Ask Gary: Why did ground effect get dropped?

If undercar ground effects are the solution to cars not being able to follow closely, why were they banned originally?
Bruce Merchant, via Twitter

Bruce, ground effect cars of the past were getting a bit out of control. The complete underbody was an inverted wing and with sliding skirts the cars were producing huge amounts of downforce for their day.

The problem was that any sort of track or kerb damage meant that the sliding skirt system would get stuck in position. This would lead to around a 50 per cent loss of downforce and usually a huge crash as the driver would know nothing about the damage until he reached the next corner.

As time went by the venturi section of the underfloor was moved further and further forward. This gave more overall downforce and, as can be seen from pictures from the era, the cars didn't really require a front wing, or could get by with a small trim.

The underfloor of a full ground effect car was probably responsible for around 80 per cent of the car's total downforce. The current cars still produce downforce from the underfloor but it would be no more than 20 per cent of the total.

The underfloor works as a venturi, with the other half of the tunnel being the ground. This is more resilient to turbulent airflow than a very complicated wing section.

If Formula 1 is to introduce an underfloor that will produce more downforce it needs to be done in a tightly controlled way and it needs to be done in conjunction with a revised front wing regulation. Otherwise once again it will all be for nothing.

What set-up changes are allowed at a tyre test and could teams use this Pirelli running to get a better idea of wet set-up? Shouldn't Williams have jumped at the chance to join the test given how much it struggles in the wet?
Gary Stam, via email

Gary, the teams participating in a tyre test will be given a couple of sets of baseline tyres. This will allow them to tinker with the set-up and also allow the driver to come to terms with the conditions.

After that it is just down to fitting the sets of test tyres and the driver coming back after the requested run distance and informing the tyre company of their impression.

If the tyres are a bigger diameter or the tyre company already knows what it is doing will cause a fairly dramatic balance change, it will allow a controlled ride height change or a small front wing adjustment.

Normally such tyre tests are carried out 'blind' - in other words the team and driver have no idea of what the objective is with each new set of tyres, they just work to a schedule laid down by the tyre company and report their findings.

On your last point if I was at Williams I would have been the first to sign up for the test. Even though you can't do much set-up work to the car during the test, it does allow that bit of extra thinking time and with any wet test it is a fairly well controlled environment.

Reading about the recent wet test made me think about what effect rain and spray has on the cars aerodynamically. I've never heard anyone talk about it, but surely the water will make a difference to airflows, downforce etc?
Mike Bates, via email

Yes Mike, rain definitely makes a difference to a car's aerodynamics. It does depend on the amount of rain, and if it is heavy anything that is near the ground is affected dramatically. This is mainly because of the water changing the road surface.

The teams will normally reduce the front wing angle to make the car understeer a little bit. This helps with this problem and gives the driver more confidence.

In the wet a car's aerodynamic characteristics will alter slightly and the downforce levels will be a little bit higher than in the dry.

On some cars this can lead to a performance switch from the dry. It is not only the driver that contributes to a car's performance when conditions change.

I read the designs of the first driverless racing cars will become public soon. If you were building a single-seater conventional in every sense except being driverless, what would the main differences be?
E Martin, via email

Actually when I read this myself it brought back some memories.

When I was building my own Anson Formula 3 and Super Vee cars in the early 1980s I had contact from what at that time I considered a mad man (and probably still would) who wanted to build 20 radio controlled F3 sized electric racing cars to allow all the past and present motorsport champions to compete together at Indianapolis.

The drivers would be in a control tower with a full view of the track to allow them to see and control their cars.

Can you imagine the carnage?

Nearly 40 years on and with driverless cars becoming fashionable, perhaps it is not such a bad idea.

It could be done for many reasons and I would initially drop the idea of it being a champions' competition.

If I was involved it would be to see which manufacturer had got the best handle on what a driverless car needs to minimise risk, it would consist of a total distance and seeing who could get there just using the on car sensors with minimal bumps and bruises.

In your opinion, are there things that are standard on F1 cars that are technically against the rules even though they have been allowed?
Ben Stern, via email

Ben, that's a very tricky question to answer. Every team pushes the limits to the maximum and sometimes they will go past the control line just like a driver will go slightly outside of the track width line. Every little bit counts.

The regulations are more about interpretation. Take for example body and wing flexing - this has been around for many years and if you look closely at the TV pictures you can see the front wings on some cars deflecting.

As time has gone by the FIA has added points where body and wing deflection tests will be carried out.

This is great because it allows the teams to make sure their systems withstand the test levels but it also lets them look at other non-tested areas to find advantages from deflection.

Some teams (who shall remain nameless) do push the limits above and beyond what is I think is acceptable. A component deflecting slightly under load is very different to a component specifically designed to deflect under load for aerodynamic benefit.

Why is driver extraction in F1 different from closed cars like those used in sportscars and rally cars? If it's not different, why are canopies seen as a problem?
Bruce Merchant, via Twitter

Bruce, change is seen as a problem.

I agree with you - sportscars, touring cars, rally cars, NASCAR etc all use some form of closed cockpits and still the drivers manage to get out of them.

Personally I am not a fan of this new proposed 'halo' system that F1 is considering. It not only looks stupid but I think it is being conceived for one type of incident: a loose wheel and tyre.

It will do nothing to reduce the risk of head injury when smaller objects are flying around.

At the moment the cockpit opening size is defined in the regulations and with a little bit more definition I am pretty sure that a see-through canopy could be made that would slide into fastenings as it was pushed rearwards into place.

Current F1 cars contain a huge amount of energy both in their ERS storage and in the hydraulics that operates most of the systems.

I am sure it is not beyond the thousands of very clever design engineers that work in F1 to come up with a system that will clamp a canopy into place when the driver requests it and also release it when the driver or marshals request it.

When designing an F1 car, how do you go about setting and testing the strength of components? For example, can you manufacture suspension components in a way that they are less likely to break when there is an impact?
Ben Davies, via email

Ben, components breaking off a car in a crash is actually good for reducing the 'g' levels seen during an impact. It is when components break off of their own accord that it is worrying.

Weight is vitally important to performance and every team pushes the limits to the maximum to save grams. Every accident is different so you would end up with very heavy components if you were to cater for every possible incident.

Teams gather enormous amounts of data as the car is going around the track and this data is used to understand the working loads that every component on the car has to withstand.

Each of these components will then be designed to to this level plus a safety factor that is based on the level of confidence the team has in its data and in its design capabilities.

Budget will also be an influence - if you have lots of money then the component life can be reduced and parts replaced more regularly.

If a component has to withstand say 1000kg with a safety factor of 1.5 then the design load will increase to 1500kg.

After manufacture the first of each component will be tested to destruction to see that it meets the design criteria.

As that happens, a force to deflection graph will be produced to monitor the component's stiffness and strength.

After this every component manufactured will then be proof tested to around 80 per cent of its working load and a load to deflection graph will be produced. This allows this component to be proof tested again at any time during its life and the load to deflection graph to be compared to check for component defects.

You've talked about doing windtunnel work with the car effectively in yaw and with steering lock on etc. What sort of range of settings do teams run through, number of degrees of lock etc?
Darren Yates, via email

Darren, there are basically four states that a racing car goes through as it goes around a racetrack.

1) Going down the straight and as it is gaining speed the downforce is increasing. During this period it is the drag level that you try to reduce - succeed in this and it will give the car more top speed.

2) Braking at the end of the straight. During this you get a weight transfer from the rear to the front axle and the front ride-height will decrease and the rear ride-height will increase. Rear stability is the priority during this period, succeed in this and it will give driver more confidence to brake later.

3) Mid-corner is the most complicated area. The car is loaded up so some roll will be evident, and also some yaw to generate the tyres' slip angle, plus steering lock. How much steering lock you have will depend on the corner radius, but something like three degrees for fast corners, six degrees for medium-speed and nine degrees for slow-speed is about right.

Car balance mid-corner is what generates the corner speed and normally the slower the car is going the more understeer it has, so steering lock characteristics are vitally important to tuning this balance.

4) Exiting the corner when the driver wants to get the power on, the rear of the car sits down slightly which helps with the traction but can upset the aerodynamic balance.

Get a good balance compromise between the weight transfer and the aerodynamic balance and the driver will get on the throttle earlier and with more confidence. Any speed gained will be carried all the way down the straight until it all starts again.

Obviously each of these four states gets broken up into many different configurations but for me the most important is the transient aerodynamics.

A racing car is never in a steady state when it is going around the track and it is the transient aerodynamics that are prime mover in giving the driver confidence, where they are standing on the brake pedal, turning the steering wheel or getting on the throttle.

If a driver has to wait before they have confidence in putting an input into the car that is laptime lost.

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