Tyre Friction Coeficients on official vehicles, are they accurate?

In BeamNG, the Friction Coefficients defined for "sport" class tyres is defined as 1.0

Continental suggest that their Sport Contact 6 tyres can do better than that with up to 1.2 (or near enough).


In a recent video*, "Engineering Explained" has suggested that modern road tyres can do over 1.3.

So are the friction coefficients used in BeamNG comparable to real world? If so, are the tyres on the official vehicles correct?
Or is this done to account for other potential inaccuracies such as contact patch etc?





* Didn't want to embed it, Search following on YouTube: "How The 2020 Corvette Achieves Its Fastest 0-60 Ever"

 

Hmm, this is a good one. I believe that they feel correct, but I'm not sure if they used real-world coefficients for the games own tires. Probably not, as I suspect that a coefficient of 1.0 is comparable to 100% in-game traction. Still, not sure though. I think @Diamondback has come up with the tire system now used, if not please correct me. If so, best thing to do is to ask him I suppose...

 

Yeah, I find it kinda odd though considering that all other in game numbers are real world comparable metric values.
My understanding is that they replicate real world tyre tests in game and then compare behaviour. So the numbers they have must be the most realistic they could get with the current tyre grip model. But as mentioned above, while close to real world values, they don't seem to match up*.

As you say, the current tyre model feels pretty reasonable overall when compared to other simulation games, and in the recent refreshes of cars, some have leapt forwards considerably in terms of how they behave.

*In value, as opposed to behaviour.

 

I think part of it is the fact that tires have improved a lot in the past 10 years, and continue to do so. The grip coefficient values might have been reasonable 2012/2013 when the initial research was being done for BeamNG, but with new, higher end tires, way higher grip coefficients are possible. For example, the 2019 Prius is able to beat a lot of slightly older sports cars on a figure 8 track, purely because it uses new tire designs that give it a lot more low speed grip.

 

The math that BeamNG uses is correct... whether the values they use are correct... well... that can be up for debate I suppose...

The best of the best tires out there can really stick to the road with some absurdly high coeff of friction, like drag radials for instance... I think some of them can hit as high as about 4 which is just nuts. More down to earth road tires that are used on supercars like the Michelin Pilot Sport Cup 2 tires are also fantastic at gripping the road having been noted as practically being a cheat code for cars... I have seen people hopping between 1.3 and 1.4 coeff as guesses for what it can do... which... for a roadleagle tire... is also just nuts...

This brings us to the Beam tires though. Naturally there are a million and one sports tires out there on the market with all sorts of different levels of grip... where do Beams tires fall? Are they meant to be the top of the top tear or are they supposed to be more mid range... something just a little over your average consumer could still afford (like high performance wipers or something of that nature) That is what needs to be looked at... because if they are just mid tear sports tires... 1.0 coeff is probably bang on.

 

The grip coff beam uses is correct, the guy that made the current tires mechanics actually made a test rig as well that they use IRL to test the slip angle and such of a tire.

 

But then the race tyres in game (slick tyres) use 1.3 as their grip coefficient, which as you say, can be had in a road tyres.

 

Right, well, as you said you think the Coefficients might be off... and it is possible that they are off for one tire type and not for another.

As far as racing tires go... there are many different rubber compounds used even when we are talking about straight slicks. In Grand Turismo they often simplified it to just soft, medium, and hard racing tires... but naturally there are many more grades then that. Each of these has its own differing ability to generate friction basically following the rubber stiffness. The softer the rubber, the higher the grip (to a point anyways).

So... where am I going with this... I am going to adhesion. Sticky race tires (or even hp road tires... the ones on my corvette do this too) will become softer with heat and actually begin to adhere to the road surface, pressing into all the little voids and imperfections, thus adding to the the overall grip the tire can achieve. This only happens when the tire is up to temperature though... when cold, they will usually follow much more standard laws and have much lower coefficients of friction (unless they have really soft rubber like drag tires).

So what I am saying is that for a cold tire... their numbers seem reasonable... and we all know that tire temps are coming (eventually) so perhaps that will be a part of that update and these values are just in preparation, and until then, we are all just driving around on cold tires (well... warm tires... cold tires can be very slippery)

Now that’s just speculation... but honestly it would make sense.

Granted it could also just be that the tires they modeled in game are meant to be older technology tires... that would also justify it... even just the past few years have shown serious progress in tire technology, never mind tires from the 90s or earlier...

 

I do seem to recall that they referenced a real tire for the "sport" tires specifically.

I suspect a lot of it has to do with suspension design; try as I might I can't get a Grand Marshal to exceed 0.87-0.88g at a constant speed and that's with some pretty extreme sway bar stiffness; an ETK K-series at the same speed with the same wheels and tires can get much higher.

 

https://www.beamng.com/threads/guys-come-to-talk-about-tires.58720/

There is a ton of good info here...

 

Suspension is a pretty large part of that, the GM is not gonna be able to keep it's rear tires angled correctly for peak grip with it's live axle. Weight distribution also matters, GM is front-heavy and so it will overwork the front tires, while the K should be pretty well balanced. Weight also matters, BeamNG cares about the loading of tires too. Both tire deformation through the Jbeam and the fixed tire properties matter, tires won't achieve their grip coefficients if they have to little weight on them to properly conform to the ground, or too much that it compresses the sidewall.

That begs the question of how Continental tested their grip coefficients. Some kind of testing machine applying optimal force at the optimal angle on a prepped surface would be able to inflate numbers quite a bit beyond what could be expected in real life. Testing them on a supercar being driven in a very specific way could probably also inflate the numbers a bit.

 

That's the funny part - the GM, back down on 235/15R15 sport tires (pretty much no difference between that and the ETK rubber bands in terms of constant-speed grip), can momentarily exceed 1g with heavy trail braking (though road topography may have played a part, for reference this was in the "pistol grip" of the Utah canyon heading westbound), but as soon as it stops decelerating, it starts understeering. If the fronts were already overloaded you'd think trail braking would overload them more but this appears not to be the case. Somehow adding weight to a tire that's properly loaded overloads it and reduces its grip, but adding weight to a tire that's already overloaded gives it more grip. That's always kind of confused me, not going to lie.

 

I'm not exactly sure about that, but I'm pretty sure it's due to the ratio of downward force to lateral force. The GM's center of mass is up front, so the front tires need to work more to turn it. When you weight transfer forwards, the GM isn't heavy enough to exceed the design limit of the 235s, so extra weight up front increases, rather than decreases grip. While a tire is not exceeding it's design limits, it's grip should be proportional to it's downward force. That's what the grip coefficient is, how much a tire can oppose lateral force given a certain downward force. This means the weight transfer increases your front grip at the cost of rear grip, but since a little slip in your rear tires (oversteer) will help you corner even more, you don't notice it unless you manage to unsettle the car a lot.

This is also why trail braking can be super lethal in rear engined cars, your grip gets thrown forwards in the same way, but since the center of mass is in the back, the rear tires should need to work harder to turn your car. This is what causes in snap oversteer, with very little weight on the front tires normally they slip and understeer easily, but weight transfer can suddenly give the front tires more grip while taking away rear grip, resulting in the weight of your engine pulling you around. On the flip side, you get way higher corner exit speeds since you don't really need extra grip up front once you've entered the turn (your inertia will keep you turning), so you can get on the throttle very soon in the corner, weight-shifting almost entirely to your rear tires for grip under acceleration.

 

Thank you, I feel like I understand it much better now. Also feel kind of stupid that I needed someone to explain it to me.

 

There are two categories. Everyday and special.
Everyday tires are cheap and mass produced. They include standard, all-terrain and sport tires
Special tires use more advanced rubber and mostly used in motorsport, They include race and rally tires. They have better grip than everyday tires.

Tire tread, The more the tread, the better the off-road traction and the worse the tarmac traction.
In BeanNG:
Standard Tires : {"frictionCoef":1.0},{"slidingFrictionCoef":1.0} {"treadCoef":0.7}
Sport tires: {"frictionCoef":1.0},{"slidingFrictionCoef":1.0},{"treadCoef":0.5},
Rally Tires: {"frictionCoef":1.05},{"slidingFrictionCoef":1.15},{"treadCoef":0.8},
Race tires: {"frictionCoef":1.3}, {"slidingFrictionCoef":1.3},{"treadCoef":0},
All-Terrain Tires:{"frictionCoef":1.0},{"slidingFrictionCoef":1.05},{"treadCoef":0.85},

Actual grip friction:
Tarmac grip: Race>Sport=Rally>Standard>All-Terrain;
Gravel grip: All-Terrain=Rally>Standard>Sport>Race;
Mud grip: All-Terrain>Rally>Standard>Sport>Race.

As you can see. Rally tires are as good as All-Terrain tires on gravel, and as good as Standard tires on tarmac.

If you grab a race tire and apply 0.5 tread, it will become the best road legal sport tire. That's why supercars and luxury sports cars claim their 1.1~1.2G lateral acceleration.

Several factors can affect the actural grip:
Loading: If the tire is heavily loaded, tha contact area will bend inward. If the tire is too lightly loaded, the contact area will bend outward, both will lower the contact area and lower friction.
Tread: If the tire is ideally loaded, and the frictionCoef is X, if the tread is 0 then the actual friction is X on tarmac. if the tread is other than 0 the actual friction is lower than X on tarmac.
Temperature: Not simulated in BeamNG
Width: Wider tires have sightly better grip, up to an extent, and cannot exceed the theoretical frictionCoef.
Downforce: BeamNG doesn't simulate slipstream and ground effect. All coltris are calculate one by one and add together. You cannot increase downforce by lowering the suspension.

 

This is interesting because in real life, don't sports-type tires (summer tires, especially at the high end) actually use a different (stickier but shorter-wearing) compound than normal all-seasons or is it really just a difference of tread depth?

Also, as to gaining grip by lowering, wouldn't the lower c.g. still have an effect?

 

My mistake. The meaning should be You cannot increase downforce by lowering the suspension. Edited

 

Suspension geometry (and vehicle handling) often feels like a dark art to me. For example, it was only the other day that I found out that fitting wider front tyres to a front engine rwd car can help to mitigate oversteer. Not only does it make the front end a little less twitchy, but once the limit has been exceeded, it make it easier to catch the car and bring it back in line. I was aware that moving towards square tyre setups had advantages (as well as disadvantages), but it wasn't until I had done some testing that I realised how much front tyre widths affect the rear end.

My understanding is that you don't really gain grip by lowering a car. However, you do change its handling characteristics. For the sake of simplicity, I am going to assume lower with correctly adjusted suspension for said ride height and surface.

So you get less suspension travel for pitching and rolling under accel/decel and turning, this will affect how the car transfers it weight, often allowing the car to settle more quickly after an input.
Often you will lower your roll centre. You can look up roll centre if you want a proper description. But in effect lowering the car will change the direction that any wishbones/links are angled at, especially in a double wishbone setup. Lower roll centre also affect weight transfers and has a surprisingly big effect on vehicle balance and handling dynamics.
Ride height is incredibly important for cars that are running down force too. Since they need to suck themselves to the floor.*
Many suspension designs will actually change the alignment of the vehicle as they progress through the travel. If you start off further into that travel because you are lower, it will change how that setup behaves.

When you lower a car, you can actually be making some very major changes without realising it.

*but as @default0.0player mentioned, beamng does not fully simulate aerodynamics. That means ground effect is not simulated at all, so diffusers don't work. That said, similar affects can be achieved by placing "fake" wings where a diffuser would be found etc.

 

Older sports cars have far worse weight distribution, just test this.
Put the same tire onto the 200BX and the Barstow. Choose the Street Tuned and the Nightsnake, put Clockwise 17X12 Rear Wheels and 285/35R17 Sport Rear tires. In 0~60 mph test, the 200BX is slightly faster despite much higher weight/power
--- Post updated ---

For a given splitter/diffuser geometry, the lower the ground clearance, the more the downforce. "Fake" wings does work, but ground clearance doesn't has any effect to it.

In the above picture, the lower the ground clearance, the ligher the V2, the lower the P2, the more the downforce.

Since aerodynamic parts are calculate separately, removing the bumper, trunk or window can actually reduce drag

 

Lowering a production car without adjusting or rebuilding the suspension geometry is actually bad for handling because your wheelcamber and roll center will be off the only advantage you are getting is less drag