How F1 teams prepare for winter testing

With just a few days to go before Formula 1 pre-season testing starts in Barcelona, the teams will be working frantically through their final preparations.

Having been an F1 technical director overseeing this process many times, I thought it would be worth answering some of the key questions about this crucial and high-pressure part of the season.

At what point in the car-design process do you first have something physical in place?

To get the best possible package for the start of testing, you really have to start with the date the car has to go into the truck to head to the first run. Then, you can work back from there to produce an hour-by-hour schedule for the car-build process.

Normally, the car build would take around a week, but with 24-hour shift work this can be reduced dramatically.

From that, you will get the date by which various components must be ready for assembly. Then, the component manufacturing time will dictate when the design drawings need to be released by the drawing office.

Obviously, the long lead-time items such as the monocoque and the gearbox housing will have to be released first. These items can take about 12 weeks to produce. So if the car build is going to be during the first week of February, then these drawings will need to be released before the end of October.

Items such as rear-view mirrors will not be required until the last minute, so to have as efficient a car build as possible, a very detailed manufacturing schedule will give maximum time for research and development.

Once a drawing is released, it means that component is committed to. Given the speed of development in F1, by the time it is manufactured it will be out of date, so reducing the manufacturing time can make a major contribution to the car's performance.

Crash testing can always throw up surprises. Detailed analysis will reduce those risks, but any analysis is only as good as the data you use for it.

To get the right stiffness/strength/weight ratio for various components, you always have to push the limits to the maximum. That's not easy, but it's the old story of 'no pain, no gain'.

Again, these all need careful planning and must fit in with the manufacturing schedule. One slip up on a major chassis crash test can cost you a month as you replace that chassis. So most teams will actually produce two chassis moulds and two chassis more or less in tandem, one for crash testing and one for the car build.

It is not only the crash tests that can set you back because there are many component strength, deflection and stiffness tests. All of the suspension, front and rear wings, underfloor and floor tray will go through rig testing, and should any of them fail, they will need to be redesigned and remanufactured. That can very quickly eat up any time margin you have built in.

Why is the first fire-up of the car so vital?

Fire-up is an important day. It means the majority of the systems, oil, water, fuel, hydraulic - and most importantly these days electronics - can be signed off. And if that's not possible, any problems that have cropped up can be rectified.

It is a milestone achieved and if it can happen without any problems, everyone has a bit of a smile on their faces.

There are many problems that can crop up, but the one I remember best was when Jordan changed from a Hart engine at the end of 1994 to a Peugeot engine for 1995. We fired up the car but we couldn't get any hydraulic pressure.

The hydraulic pump was mounted on the end of one of the camshafts, and on the test rig the system worked perfectly. On the car, with the hydraulic rig connected, the system worked perfectly but when the engine was started there was no hydraulic pressure.

After a serious amount of head scratching, we discovered to our amazement that the camshafts on the Peugeot engine rotated the opposite direction to the camshafts on the Hart engine. How stupid did I feel?

But it wasn't really a question I felt needed to be asked! With a little bit of lateral thinking and replumbing we were up and running.

Does the modern-day short, sharp pre-season test schedule make a big difference to the task of getting the car built and working, or was it always like this?

Having a compressed test schedule really doesn't hurt the initial car-build process. What it does do is reduce the amount of time you have for updates or developments between the start and the finish of testing. This will have an effect on the final specification of the cars for the first race.

If you have any major problems that go unrecognised until the car starts running, you will probably be taking those problems with you to Australia.

What was the toughest car build you were involved with?

In the past, tough car builds were the order of the day. If you go back to when I started in F1 in the 1970s, the car build was actually the car build - or to put it more plainly, the car build was the car manufacturing. Basically, the two mechanics plus a couple of other people would make and build the car.

Things have changed dramatically over the past forty years. Now, with the levels of staffing these teams have it is all about sub-assemblies. The upright assembly, the gearbox assembly and the steering assembly - to name but a few - will come out of different departments having gone through various independent rig tests before they are bolted to the chassis.

One of the most frustrating car builds I can remember was when we changed from the Peugeot engine at the end of 1997 to the Mugen-Honda engine for 1998. We got the car built with no real problems, but the electronics package that had been created for Honda by an outside contractor to control the engine and the chassis just wasn't up to the job.

We attempted quite a few shakedown tests at Silverstone, but to no avail. The system just seemed to have a mind of its own. One night, in frustration, I made a quick phone call to TAG Electronics - who had made the Peugeot system - and in one week they had a system packaged to suit our requirements and we were up and running.

When the car does hit the track, what are the objectives of testing?

Testing is all about understanding the car and its inherent positives and negatives. Every car has its own DNA, that is something that you can never change, but it is very important to understand if you are going to get the best from what you have.

Testing used to be about heading blind into running the car, and then you would go through various different setups to identify the car's strong or weak points. Today, the majority of this will be done on the teams' simulation rigs.

You could say that nowadays there is actually a huge amount more testing than there used to be because the teams will run their simulation rigs 24/7. Doing this allows them to arrive at the first test with a better understanding of the car's setup requirements.

But any simulation rig is only as good as the data you are using to drive it. If there is something inherently wrong with the data, then it won't show up in the car's performance until it hits the track.

How do you go about scheduling a first four-day test?

The first test will all be about making sure the car is producing what your research says it should do.

The basic mechanical setup will have been fairly well optimised from the simulation rigs. It will need a little bit of tinkering with, but this will not be a priority for the first week of testing.

We hear a lot about aero testing, and this is the most important part of a chassis' performance. Teams will fit various rigs for measuring aerodynamic flow and pressures - many of these rigs cost more than some of the F1 cars that I have designed in the past!

They will do constant-speed tests down the straights, preferably in two or three different directions (if the track configuration allows) on each lap to reduce any wind influence.

During these runs, they will also measure the load on the suspension components. All of this data will then be compared with the theoretical data to make sure the car is not suffering from some strange characteristics that the research data doesn't show.

As the week progresses, you will start to discover any potential aerodynamic problems. Time for rectifying such problems before the first race is short, so identifying these (and there will always be some) early is very important.

What are those first few laps like - how quickly do you get a feel for whether the car is good or bad?

Those first few laps with any new car are a bit like when your are present at the birth of a child! After all, it is your new baby and you actually know just about as quickly, as you do with your children, if everything is OK or not.

If the driver can do a run of about five timed laps, he will know how the car feels. You can't hold a car back, you have to let it have its head, and taking into account the fuel load you are running the driver will have a very good feeling of the balance and the potential laptime. Those two things never lie.

Just look back at when the Brawn was born for 2009. Immediately it was on the pace, to the extent that all the other teams were saying it must be underweight.

What are the areas of data that the engineering teams will be looking at closely?

Aerodynamics will always be the prime mover in chassis performance. Since the early days are about making sure the car is performing as predicted aerodynamically, you will be gathering data on that while also assessing reliability.

If reliability problems are cropping up which reduce track time, that's something you must eliminate.

Over a day's testing, you can normally recover from a few niggling problems. But that's not the case when you get to a race meeting where a 90-minute session starts when the light goes green at the pit exit and ends when it goes red. Ready or not, that's it.

And if things do go wrong, you have to react and adapt quickly. If they don't, you will lose track time and fall behind.

How draining is testing compared with race weekends, and how difficult is it to keep everyone focused?

Testing is about long, hard days. And the end of a day's running is, in reality, when the work really starts.

The teams will run a shift system for testing. Normally, the guys that run the cars during the day will hand over to another group of people and they will carry out the overnight maintenance.

There will be some personnel who will overlap, but in general over a four-day test a ten-hour day is enough for anyone. If you are asking personnel to do more than this, then the risk level of a mistake increases dramatically.

The current cars are so sophisticated that if you don't dot the 'i's and cross the 't's on everything, you can lose hours of running the next day. Why would a team spend all that money just to potentially throw it away by overworking the staff?

How about keeping the drivers focused?

Drivers want to drive racing cars, and as long as the car is good to drive it is normally difficult to get them out of the car. It is when things aren't quite what they should be that they tend to go off the beaten track and lose motivation.

This is actually when a reserve driver comes into play. If you just need someone to pound around trying various setups, then as long as they can be consistent a reserve driver is ideal.

A lot of teams have used a reserve driver to do race simulations in the past, but if you have a decent car then the race driver is the best, because he will learn a lot about how the car treats the tyres over long distances.

How do you go about running race simulations and what do you learn from them?

Race simulation is actually more important that one-lap testing. It is the little things that can let you down.

You need to simulate a complete race weekend, and that includes all the heat soaks that the car goes through while it is sitting on the grid or during a pitstop, since these are the things that can bite you. It also lets you better understand the running procedure and how long some things take to do.

Also, the mechanics get a better understanding of how to work together. Remember, a new car is new to everyone and it will respond to that little bit of tender loving care that is only possible when you know someone.

Even if you have done all this you can still get bitten. When I was with Stewart, we did all our preparation for the start of the season but still when we got to Australia, both our cars suffered a problem sitting on the grid.

To minimise the weight of the engine, it had carbon cam covers. These had a heat shield between them and the exhaust system, and during testing we didn't have any problems.

But sitting on the grid, the heat from the exhaust pipes overheated the cam covers and they started leaking oil onto the exhaust system and covered both cars in a plume of oil smoke.

Later we found out that a different supplier had made the cam covers for the race engines and had used a different, lower-temperature material.

Technical directors usually pretend not to be paying too much attention to what the competition is doing, but surely that's not the case?

You have to take it all in. If someone says that they are not, don't believe them.

Your competition is your competition. It doesn't matter what the laptime is, it only a matter of being faster and better than your competitors.

Every team will have a person, if not two, that will sit down with every laptime - actually, every sector time - that the other cars have done, and do their best to build a performance model of the other cars. This will be compared with their own car and used to try to identify where they are better or worse than the opposition.

Teams will also be measuring other cars' speeds at various locations around the track. They will then use this to try to determine acceleration and derive from that power-unit performance.

It is actually one of the most important things that comes out of testing. You normally know pretty quickly where you are as far as your car performance is concerned, but it is only from testing and racing that you know about the opposition.

Summing up, if someone tells you that they are not worried about other teams' performance compared with their own and that they are just getting on with their own programme, then they normally have a fairly dramatic problem.