How can teams resolve tyre issues?

[mep]

Thanks 747heavy I was waiting for such a post.
I have a few questions about those test rigs:

Do you know any company who produces them?

On the pictures it looks like the tires don't run on real tarmac. Its either some steel band or a drum maybe covered with something.
How does this affect the results?

The tires that run on a drum should have a reduced contact patch due to the convex surface. On the other side the one running on the steel belt may not have the tarmac like surface cover because it might break away?
How does the convex surface of the drum affect the results?

Do they simulate the heat effects you normally have on a racing tire?
I think about heating up the surface they run on to track temperature.
Furthermore I think about heating a disc between the tires to simulate a hot brake disc who transmits heat over the rim to the tire.

What kind of data do they measure?

[DaveW]

..to turn your idea the other ways round... could it be that a tyre at optimum grip shows a certain acoustic pattern-and monitoring this could help to find out where to go in terms of setup to get back to this sweet spot??

Good lateral thinking, Marcus. It would have to work on-track, however, & I would be a little worried about interference from other acoustic sources.

Back to 747's last thought:

Hysteresis (modelled by springs placed in series with the dampers) would be one candidate, but experience would suggest that this is likely to be counter-productive (for a "circuit" vehicle, anyway).

"FSD" (frequency sensitive dampers) might be another, although they are not strictly frequency dependent. They might be helpful in some situations, but current implementations leave something to be desired, I think.

Inerters will modify the way dampers (& springs) work dynamically, & are used by F1 teams & I think they are frequently used to manipulate the way tyres are "worked".

Do you have other ideas, 747?

[DaveW]

The right frequency for maximum grip is 0Hz. Any oscillations will normally result in a net decrease of available grip. This is due to tyre load sensitivty, lag due to relaxation and a whole bung of other things.

This is the primary function of the suspension; i.e to isolate the bulk of the vehicle mass from the wheel mass to reduce contact patch force variations.

Might be true for road cars (or some race cars on wets), Tim, but not a race car shod with slicks. The latter have to be "worked" to reach & maintain the correct operating "states". The main cause of "tyre issues" in F1 is the fact that it is difficult to achieve that consistently at both axles, especially when aero control is (considered to be) the priority.

[Tim.Wright]

I can answer some of these. I can also say that tyre testing is a massive field with lots of literature available if you bother to seek it out.

Thanks 747heavy I was waiting for such a post.
I have a few questions about those test rigs:

Do you know any company who produces them?

The last picture is of the CALSPAN TIRF machine. No-one currently manufacture this machine as far as I know since it was built specifically for he facilty. They just use it to measure for customers. This machine is quite old.

MTS do a flat trac tyre testing machine. I dont know of any other company names

On the pictures it looks like the tires don't run on real tarmac. Its either some steel band or a drum maybe covered with something.
How does this affect the results?

I heard of a 3M grit surface, probably not unlike sandpaper to be used. It will affect the results, this is why tyre data is scaled and re-validated on every different surface you want to simulate.

The tires that run on a drum should have a reduced contact patch due to the convex surface. On the other side the one running on the steel belt may not have the tarmac like surface cover because it might break away?
How does the convex surface of the drum affect the results?

Like any test bench, there are comprimises which affect the data and it is down to the engineer to understand the restrictions on how the data can be used. There is enough data available from these tests to be useful without the complication of specifiying a tarmac surface

The convex surface or the lack of tarmac in the test machine is obviously not representative of the perfect road but they are done for practical reasons. If you really want to run the best test of a road on a real track then you need to go in the direction of the Bridgestone or Delft tyre testing trucks

Do they simulate the heat effects you normally have on a racing tire?
I think about heating up the surface they run on to track temperature.
Furthermore I think about heating a disc between the tires to simulate a hot brake disc who transmits heat over the rim to the tire.

Typically a warm up sweep is done to get the tyres up to temperature. Adding a braking disc would be quite a complication in the data.

The more and more complications you add to the test data, the more and more isolated your dataset becomes. I.e. you wont be able to compare you tyres to any others since they are measured in a different way. Not to mention the costs of the tests would be completely impossible and at the end of the day I guarantee you will have more data than will ever be of any use to you. Not only that but it will only be valid for one tarmac surface and one brake setup

Like I said before, the test data is useful enough with just the basic setups you see in the pictures above. Generating gigbytes of test data still wont engineer your car for you. This you need to do yourself.

What kind of data do they measure?

Usually they measure 6 forces and 6 moments at the wheel centre of contact patch depending on the axis system. They are;
Vertical force
Lateral Force
Longitudinal Force
Aligning moment
Overturning moment
Braking moment

Then some displacements
Vertical deflection (rolling radius)
Slip angle
slip ratio

General test parameters
Tyre temeratures
Tyre pressure
Road speed
Water flow (wet tests)

Tim

[DaveW]

Incidentally, & apologies if I am repeating myself, I had a customer GT vehicle on "my" rig a little while ago. The engineer declared the tyres to be "s--t" & wouldn't last a stint. The balance of the vehicle changed through a stint largely, he thought, because the fuel tank was away from the c.g. He reported back after the test that balance was now consistent, the tyres lasted two stints easily, & could probably be used for three. The intrinsic properties of tyres (& dampers) are frequently blamed for poor performance....

[Tim.Wright]

Might be true for road cars (or some race cars on wets), Tim, but not a race car shod with slicks. The latter have to be "worked" to reach & maintain the correct operating "states". The main cause of "tyre issues" in F1 is the fact that it is difficult to achieve that consistently at both axles, especially when aero control is (considered to be) the priority.

Ok we are talking about slightly different things here. Yes I agree that until the tyre is at temperature you need some "working" performed on the tyre. And probably also some "working" to maintain tyre temperature.

What I'm saying is that when you are at temperature, oscillations in the contact patch are a grip loss mechanism. I suppose the great challange here is to optimise the trade off between grip lost by oscillating the contact pacth, and grip gained by working the tyres.

Tim

[DaveW]

I suppose the great challange here is to optimise the trade off between grip lost by oscillating the contact pacth, and grip gained by working the tyres.

Absolutely....

[DaveW]

Tim,

That Delft/Bridgestone(?) test trailer is an horrendous device. I think it has a road vehicle tyre fitted to it, however?

I once (or several times, actually) saw a truck converted in a similar(ish) way (that was intended) to characterize F1 tyres. I'm fairly sure that test tyres invariably finished a run with no compound & without ever having reached operating temperatures.

The major problem with such test vehicles is that they really require a performance similar to that of the target vehicle in order to "work" the tyres in a representative way.

I believe the same is true for static tyre test machines. The main problem with "flat trac" devices is keeping the belt in place (& rotating) when slick tyres are loaded through their performance envelope.

F1 & road vehicle tyres are very different animals....

p.s. I recall suggesting, in my innocence & frustrated by test days terminating with unfinished business after an engine failure, that it might be a good idea to use a DFL engine for testing instead of the (then ubiquitous) DFV. I was told politely (considering) that tests with such a vehicle would be entirely unrepresentative.

[mep]

Dave do you want to say that those rigs are not optimized for race tire tests?
That is the idea I get.

The more and more complications you add to the test data, the more and more isolated your dataset becomes. I.e. you wont be able to compare you tyres to any others since they are measured in a different way. Not to mention the costs of the tests would be completely impossible and at the end of the day I guarantee you will have more data than will ever be of any use to you. Not only that but it will only be valid for one tarmac surface and one brake setup

Like I said before, the test data is useful enough with just the basic setups you see in the pictures above. Generating gigbytes of test data still wont engineer your car for you. This you need to do yourself.

I think this statement is correct for commercial tires.
For race tires I think you should get as close as possible to the actual working situation. This doesn't make the things more complicated it makes them easier because you don't have to interpret your data that much you just carry them forward to your car. The tires should at least have the temperature they will see on the racecar.

I am impressed by the Delft machine but I don’t think that’s the way to go. I guess test with this truck take much time and are therefore quite expensive. Also the speed of the truck doesn’t fit to racecar speed. I wonder if they tested there formula student tires with this truck?
I think a good rig should test the tire automatically. Just as example changing the chamber angle incrementally and measuring the data. So you mount your tire on, let it run for some time meanwhile you drink a cup of coffee and then you get the final plots and graphs for the tire.
(That’s just my opinion about it)

[autogyro]

I always think the simpler you make the tests the better.
Surely it is possible to do static tests with pre heated tyres.
By now the tyre companies should have sufficient data to transfer the data to dynamic.
Then they would have a range of suggested set up figures for the car.
Of course it still remains the job of the teams to develop this.

[Jersey Tom]

Tire testing is an immensely ... difficult field. I'll put it that way.

But it keeps me employed.

[autogyro]

Thats as maybe JT but as a team principle I would demand a great deal more information from tyre suppliers for these important components, than seems to be the case at the moment.

[mep]

Well in fact we don't know how much data they get.
Maybe we make a big deal about a lack of data and they have all what they need.
Frankly I don't believe so.

[marcush.]

In my view the key is to gather this data during FP1-3 .there is no sense in waiting for the track to rubber in but being able to adjust the car to the improving situation.
with a track known from years before you should be able to very early realise when
things do not develop as predicted and it creates further input earlier as just how to react to this new situation.

You can sit in the pits waiting and changing nothing when you feel you know it all.
But ending Q3 or even worse Q2 and having to admit you did not have a handle on just why the car did not work in those important moments there is no justification for
waiting in the pits for conditions to come to you.

[Pierce89]

..to turn your idea the other ways round... could it be that a tyre at optimum grip shows a certain acoustic pattern-and monitoring this could help to find out where to go in terms of setup to get back to this sweet spot??

Good lateral thinking, Marcus. It would have to work on-track, however, & I would be a little worried about interference from other acoustic sources.

Back to 747's last thought:

Hysteresis (modelled by springs placed in series with the dampers) would be one candidate, but experience would suggest that this is likely to be counter-productive (for a "circuit" vehicle, anyway).

"FSD" (frequency sensitive dampers) might be another, although they are not strictly frequency dependent. They might be helpful in some situations, but current implementations leave something to be desired, I think.

Inerters will modify the way dampers (& springs) work dynamically, & are used by F1 teams & I think they are frequently used to manipulate the way tyres are "worked".

Do you have other ideas, 747?

I F1 heard one team refer to using their inerter to "beat up" the tires to get some temp