How a new F1 design can go wrong

February is looming, the start of pre-season testing is just around the corner, and the Formula 1 teams are almost ready to reveal and then run their 'new' cars for the first time. But what exactly constitutes a new car?

The idea sounds simple enough, it's just the new machine for the coming season. All 10 teams will have one on track when testing begins at Barcelona on February 18, but as always in F1 things aren't that simple.

The regulations are now so closely defined that the opportunities to do something visually and 'engineeringly' different - the ground-effect Lotus, the Tyrrell six-wheeler or the Brabham fan car spring to mind from the '70s - are very limited. Back then, and for much of the history of grand prix racing, teams had enough latitude to stick their necks out and do something different, and only later would it be deemed legal, illegal or somewhere in between.

Today, F1 cars are all about optimising every small detail and piecing together the thousands of components to maximise performance. So it's no wonder that creative minds like Adrian Newey's get a bit bored with the predictability of it all.

In the old days, there was much more carry-over from season to season. On many occasions we would actually start the new season with the previous year's car, then introduce the new one at the first European race - or even later.

Going back to my early days in F1, the Lotus 72 (1970-75) and the McLaren M23 (1973-78) had long lives in the world championship. But that all disappeared in the late '80s and now the pattern is that teams introduce a new car at the start of the season and a major upgrade, if not another 'new' car, mid-season.

Last week we heard from Renault's new technical chief Marcin Budkowski, who explained that the team's 2019 offering was all new except for the power steering.

My worry is the unnecessary risks Renault might have taken to achieve its 'all-new' car

"It is always challenging to get a car with thousands and thousands of parts, and the only thing we carried over from last year's car is the power steering," he said.

"That gives you an idea that all the rest [of the car] is changing, and it is changing because we are trying to squeeze more performance from every area. It is a big effort to do everything from scratch, plus the [new] regulations, but it is a performance goal.

"The later you leave it, the more performance you have, but the more stressful it is to get it out on time."

My worry when I read those comments is the unnecessary risks Renault might have taken to achieve this.

If the regulations stay stable. the step over the winter should always be about more downforce, better aerodynamic efficiency, stiffer structure and cutting back the weight.

The aerodynamic changes to the front wing, especially the endplates, along with the bargeboards and the new rear wing location (including the larger DRS opening), mean there should be a focus on just getting back to where you were before the changes.

Until the cars appear and start testing, no one will really know how well they have done with their recovery plan. Some will have done better than others, because no matter how tight the regulations are there's always scope for getting it right or wrong.

But even with the extra demands and very expensive regulation changes, you can't neglect the other potential but more basic improvements.

We've often heard over the past few seasons how difficult it is to get the cars down to the minimum weight limit, and that the drivers have had to go on some very serious diets. Weight is a major performance penalty, and on average 10kg costs you 0.35 seconds per lap (it varies according to the circuit).

So if, for any reason, your car is that much over the minimum weight limit, that's the penalty you are paying. And that's even before you consider the actual car performance because one that's overweight will cost you in terms of tyre degradation because the rubber is doing more work.

For 2019, the minimum weight of the car and driver increases to 743kg. As is often said, about 4000 drawings make up a car so attention to detail in every component is critical, otherwise weight inefficiencies can add up and you end up with a car that is too heavy.

If every component can be analysed in minute detail and you can save 1% on every one, that equates to a saving of 7.4kg over the complete car. If 2% can be saved then you're really on your way. If you can make one component do two jobs then that reduces the component count and the risk of an under-designed component failing - as well as potentially offering packaging gains.

To achieve this sort of weight reduction isn't easy. It means you have to either interrogate the data you have and make sure you really understand the load cases in minute detail, or you reduce your safety margin, which reduces the life of each component. But this increases your risk of failure if, for some reason, you exceed the design loads.

You can reduce the weight of the existing component by increasing the detail of the machining, or you could achieve it by coming up with a slightly different geometry. But either way it is a lot of very detailed design work and structural analysis before you can sign it off.

Just making sure the bolt lengths are optimised - for instance with a 1mm reduction in the length of every bolt holding the car together - or a small machine detail on the ends can add up to well in excess of a kilogram.

As far as stiffness is concerned, basically the stiffer the better, but not at the expense of adding weight. Normally this can be achieved with optimising the geometry and/or load paths of every component.

With efficient load paths, it's all about ensuring you feed the forces from the suspension into a stiff structure instead of having to create a new one just because the aerodynamicist doesn't want a wishbone to attach to the chassis at a certain point or they want a 'bent' wishbone leg.

For 1998 we built an all new car at Jordan but it had some aerodynamic flaws. Basically, it cost me my job

Sometimes this is unavoidable, but it will lead to a heavier component, so something else will have to be lighter to compensate. As a visual example, if for some reason someone wanted to have a bent suspension pushrod it would be fairly difficult to do and would require a significant weight increase, so the positives always need to outweigh the negatives.

On many occasions, suspension components are placed to optimise the aerodynamics and everything is a compromise. But if you need to add weight to achieve just a small amount of extra downforce it needs to be justified against the weight and tyre degradation penalty.

So the big question I posed at the beginning of this column was when can a new car be considered a new car? Well, in reality it is always new - but not necessarily a new concept. If you just optimise everything then by definition everything will be new and you can make serious strides forward.

Taking Haas as an example, when it showed up for pre-season testing last year we said its developments looked a bit tame. But the season showed that Haas had made very good progress and had, on average, the fourth fastest car, even though Renault beat it in the championship.

What you need to do is recognise the areas that are compromising your performance and do your best to reduce that compromise. If you go blindly into doing everything differently, you will normally just move those compromise areas somewhere else.

On the other side of the fence were Williams and McLaren. They basically changed everything and for pre-season testing last year we said that their cars now had all the right pieces in all the right places. But I questioned whether they had got them all working together because this would determine if the cars were successful or not.

I think the season proved they had what could be called new cars, but they weren't a step forwards.

Stability is the thing that enables you to make the correct decisions and dedicate the time to the smaller details. That is stability in personnel, regulations and in engine supplier. All the teams will be optimising their power unit installations but this year Red Bull has the extra work of changing from Renault to Honda engines. Everyone has to adapt to the new aerodynamic regulations, though.

That stability wasn't always so common and at Jordan we were faced with changing engine suppliers on more than one occasion. So for a small team, especially in the early days, we focused on just getting the car on its wheels. A lot of the nice stuff, optimising the detail of all the small parts, we had to forgo.

During those changes there were two examples that caused problems but for different reasons. We were a bit short of budget so for the Hart-engined Jordan 193 we kept the chassis from its Yamaha-powered predecessor and basically wasted a year's potential development. By not keeping our programme up to date we were a year behind and always playing catch up.

And for 1998 we built an all-new car but it had some aerodynamic flaws. Basically, it cost me my job at Jordan but by the time I had left we had the fixes in place and it went on to win at Spa. At the end of the season, it was a very competitive car.

Getting the balance of how much to carry over and how much to change is one of the biggest decisions that a technical director, or technical manager, will have to make. It is all about 'return' versus 'risk'.

It's not long now until we see who has made what decisions, and more importantly who has made the correct decisions. The one thing that you can't change is time, and the stopwatch never lies - no matter how new your car is.