The philosophical and practical difficulties facing Mercedes in its design concept shift

In my time as chief technical officer at Williams I had the chance to engage with Toto Wolff on a regular professional basis. I really respected his management ability: he knew when to use the carrot and when to use the stick.

At this time, while still a shareholder at Williams, he had relinquished any close role to avoid conflicts of interest with his new role at Mercedes. My relationship with him, therefore, was as a customer for the power unit. I perhaps saw more of the carrot than the stick, although there were times when the string that the carrot dangled from was visibly shortened when our team challenged his. My colleagues at Williams, however, were only too aware of his ability to wield the stick since Williams had been singularly unsuccessful in the years up to 2014, and Toto was forthright in his criticism of certain aspects of the operation.

Toto has been very vocal in his feelings about the Mercedes performance of 2022 and early 2023. He was always the first to praise the team during its immensely successful years from 2014 to 2021 and, therefore, had every right to question the direction the technical team took with the design for the new-era car for 2022.

Being the hands-on manager he is, one can imagine he’s been involved in some difficult and probably heated discussion in Brackley over the past year or so. He will undoubtably have been wanting to hear the outcomes of post-race technical analyses and will also have been briefed on the content of regular design reviews. He will have been demanding explanations as to why the team has lost the competitive upper hand and what actions are in place to rectify the situation.

Unfortunately, even with the sophisticated data tools available to the teams today, the results of competitor analysis aren’t always clear and, even when they are, don’t immediately lead to solutions to reverse an uncompetitive situation. The data can confirm a car is slow in a medium-speed corner or is lacking top speed, for example, but the solutions need to be determined through an optimisation process by the engineers. As an example, lack of top speed can be conceived as a car having too much drag but the route to better performance isn’t simply to reduce the drag, it’s to increase the aerodynamic efficiency – a much more complex task.

Drag reduction is easy – a smaller rear wing will achieve this, but it will also lead to a drop in downforce and a reduction in cornering speed. Performance optimisation is a multi-dimensional problem and not an easy one to understand, particularly if the data you have is sparse.

Concept vs execution

So for Toto to say the concept of the W14 is wrong and that the W13 left no development runway from which to launch improvements is a significant statement. The problem is that for a concept to be wrong, one must first decide what the concept actually is. Is it the concept itself that’s wrong or is it the path taken to implement that concept? How do you examine and quantify the multiple paths that may exist? One of these may lead to the breakthrough that’s needed, many of them may be just blind alleys.

Symonds has hands-on experience working with Wolff in good and bad times

Photo by: Keith Sutton

It’s very easy to pursue a direction because you become heavily invested in its success. You may feel responsible for a particular direction that has been taken – or you might firmly believe that, in spite of repeated failures, success will appear with the next design iteration. I worked with one very talented engineer at Renault whose ideas led to a breakthrough in our performance and yet who, at the same time, spent two years trying to make another innovative design work before he was told to throw it in the bin. It takes a strong mind to stick with something through adversity, but it probably takes a stronger mind to abandon a project which has become the focus of your life but isn’t yielding positive results.

When the 2022 regulations were released and teams started work to interpret those rules to best gain performance, a few things became obvious. It’s been the case for many years that part of the secret of aerodynamic performance has been to get clean, high-energy air to the rear corners of the car. This became more important in 2022 as the front wheel wakes were no longer pushed sideways but instead were contained closer to the car itself.

Does the problem lie within the realms of what may be termed concept, or is it firmly ensconced in the execution?

The need to get clean air to the rear was ever more important since this air serves three purposes. Firstly it acts on the rear brake duct winglets, producing downforce directly onto the wheel without going through the suspension; secondly, as it passes over the diffuser top surface it will entrain air from underneath, thereby increasing the underbody flow; and thirdly, and perhaps most importantly, it blasts the rear-wheel wake away from the car centre line. Any encroachment of the rear-wheel wake into the diffuser area is hugely detrimental to total downforce since it destroys the all-important flow of air under the car.

PLUS: Why Mercedes is fronting up to its F1 mistakes too much

Equally important to the capability to produce downforce is the ability to use it. The current cars, in total contrast to their flat-bottomed predecessors, reward a low rear ride height. To maintain a low ride height under high and low load conditions one needs stiff suspension. One also needs stable aerodynamics that are predictable and give the driver confidence the car isn’t going to misbehave in any area of its operating envelope. So when a car isn’t performing it may be that the concept is correct but the execution is flawed. Maybe the sweet spot of ride height is simply too close to the ground to be useable; maybe the bodywork stiffness is inadequate, leading to airflow detachment from the surface and instability.

So does the problem lie within the realms of what may be termed concept, or is it firmly ensconced in the execution?

Mercedes is set to give up on its current F1 car concept as it doesn't have the maximum performance potential to win world titles

Photo by: Glenn Dunbar / Motorsport Images

History’s near misses

Before we get into the detail of the Mercedes W13/W14 cars, let’s look back at some history to examine the provenance of some possibly flawed concepts. When the Ferrari 246 won the Italian Grand Prix in 1960, it continued a run that one would have to consider successful – but this was to be the last time a front-engined car won a grand prix. The debut of the Cooper T43 three years earlier meant that the Ferrari, although still competitive, was a flawed concept. Ferrari simply didn’t look at what the competition were doing and evaluate their designs with an open mind.

In the late 1960s there was a blind avenue of four-wheel drive in Formula 1 which many followed, including Cosworth. It’s an interesting example of a concept which may have been correct but an execution that failed for reasons not obvious at first sight. I recently spoke to Cosworth co-founder Mike Costin, now 93 years old but as bright as ever, to ask him if he felt the concept was wrong. He felt it wasn’t.

“Traction had always been a problem, even with only 400bhp, since we had no aerodynamic downforce,” he told me. “The concept was to run 55% of the drive torque to the rear and 45% to the front and this would have worked.”

Now Mike wasn’t only a brilliant engineer, he was also no mean driver and it was he who first tested the car. “After five laps driving at Silverstone I was too exhausted to turn the steering – it was so heavy.” The Formula 1 cars of that era had no hydraulics and so power steering wasn’t really something that could easily be added. The only way to lighten the steering was to move the torque split backwards.

“When we got to 70% rear and 30% front the car was drivable but the advantage had gone,” Mike concludes. So the concept may have been correct but the necessary enablers for success were not there. As we progress to the era of ground effects, Jody Scheckter won the 1979 drivers’ championship in the ungainly Ferrari 312 T4. This was powered by a flat-12 engine, the architecture of which was completely unsuited to the requirements of a ground-effect car: its width encroached on the area where the underwings of other cars were placed. Nevertheless it won the championship and is a rare example of a flawed concept that was ultimately successful.

In my own experience, the Renault RS22 engine was similarly flawed in concept. The ultra-wide-angle V10 engine was conceived to lower the centre of gravity of the engine but the lateral spacing of the inlet trumpets meant that cylinders were unable to cross-feed, something that hadn’t been understood before the engine was built. The result was an extreme lack of torque arising from a conceptual decision which could not be changed without starting again.

Scheckter's 1979 triumph in the Ferrari 312 T4 was a rare example of the flawed concept proving successful

Great ideas, bad consequences

Sometimes the concept is good on paper but anticipated problems prove harder to cure than expected. Such a car was the Brabham BT55. Brabham had used the BMW four-cylinder engine for a while and felt a worthwhile improvement to the car could be made by canting the engine over at an angle rather than running it upright. Oil scavenge problems were anticipated but never really resolved despite multiple re-designs of the sump and scavenge pumps – another brave concept to bite the dust.

Even the best have their occasional off-years. Adrian Newey, in my opinion the best technical leader of recent years, has overseen a car so flawed it never raced: the McLaren MP4-18. This car from 2003 incorporated many ideas and was perhaps a step too far in concept. In his book How To Build A Car, Adrian states that the problem was “related to the shape of the chassis and the front of the sidepod overloading the vortex that forms off a delta wing just in front of the sidepod, causing the vortex to be unstable and burst in certain conditions. The problem could be alleviated by trimming the wing, but this lost downforce”.

Lewis Hamilton says that he isn’t at one with the car – is there a clue in this? A driver needs confidence and this confidence can be destroyed by an unstable car

With all due respect to Adrian I suspect this was one of a myriad of problems. The car featured a blown diffuser, something we were struggling with at Renault at the time. The only solution to stabilising this was drastic engine remapping – something Mercedes wasn’t doing, I believe. Mark Preston was the chief structural engineer in charge of the chassis. I asked him about the car and he told me: “Adrian wanted to use unidirectional carbon everywhere. In some places this was good but in others difficult. The front wing main plane would alter its angle of attack by five degrees at maximum speed due to a lack of stiffness and this made the car hard to balance.”

The problems weren’t confined to structure and aerodynamics. Mark went on: “The desire to lower the engine in the chassis and the extremely tight bodywork both conspired to make something that was theoretically good but practically bad.”

It was a car where the concept was perhaps too ambitious and maybe ahead of its time. Size-zero bodywork and low-mounted drivetrains are, however, now the norm.

Is the Mercedes concept the right one but the team is unable to execute it fully?

Photo by: Lionel Ng / Motorsport Images

Sidepods may not be the whole problem

Returning to the current Mercedes, if the concept of all cars is to get high energy air to the rear corners, it’s more in the execution that one might see differences – and where the architecture of the car may either enhance its performance or be detrimental. In early 2022 we saw two approaches to dealing with the sidepod aerodynamics.

Some chose to flow the air over the top of the sidepods and then encourage the air to dive downward toward the diffuser. The most extreme example of this was the Williams, which relied on a strong Coanda effect to keep air attached to the extreme curvature of the top of the pods. This was soon abandoned for a more gentle shape. The other approach was more akin to the pre-2022 cars with a deep undercut beneath the radiators to allow a low-positioned flow of air to the rear corner.

History shows the former to be more successful (Red Bull, Ferrari and Aston Martin) but this isn’t to say the latter approach is incorrect. Mercedes has undoubtably taken that concept to the extreme by means of an extremely clever cooling layout. While Red Bull and others have maintained an air-to-air intercooler to cool the engine air coming from the compressor, Mercedes, like Ferrari, has used a water-to-air intercooler. In the case of Mercedes this has led to an extremely tight design of the rear of the car, which is enabled by the most complex fuel-cell area and rear of chassis of any car I’ve ever seen.

PLUS: Why "grounded" Hamilton remains confident in Mercedes' F1 recovery

Lewis Hamilton says that he isn’t at one with the car – is there a clue in this? A driver needs confidence and this confidence can be destroyed by an unstable car. In turn this can be caused by aerodynamic instability or a feeling of movement. Maybe the complex chassis shape doesn’t provide the stiffness required. Only Mercedes knows the answer to that.

The limits of copying

Perhaps more importantly, what can Mercedes do about it? The budget cap assumes a regular development programme but limits a total redesign – if indeed that’s what is needed. Copying someone else’s concept is always difficult. One designs to a maximum of performance but one never knows if this is a local maximum. Moving away from where development is leading you always starts with a negative result until the nuances are understood and changes get positive.

Aston Martin found success in following a Red Bull design concept last year, dubbed the 'green Red Bull' but would that process lead Mercedes to its overall goal?

Photo by: Zak Mauger / Motorsport Images

In my time at Williams, the first car I was involved with was the FW36 of 2014. This was extremely successful, largely because we pitched many of our design objectives in just the right area for the then-new turbo-hybrid era. After some bad luck in the early season we regularly took the fight to the dominant Mercedes team, claiming the front row in Austria, leading before the rain in Silverstone and chasing Lewis hard to a second and third place in Abu Dhabi.

The car had been generally reliable and carried around 17kg of ballast. For the 2015 car, I wanted to turn that ballast into performance and there was one obvious place to do it. At the time every other car had their lower rear wishbone aligned with the driveshaft, seeking an aerodynamic benefit. The FW36 had the wishbone positioned lower which meant loads in the wishbone were reduced relative to the more common design. The lower loads meant a lighter structure and so this was where we turned our attention. Lifting the wishbone in line with the driveshaft led to a significant weight increase and, initially, a drop in aerodynamic performance. We had optimised the car around a local maximum of performance and when we moved away from that, other surfaces needed to move in harmony with the new airflow. 

The initial drop in performance was high and it took some time to re-optimise. Although ultimately we got the performance we were seeking, it was a classic example of how one cannot just copy the detail of another car without fully understanding and working on the concept that surrounds it. 

Did the win in Interlagos persuade the leaders there was nothing intrinsically wrong with the car – and, therefore, was this triumph actually the foundation of further grief?

If Mercedes’ problem is aerodynamic it needs to rapidly evaluate the concept of the successful cars and I’m sure it’s doing that. If the problem is more systemic, such as a structural problem, the road to success is much longer.

I think what makes the answer to the problem more difficult is that towards the end of 2022 it appeared Mercedes was on top of the car and competitiveness was returning. Was the end of last season, and particularly the result in Brazil, a false dawn? Did the win in Interlagos persuade the leaders there was nothing intrinsically wrong with the car – and, therefore, was this triumph actually the foundation of further grief?

Results in 2023 so far suggest this was the case and, if the budget cap boat has sailed, this will be a long season for Mercedes.

If the concept flaws run deep it could be a painful season again for Mercedes

Photo by: Glenn Dunbar / Motorsport Images