Beginner’s guide to F1 suspension

[DaveW]

Extract from the quote in Giblet's post (last paragraph):

Other innovative tuning elements featured on the FW31 include front and
rear devices called inerters. These inerters literally add inertial mass to the
un-sprung suspension without actually adding weight to the wheels, thereby
avoiding the pitfalls associated with extra weight. More inertia will reduce
the ride frequency or response time of the suspension just like a heavier
wheel would, and may be beneficial over different types of track surfaces.
In some instances a lot of inerter mass will be beneficial and at other times
no inertial mass will.

With respect, I don't think the paragraph is quite correct. In the configuration used in F1, inerters & increased unsprung mass have an effect that is similar for some vehicle response properties, but the effect on others is quite different. Overall, used sensibly, inerters can help the aero/mechanical set-up compromise. Whilst they may not always deliver an improvement in single lap performance, I can't imagine a situation where inerters would have an adverse effect on the performance of a vehicle with good structural integrity.

[Jersey Tom]

A few items:

1) Don't care how fancy the tire model is, it will never be an absolute recreation of reality. The important thing, IMO, for a game like this is to get something that's fun, somewhat representative, and responds directionally the way it should. That, I am fairly certain, is at least captured in their current model spec. So to say setups "don't correlate" isn't accurate IMO.

2) It's one thing to have a model that is capable of doing a very good recreation of the real deal. I have no doubt they've put something together to this effect. However, you still need real world data to form the basis of the model. I am near positive that iRacing does not have data for NASCAR tires, and would be very surprised if they had data for F1 or Indycar tires. As such, whatever model they construct for those may just be some extrapolation. Might "feel" good to drive - doesn't mean it's an absolute recreation.

[Giblet]

Tom, if iRacing was given full access to one Williams FW31, and were given full access to data, is that not the same data that is shared with (at the time) Bridgestone, or would whatever tire data they have collected be the property of Bridgestone as well?

I know the tire companies take the lions share of the tires back and shred them, sp there are limited sets.

[Jersey Tom]

To my knowledge, under Bridgestone F1 teams were not allowed to do F&M testing on their own. This is not uncommon. A tire company performs testing, and then "licenses" that data to teams under a non disclosure agreement. I would be surprised if that data - particularly the raw test data - was made available to iRacing. I could be wrong though. (There's also the question of what kind of data, how much, and how good it is)

The Williams chassis... all IP belongs to them, so they can send that out to whoever they want.

[HampusA]

A few items:

1) Don't care how fancy the tire model is, it will never be an absolute recreation of reality. The important thing, IMO, for a game like this is to get something that's fun, somewhat representative, and responds directionally the way it should. That, I am fairly certain, is at least captured in their current model spec. So to say setups "don't correlate" isn't accurate IMO.

2) It's one thing to have a model that is capable of doing a very good recreation of the real deal. I have no doubt they've put something together to this effect. However, you still need real world data to form the basis of the model. I am near positive that iRacing does not have data for NASCAR tires, and would be very surprised if they had data for F1 or Indycar tires. As such, whatever model they construct for those may just be some extrapolation. Might "feel" good to drive - doesn't mean it's an absolute recreation.

All i can say is try it, and form your own opinion.

Setups do not work from sim to reality, it´s been confirmed by many race drivers.
You might think the PTM is good but i can assure you it will be night and day when the new one arrives.
Therefore setups are drastically changed. Because the characteristics of the car it torally different.

Black and White is a good way to describe it.

[HampusA]

To my knowledge, under Bridgestone F1 teams were not allowed to do F&M testing on their own. This is not uncommon. A tire company performs testing, and then "licenses" that data to teams under a non disclosure agreement. I would be surprised if that data - particularly the raw test data - was made available to iRacing. I could be wrong though. (There's also the question of what kind of data, how much, and how good it is)

The Williams chassis... all IP belongs to them, so they can send that out to whoever they want.

To clarify. iRacing has nothing to do with Bridgestone.
They only use their text on the tires.

But, Eric Hudec said that the tires we have today is more like the hard tire used in 09. Which should explain some of the variation in laptimes on identical tracks.

the car and tracks are laserscanned down to the millimeter. Parts are also weighed individually.

[Jersey Tom]

I have tried it. It's fun. I agree that if I were to fire up iRacing, go to Charlotte Motor Speedway and enter what I know is a real world race winning setup.. it may not perform worth a damn in iRacing. However, that will likely not change with a new tire model - nor does it matter. "Different" doesn't necessarily mean better, btw.

If iRacing is not connected with Bridgestone or Pirelli and doesn't have access to their data, then I doubt they have a "true" representation of a F1 tire. Doesn't matter how fancy your model is if you don't have real world data to build it with. Hell, if you want a very elaborate tire model definition, just check here. Doesn't mean you can make an accurate representation of a specific entity out of it. But again, even if it's an extrapolation and more or less representative, that's fine.

In any event, to say iRacing setups don't correlate to real world is false. If I throw more rear spring and bar on a car in iRacing, it's going to free up (because of the tire load interation). That's a correlation. For a game, doesn't really matter whether 20 psi gives best performance in the virtual world and 22.5 in the real world - so long as the general performance characteristics go in the same direction.

[HampusA]

I have tried it. It's fun. I agree that if I were to fire up iRacing, go to Charlotte Motor Speedway and enter what I know is a real world race winning setup.. it may not perform worth a damn in iRacing. However, that will likely not change with a new tire model - nor does it matter. "Different" doesn't necessarily mean better, btw.

If iRacing is not connected with Bridgestone or Pirelli and doesn't have access to their data, then I doubt they have a "true" representation of a F1 tire. Doesn't matter how fancy your model is if you don't have real world data to build it with. Hell, if you want a very elaborate tire model definition, just check here. Doesn't mean you can make an accurate representation of a specific entity out of it. But again, even if it's an extrapolation and more or less representative, that's fine.

In any event, to say iRacing setups don't correlate to real world is false. If I throw more rear spring and bar on a car in iRacing, it's going to free up (because of the tire load interation). That's a correlation. For a game, doesn't really matter whether 20 psi gives best performance in the virtual world and 22.5 in the real world - so long as the general performance characteristics go in the same direction.

reserve your assements to when the NTM arrives.
Also, please use the same setups for both models.

I´ve talked to alot of real life racers and setups (like in every other game) DOES NOT correlate to th real world.

regarding the F1 tire. True, but they do not have any collaboration with any tire maker.

[Jersey Tom]

Don't think our definitions and expectations of correlation are anywhere near the same.

[HampusA]

Don't think our definitions and expectations of correlation are anywhere near the same.

I guess not.

[Jersey Tom]

Comes down to what you are expecting... and whether you expect "correlation" in the manner of relatives or absolutes.

But the most important takeaway point here is this: any model (vehicle, tire, damper, aero, whatever) is ONLY as good as the data that goes into it. Pretty much indisputable.

The vehicle model perhaps is a bit better to use as an example. You can have a very, very good vehicle model (let's say a full multi-body system a la ADAMS/Car)... but if you don't have the right weights, inertias, hard points... it amounts to ---. You must make the distinction between a "model" and a parameterization of that model.

On the vehicle side of things, I'm sure iRacing's cars are very good indeed - both as a model and parameterization since they do apparently measure hardpoints, take weights, etc.

On the tire side.. I'm sure this 'NTM' of theirs has all sorts of functionality. But, if they don't have some amount of REAL data for the specific racing tires to use as a reference, then your model outputs will not match reality. If that's the case, you will never have a 1:1 absolute correlation, if that's what you are indeed looking for.

[HampusA]

What i mean is that if you make a setup in the game for present tire model
A competitive setup that is.

And then take that setup and apply it to a real car you would not be competitive because the setup is made for a game which isn´t dead accurate to the real world.

Because there are things like slip angle that is greatly exaggerated so alot of those things are being explored with setups in the game.
Some setups you would not even get round a lap in in the real world.
Same with PTM and NTM. Setups will change ALOT.

Maybe correlate is the wrong word to use. But what i´m saying is competitive setups in iracing today is unrealistic.

[Jersey Tom]

Yes, I agree with that statement. If you were to develop great, well-balanced setup in iRacing and put it into a real car, you would likely need to do a fair amount of re-tuning because there is not a 1:1 dead-on match. There are a variety of reasons for this.

I do know that iRacing has spent time at TIRF doing tire testing, characterization, etc - i.e. data collection. I'm sure that with whatever new tire model they've developed, it will allow them to get closer to reality in certain areas.

However, using F1 and NASCAR as examples... if they do not have specific data on those tires, then there is no way of tuning their NTM to match reality. Or, they may have some limited data, observations from teams and drivers etc. Either way, I'd say whatever they come up with for their F1 and NASCAR (perhaps Indycar as well) tire models will be an educated guess or extrapolation. They might feel about right, but that doesn't mean that you'll have a 1:1 match.

However, that doesn't really matter. To you or I, the specifics of a F1 car running 22 psi hot pressure on the fronts vs 25 aren't really critical.

[HampusA]

True, things that will improve is the characteristics of how a tire should act.

Like the slip angles we have now which is extremely unrealistic.
Then you have tire wear,
More exact representation of tire temps

[Jersey Tom]

May I ask.. how do you know the slip angles are unrealistic?

Is this something that is output in the ATLAS data files?