel loads they carried in the real Q3 reflect the logic of their situation. Vettel carried fuel for an optimum strategy. Bourdais carried an extra lap of fuel because he had to avoid interference with Vettel’s first stop without being too penalised himself.

When we come to Ferrari things get much more interesting. The first thing that has to be said is that Kimi Raikkonen was faster than Felipe Massa in qualifying once you correct for fuel. More than half the press corps do not understand the issue and the others cannot be bothered to think it through. But this was yet another qualifying in which Kimi started heavier than Massa and therefore ended up behind him on the grid in spite of having driven faster. As Kimi is suffering a hugely hostile reaction from the press, and accusations of ‘poor qualifying performance’ (which has an objective basis but is actually a rather complex technical issue), I have to say that it is just as well that he is such a phlegmatic Finn. Just remember how Fernando Alonso reacted when he found himself in a similar position with McLaren. So why is Ferrari handling Q3 in this way? I feel that there are two possibilities in Valencia. Actually, I believe that both of these things are true but let’s look at them one at a time:

Ferrari was pragmatic in the face of their engine worries
In Hungary a con-rod failure destroyed Felipe’s engine and robbed him of an excellent win. It was that engine’s second race. Kimi was on a fresh engine in Hungary. So this was Kimi’s second race with an engine that may have been in exactly the same spec as the one that blew in Hungary before the end of its second race. It is quite probable that the Ferrari team were very fearful that Kimi’s engine was not going to last the distance in Valencia. I had considered voicing this concern before the race, but decided not to make it public. It seemed such bad karma.

Under these circumstances it would be perfectly logical to run Felipe light in an effort to capture pole in the hope that the race itself would then come to Ferrari as the weather improved on Sunday. No point in doing that with Kimi as he would have to conserve his engine on Sunday in an attempt to get the car home. This is what appeared to happen. In the race Massa set the fastest lap for the first time this season while Raikkonen was a whole seven tenths slower. This is what I have termed a “door-kicking” margin in an earlier posting. A gap of this magnitude can only be a car issue. As Kimi is actually faster in the races than Felipe, as well as faster in qualifying when you correct for fuel, the inescapable conclusion is that he had the suspect engine running on a very conservative ECU map. This is also borne out by his dramatic slowing towards the end of the race in Hungary. The team said something about rear suspension trouble, but the fact remains that he dropped away from Glock, who there was little chance of passing anyway at such a circuit, almost as soon as Massa’s engine let go. This was engine conservation with the next race in mind. Too no avail as things turned out in Spain.

Ferrari made an inspired decision in qualifying
My own belief is that there are no team orders in Ferrari at this stage of this season. Or at this point in any previous season in recent years for that matter. Eddie Irvine summed up the situation perfectly when he said that racing alongside Schumacher was like being hit over the head with a baseball bat every weekend. In spite of everything Eddie and his engineer could muster, Michael was either faster in qualifying, or started the race on a superior strategy and got past anyway. Intra-team rivalry reached such a pitch in later years that the technical staff were becoming frustrated by Rubens Barrichello’s insistence on running different set-ups from Michael’s. So focussed were Rubens and his engineer on getting some sort of advantage that would allow him to jump his team-mate that they took risks with set-up and strategy that often dropped him down the order in the races. The team lost valuable constructor’s points as a result.

Naturally, team orders must come into play in the end game of the championship. But this is still the development of the game and there continues to be everything to play for. The two Ferrari drivers and their respective engineers are still racing each other. Think of this as a tandem bicycle race with the driver at the handlebars and his race engineer sitting behind him. A good result needs a big effort from both men.

Look at the fuel corrected times for Q3. Felipe could expect to be way down the order if he took a standard fuel strategy. He would then be bottled up behind slower cars in a race that would be frustrating indeed. I think that Rob Smedley made an inspired move. He sent Felipe into Q3 on a very light fuel load indeed. In the cold weather of Saturday afternoon Felipe’s raw pace was slow. He was only sixth fastest in fuel-corrected terms. Smedley was gambling that the warmer track temperature that was expected on Sunday would slow the McLarens and help the Ferrari. He guessed that Felipe would have such good race pace compared with the McLarens that he would be able to pull enough time in even a very short first stint to maintain his lead after the McLaren stops. He was right, and it was the decision of a tactical genius.

The television coverage showed how shocked the McLaren people were when Massa stopped for fuel so early. They were probably convinced that Hamilton was going to be the first to stop and suddenly realised what a huge gamble Ferrari had taken the previous day. What a masterstroke for Felipe. But what a frustrating day for Raikkonen and his engineer when his engine failed to survive in spite of being nursed during the race.

Could this have been a factor in what appeared to be Kimi’s error in setting off before his indicator light was switched to green? Without this disaster he would have jumped Kovalainen at this stop. Massa’s race was also notable for the careless release of his car (in Felipe’s case the traditional ‘lollipop’ is used) straight into the path of Adrian Sutil. If Massa had not swerved to his right and let Adrian through the cars would certainly have tangled. So there may be inspirational stuff happening in the Ferrari team but there is also an ominous hint of a return to a Chinese fire-drill chaos. Already two engines have cracked under the pressure and we have to hope that the pit-wall and garage crews will not crumble as well. Will both drivers now enjoy engines free from the vulnerability of their previous ones? Spa is a race that Ferrari can be expected to win. If the weather is good. And there is still no end to the autumnal conditions as this dreadful European summer continues without end…..

Ciao

PS: The lamentable Steve Rider said that Smedley comes from Manchester. That will pain my reader who pointed out recently that he hails from Middlesborough (re-invented in modern times as 'Middlesbrough'). I don’t think Rob himself will be upset by the TV nonsense as he will still be on cloud nine at the moment. What a fine relationship he seems to have with Felipe. The wonderful TV shot of the pair of them inspecting the circuit on board a Scuderia motor-scooter is an echo of my tandem-bicycle analogy. And little Felipe didn’t even have duck his head to give Smedley a clear view of the track ahead.