home icon / Technical / Tech F1i: Mario Isola on the design of F1's 2017 tyres

Tech F1i: Mario Isola on the design of F1's 2017 tyres

The upcoming Formula 1 campaign will see the introduction of new tyres from Pirelli as part of 2017’s revamped rules. Wider Featuring a wider thread while offering lower degradation, the new rubber should not only boost performance but also see make drivers adjust their style at the wheel. We talked with Pirelli racing manager Mario Isola, to learn know more about the new philosophy followed by the Italian brand company.

Pirelli has been tasked with the sizeable challenge of coming up with different tyres for a new breed of F1 machinery that only existed on paper at first. Is it the first time that you had to produce a tyre for a car that didn’t exist yet?

Yes – for Formula 1 at least. When I started working for Pirelli, we were designing a new tyre for the Maserati MC12, which didn’t exist. We started from a prototype, and we developed the tyre together with the car.

In 2005 we went on to win the title, not just because the car was fast but also because it worked so well together with the tyres. In this project, there was no 2017 car.

Therefore, we had to work with mule cars, which was useful because we could test the new size [Dimensions have increased from 245mm to 305mm at the front and from 325mm to 405mm at the rear. Although the 2017-spec tyres are around 25% wider, their diameter remains almost the same (with a small uptick from 660mm to 670mm) while the rim’s diameter does not change.].

But the mule cars were not really representative, because the actual 2017 cars are much faster. It’s the first time I can remember Formula 1 introducing regulations to increase performance rather than restrict it.

This is why we struggled with the mule cars: if you are restricting performance you can just take a recent car, limit the performance and you get good data. But if you increase the performance it is not easy for the teams to make a big step up.

How did you combine these runs with team-supplied simulation data on downforce and loads?

The flow with the teams was good. They provided simulation on the expected performance of the car, while we developed a model of the tyre. They incorporate our virtual tyre in their car model, and came back with an updated model.

Their findings were sent to the FIA, which then fed these back to us so we could refine the design of our tyres and computer models. Every two months they gave us new data, anonymously and filtered by the FIA, and we updated our models accordingly, three or four times. It’s a loop: the more times we go around that loop, the closer we are to reality.

At the end of the process, all the data were converging to something quite reliable, after the checks we made during the tests in Barcelona. But as we expect a huge rate of development, we asked the teams to provide data also for the likely performance level at the end of the year.

When it came to working out the required integrity of the tyre we looked at the data for the end of 2017 and added a margin. The range for the end of year simulations is a bit wider than for the figures regarding the start of the season (the spread is around 20%), but at least we have a direction to follow.

As advanced as F1’s simulation technology has become, this sort of testing remains in the realms of a virtual world where not all factors can be reproduced faithfully…

Yes, if tyre structure can be designed at the factory and tested on the dyno, the chemical composition of the rubber can only be assessed in live conditions. We can say we have two steps in the development of a tyre.

First, we use our indoor machines to test the construction (high speed integrity, etc.), using the loads that teams expect at the end of year. And second, we need to better understand the performance of the compounds, for example degradation, overheating of the surface, and so on, because when you stress the tyre in a different way, you get different results.

We have to assess it on track, this is the second step. There are predictive models to simulate the thermal degradations, but they are not completed yet. This is an area where we are working a lot.