Why Rosberg is wrong about the 2019 Ferrari

What did you make of Nico Rosberg's recent comments that the Ferrari is not sat in the correct aerodynamic window?
@Professor_IPC, via Twitter

I can understand his comments but that doesn't mean I agree. Barcelona has always been a circuit that requires a bit of everything from a car. It has a reasonably long straight, braking entry into Turn 1, then it's traction and understeer limited out of Turn 2, which is followed by the long, demanding high-speed Turn 3.

Turn 4 is a 180-degree medium-speed corner that requires good balance both on entry and exit, Turn 5 is a slow, downhill-braking, about 170-degree corner that has a lot of camber change so it's easy to lock the front wheel on entry, and Turn 6 doesn't really exist. Turn 7 has a compression in the middle and Turn 8 is a kerb-clouting corner as you try to get the power down heading up to Turn 9, Campsa, which is a blind, fast right-hander.

Then it's downhill into the Turn 10 hairpin, the kink of Turn 11 and the long Turn 12 where it's very easy to enter too fast. Finally, the downhill right-hander at Turn 13 leads to the kerb-hopping chicane before the last turn onto the straight.

So, it has a bit of everything, and if you are fast there you should be in pretty good shape at most tracks. That's why the teams go there to test.

Every car will have a sweet spot when the engine power, drag and downforce levels all work together at their best. Racing Point is a very good example of that. Historically, its performance around Spa was always right up there - even going back to the Jordan days. And that's because the team focused a lot of the aero testing around that sort of level of downforce-to-drag and the other circuits just fell out each side of that requirement.

Basically, you have a graph starting at your lowest level of downforce and drag for Monza and going to your highest for Monaco. All the other circuits fit in between those two somewhere, but it isn't linear - so you may have an area that is a little more efficient and that probably comes from the level you do most of your research around.

I think Ferrari have the downforce and efficiency, it just hasn't got the best from its car at all the circuits so far this year. And when it did, reliability gremlins and driver over-enthusiasm crept in.

Reading that Haas had problems getting heat in the tyres in China, I was wondering how a team can sort that issue?
Thomas Martensson, via email

The tyres are a problem for everyone and to get them into their working window is not easy. But there is a lot more to it than just temperature.

Pirelli requires that the tyres are heated to a maximum of 110C front and 80C rear and they also stipulate a minimum tyre pressure. This varies for each circuit, but let's say a minimum of 22psi front and 20psi rear is the average.

These pressures - especially the rears - are, as far as Pirelli is concerned, required to reduce tyre sidewall deflection that could cause structural damage. But it is not optimum for grip, especially the rears. You can live with it at the front, as it offers increased stability - a couple of psi lower wouldn't do any harm.

When you consider that a Formula 1 car has a weight distribution and aero distribution of around 40% front, it means that the rears are responsible for around 60% of the cornering load. The fronts have to also turn the car but the rears have to cope with all the acceleration, so the demand on the rears is very high.

When you heat the tyres in the blankets, obviously the pressure goes up so the teams will heat them to the maximum allowable temperature to achieve the minimum pressure when the blankets are taken off. That is when it all gets a bit silly - you will see cars sitting in the pits with the rear blankets removed just to let the pressures drop, and then on the out laps they all creep around. Heavy braking will bring up the front temperatures and if for any reason the rears get too cool, a quick dab on the throttle instigating a bit of wheelspin will bring them up to temperature - at least on the surface.

If the tyres are left too long in the blankets, this tends to heat treat the tyre compound. This, in turn, will harden the compound and offer the driver less grip. Also, consider that the tyres used in qualifying are brand new and have not been loaded up. When brand new, they are like a new pair of shoes and wearing them for a little while frees them up. Tyres are very similar, so if you checked the vertical or lateral stiffness between a new and a used tyre, the used tyre would be more compliant.

If a tyre is more compliant, it allows the driver to feel the performance limit more easily without going over the top of what's called the slip angle. This is when the tyre slides and offers no grip, just like when a driver locks up under braking.

The best situation is to switch on the blankets, giving the tyres exactly the correct amount of time in them so they are just coming up to their maximum allowable temperatures. Then, take them straight out of the blankets and get them on the car, let the rears cool down something like 20C and head out on the track doing a reasonably fast outlap braking hard, but not using the throttle too aggressively.

This will keep the rear temperatures down, keep the front temperatures up and, more importantly, the extra load from cornering reasonably fast will free up the tyre carcass and offer the driver a less peaky performance tyre on their actual fast lap.

Why, if a team has so much money to spend, can they still not always compete at the front? Is money not the key in F1?
levidg15, via Instagram

Money buys you the key to being in F1, but then you have to spend it wisely and on the things that allow the designers and engineers to come up with the best package.

There is no magic bullet or god-given right to be successful in F1. It is all about putting together an organisation that functions at every level. There are many layers of management structure in place to achieve that, and the days of one individual steering that ship have long gone.

Leading the design team like James Allison at Mercedes or Adrian Newey at Red Bull do is still critical to give the team direction, but behind them is a huge structure of very competent people that all need to be managed. More importantly, they need to pull in the same direction. A few people prioritising their own egos can very quickly undermine a team's overall efficiency.

So, to sum up your question, money is critically important if you are going to be successful, but on its own does not guarantee success.

This is a follow up to my last Ask Gary question in regards to wheel wake and diffusers. You said that designing a car isn't about applying what's in a textbook. But the textbooks contain the physics of fluid flow and the equation I referenced is used by all steady-state CFD solvers. So how is designing an F1 car not about applying what's in the text books?
f1tech_blog

I have yet to see a successful F1 car that has been solely designed using CFD. It will happen, and probably in the not-too-distant future, but for now I believe you still have to think outside of the box - and that even applies to windtunnels.

CFD is an excellent tool when it comes to optimising a closed duct, like a brake duct or underbody cooling ducts, or detailing a turning vane or wing tips to increase or dissipate a vortex that is, in reality, already there. Also, it's effective to help with visualising the changes in airflow of certain components to allow the engineers to optimise or come up with the next step for that component. But coming up with a complete car design based solely on CFD is still a long way off.

When you see teams with their massive and very expensive aero rakes fitted to their cars, they are there to gather data - helping them correlate with what they are getting from their CFD and windtunnel research.

I have yet to know of a car that actually works on track as was predicted from CFD or windtunnel data. The track is a much harsher environment than both of these because the car is never in a steady state - it is always on the move. Something is always happening - from riding high or low frequency bumps to slowing down or acceleration, applying or reducing steering lock, to clouting the kerbs etc.

All that combines with the rotating tyres, which on one side of them have the low pressure of the underneath of the car and on the other they have reasonable atmospheric pressure. Unfortunately, you want the tyre-displaced airflow to go towards the latter, otherwise you will lose a high percentage of the underbody downforce.

We have seen, with all teams, developments arriving at a test or race meeting and then disappearing again. All of these will have been signed off by the aerodynamicists as a positive step forward, but they get track tested and end up in the skip at the back of the factory.

It is no fault of the aerodynamists, it is just that components on an F1 car have to work in such a harsh environment that modelling or simulating it is still nigh on impossible.

Perhaps Mercedes is using the textbook and Williams hasn't received its copy yet - or perhaps it's the other way around? Whatever it is that makes one team better than another, I'm pretty sure that thinking outside of the box and 'interpreting' the results from experience still counts for more than the answer at the bottom of a complex spreadsheet.

Will Williams, McLaren, Toleman (Renault) and Lotus still have Senna's telemetry data from when he drove for them?
widlarzjakub, via Instagram

I'm pretty sure that Williams and McLaren might just have something in their archives somewhere. Toleman - Benetton, Renault and Lotus - has been through many owners since those days so I would imagine it might have been lost in the changes.

Data between when Senna drove and now is vastly different and how it is managed is a different dimension. Currently, data is everything and the engineers and design are driven by data. Every team gathers loads of it and within analysis tools they will have set some working parameters.

Taking a simple one like water temperature, a team will have set a working window of, let's say for simplicity, 115-125C. If it is within that then the team is happy, if it moves higher or lower, then that channel might just flash up red - drawing the engineers attention to it. Then, the car engineer will instruct the driver to perhaps turn down the engine or drop back from traffic.

The same will apply to car performance parameters such as ride height, centre of pressure brake balance, tyre temperatures, tyre pressures, brake temperatures, etc. If they all stay within what the team has set as the working window, then everything is fine. If not, it will alert the data engineers and they will then look into it a little more deeply.

There are probably 200 channels of datalogging on each car, so far too much for anyone to monitor while the car is running unless something draws their attention to it. Sometimes it can be too late to detect and analyse before a failure occurs, but as far as car performance is concerned something like a major data reduction should be considered. I think even Ross Brawn agrees that a car's performance is too driven by the garage and from back at base.

My suggestion, as I put in a column a few weeks ago, would be that each team would be allowed something like 10-15 channels to monitor reliability and safety stuff - engine and gearbox temperatures and pressures, tyre pressures, brake pressures to name but a few. The rest could be logged but not available to the team until the chequered flag falls post-race. They can then analyse the data and see where they went wrong, but at least it would increase the driver and engineer's technical relationship.

In the evening after practice, the engineers will go through mountains of data in detail trying to optimise that working window and refine it to suit what the driver is saying about the car balance. Then, changes will be instigated based on the data.

This might mean the data is showing that the front ride height is a little too high - 0.5mm on an F1 cars front ride height is critical, but that will give more front grip at all speeds and the driver was complaining about oversteer in the fast corners. So perhaps they will drop the front wing angle a little bit, which can be the overpowering feature on the car's balance as the speed increases.

I remember Patrick Head once telling me that Nigel Mansell was complaining that he wanted to go faster around Copse corner but he didn't feel he had the grip. Patrick told him about the downforce increasing with speed and that the faster he goes the more downforce he would have. Nigel went out and simply went faster, no problem.

What is the biggest change you would make to the front half of the cars? For me it would be no bargeboards and the floor to be in line with the body, more like these were in the past, and to go back to a similar front wing.
Rob Lee, via Twitter

It really depends on what the objective is. Is it to lose downforce, make the cars less critical to turbulence or just for the sake of change?

I am not a great fan of the last point, but the other two wouldn't do any harm. But we must make sure we don't end up with an engine formula. The more that is taken away from the chassis side of the equation, the more the engine will define success.

If we think that the car's performance is equal between the chassis and engine, that means each is responsible for 50% of its performance. Then we look at the difference between the performance of Mercedes and customer teams Racing Point and Williams, and Ferrari and its customers, so we can try to come up with a percentage number for chassis performance that adds up to the overall lack of pace.

It's not just about the overriding number, it is more of an all-around loss, but let's just call it grip. Relative to their engine supplier Haas are probably the closest at around 1.5% to Ferrari, Alfa Romeo and Racing Point are around 2.0% to their engine suppliers and Williams 3.2% to Mercedes. So, if we multiply those numbers by 10 we get Haas lacking about 15% grip, Alfa and Racing Point 20% and Williams 32%.

These are just rounded up figures, but I think they show roughly the magnitude of chassis grip that would need to be lost relative to the lap time loss. If we were to say a 30% reduction was the objective then it would leave the engine responsible for 80% of the car's performance. That's all OK as the engine is not affected by turbulence. But I believe that, as in the old days when teams went out and bought a Ford DFV, F1 should be primarily a chassis formula and keeping similar downforce levels to the current cars but producing much more of it from the underfloor is the correct way to go.

So, I would start with a very powerful underfloor with a window of parameters that each team could exploit to get their own individual aerodynamic characteristics. From that, you can then define the front wing and bargeboard assemblies to make sure the underfloor is the dominant downforce-producing device.

Are Mercedes boss Toto Wolff and the drivers serious when they say they are worried about the pace of the Ferraris, or are they just trolling/playing mindgames with Ferrari?
Ricardo Lopez, via Twitter

There is probably a bit of both, but I think they should be worried. As we saw in Bahrain, Ferrari is not far away and they just need to get it right on more occasions and understand why the performance drops off more regularly than Mercedes. Put that right and they could be a real threat.

Also, Mercedes is covering its arse! If they were to jump up and down shouting about how good they were and it all went wrong later in the season, they would have to eat some humble pie. Remember, the media is cut-throat and are always looking for a few column inches, so say the wrong thing and many of them will be out there waiting to jump when you fall flat on your face.

Ferrari definitely had the upper hand at the start of pre-season testing in Barcelona but then, with a few tweaks, Mercedes caught them up for the second test. In Australia, Ferrari was well off the pace, in Bahrain it set the pace, in China Mercedes had the upper hand and now the teams are off to Baku. Who will be on top there? I don't think anyone has any idea, but I don't think either will be too far ahead.

Barcelona is the one I want to see, as this will show who has really progressed since pre-season testing and who has lost their way.

What can you tell us about Mauricio Gugelmin and his reputation as an outstanding car developer? Is it paramount to have a driver with good development skills in the team?
@DeWeberis, via Twitter

Mauricio was a good driver and also good to work with at Jordan. He had a very professional attitude and it was a pity he never really got a chance in a top car.

Times have changed dramatically since I worked with him in 1992. Then, we were probably collecting something like only 20 channels of data and at a fairly low rate. Most of these would also have been for system information, oil and water temperature, pressures etc. But a few would have been chassis performance related. So, the driver was, in reality, your connection to the track.

What they said was what was happening and you used the data to try to understand it. But in my opinion, these were the best days when the driver and the engineer could work together and come up with set up solutions from their experience.

Mauricio was good at this and you knew that if you could react to what he said he needed to go faster, he would go faster. So many drivers, even the experienced ones, would want something and when you gave it to them they would come back having done the same laptime saying that the car was easier to drive. I didn't want it to be easier, I wanted it to be quicker!

Data-driven car setups and engineering overtook the driver input about the time of active suspension and traction control. To optimise all this electronic control of the car's performance meant that you needed sensor input for the control system, so then it was available to the engineers.

When these electronic aids were outlawed, unfortunately the data was still left and now what we have is a formula that is data-engineered and the driver - who still needs to be fast - doesn't need to be experienced. For confirmation of that, just look at how the young drivers of today are performing.

You at Jordan and Harvey Postlethwaite at Tyrrell were the early adopters of the raised nose concept in F1. What led you to that aerodynamic solution?
@DeWeberis, via Twitter

We both went for fairly different interpretations of the same concept, but probably for similar reasons.

Our reasoning at Jordan was that our cars were going to have to get through a pre-qualifying session on Friday morning before the event actually started in 1991. It would be on a dirty track and most of them were circuits we had not run at, so if we weren't in the top four out of eight cars we would be packing our bags before the event would actually officially kick off.

To achieve that, we wanted a car that was reasonably easy to set up and that gave the drivers confidence. We wanted the rear of the car to be as stable as possible under braking and for the car to have a good balance in the fast corners.

We had a very long-chord front wing main plane and when it got near the ground it stalled, basically upsetting the airflow to the underfloor. So, by lifting the nose and reducing the chord length of the main plane as it went inboard, we were able to manage that stall better and it became much more progressive. That, combined with different wing flaps and inboard wing expressions, achieved what we wanted.

This combined with a vortex that was generated from the trailing edge of the long front wing endplates where they went inside up to the centre of the front tyre. This vortex sealed the underfloor sides, improving the diffuser performance especially at low speed or high ride heights.

Interestingly, to optimise this vortex we used different lengths of wool attached to a piece of wire that we fed through the windtunnel wall. The strength of the vortex was defined by what length of a piece of wool it could rotate. We also added a small undercut where the sidepod wall met the start of the coke bottle area and this created a stronger vortex.

From all of this we created a car concept that, under braking when the front wing was close to the ground and the outer sections stalled, meant the airflow to the underfloor was consistent and the centre of pressure went rearwards. Helped by the front wing endplate vortex and in fast corners under cornering forces with roll, the outer section of one side of the wing would stall but only enough to give the driver confidence in the rear of the car.

Some things never change. The wheel is still round and airflow still creates vortices and forces, they can be positive or negative but that's the designer's decision.

Do you have a question for Gary Anderson? Send it to askgary@autosport.com, use #askgaryF1 on Twitter or look out for our posts on Facebook and Instagram giving you the chance to have your question answered