(5 February 2004)
In my last column I briefly mentioned the use of telemetry traces, and suggested that they were not particularly useful for race coaching. I think it is worth explaining why I feel this to be the case.
Even a newcomer to Grand Prix racing will hear or read about telemetry very quickly. Properly speaking, it is the system whereby information about the way a car is running can be transmitted "live" to the engineers in the pit lane. The actual gathering of the information itself is called data acquisition, and it relies on a large number of sensors fitted throughout the car.
Data acquisition has long since become common even in club-level motorsport (though in this case the information is not transmitted anywhere but is stored for later download to a laptop computer). Even with the more basic systems, an amateur racer can now get access to engine speed, road speed, throttle opening, g forces, temperatures, lap times and sector times, all of which can then be perused at leisure and used to find ways of improving the car's set-up.
Professional teams in F1, Formula 3000, Indycar, drag racing and so on have much more sophisticated data acquisition which gives the engineers even more to think about.
It is worth emphasising at this point that - in the words of one engineer - data acquisition systems operate like "a very large notebook". They allow teams to gather far more information than they could otherwise, but they do not explain how to improve lap times. That job is still in the hands of whoever is setting up the car.
Bearing this in mind, we come to a particularly common use of data acquisition. The fastest race or qualifying lap of each of two drivers in different cars can be superimposed on the same graph to see how one achieved a better time than the other. Realistically this is only likely to happen within a team, since rival teams are hardly going to share information, and indeed the method is generally used to find a way of improving the slower driver's performance.
Data acquisition can show where one driver is losing time. What it cannot do is explain why this is happening. If the driver and his engineer are looking at a particular corner, what are they to read into the data traces? Let's say that the quicker driver gains time by applying power earlier; the slower driver goes back out and applies power at the same time. But if his line through that corner is wrong, and the car was already on the limit at that point on the circuit, simply applying more power will more than likely make the car fly off the circuit.
If he is indeed on the wrong line through this corner, the reason may very well be that he was also on the wrong line through the last one and has therefore got his approach wrong. But how can he tell? Not by reading the graphs - data acquisition systems are highly accurate in many respects, but not when it comes to tiny but important differences in racing lines.
They can certainly record differences in braking distances, but simply knowing that one driver is braking 15 metres later than his team-mate may not be helpful. It may be down to natural ability; if the driver who brakes earlier tries to exceed his own personal limit it may cause all sorts of problems.
With all the information gained from the data acquisition system at hand, a driver can of course consult his race engineer, but the engineer can't advise him on race driving technique. There is no reason why it should be otherwise, since we each have our own specialities. If I were asked to engineer a race car I would look a complete idiot.
The engineer can, of course, change the set-up of the car, but while that may improve the situation at one corner it may screw up the handling on all the others. And we often read, particularly when it comes to Formula 1, that drivers are looking for a compromise, an overall balance over a complete lap.
The real problem is that data acquisition, as useful as it undoubtedly is, does not give enough information to improve a driver's technique. The only definitive answer is to establish a link between the driver, the engineer and the designer - someone who can pinpoint whether handling problems are down to the car or to the driver, and thereby save a considerable amount of money in testing and development.