Michelin problems at Indy 2005

[Jersey Tom]

My assessment - Michelin didn't have an appropriate duty cycle file before the race for what was going to be seen on track. They did their durability work to what was an incorrect projection of a lap at the track, and it caught them out. That they could recreate the failures on a lab durability machine afterward, using DAQ, I think points pretty clearly to this. Takes away that grooving having anything to do with it, or any "resonance" or other malarkey with the suspension, etc. Durability machines are not exceedingly dynamic or exciting anyway.

Tire failures in general generally aren't that complicated. Too much quasi-steady state load, leads to too much heat and/or strain... and pop. That or a cut down anyway. Not to say they don't happen - but I've never heard of or seen a "resonance" failure in a tire. Nor have I seen it really at all, failure or not, with the exception of noise. Don't think the Michelin tires failed from being too loud for their own good.

That said, I'm not going to turn this into a pissing contest any more than it already is. At the end of the day, I have my assessment and you have your theory. From working as a tire engineer and having had discussions with engineers working in F1 at the time - I'm pretty confident in mine.

The more interesting thing is why they didn't have an appropriate load trace to test with beforehand... and that's something I won't get into further, with the things I've heard or with speculation.

[hardingfv32]

My assessment - Michelin didn't have an appropriate duty cycle file before the race for what was going to be seen on track. .... That they could recreate the failures on a lab durability machine afterward, using DAQ

Rough what type of machine produces this information? Are these units propietary to each manufacture?

Would this machine generate the 'load trace' you referenced?

Brian

[feni_remmen]

On this topic, I recall some great pictures of these tyres deforming massively on the cars at speed at Indy. One was printed in Autosport on the McLaren showing a sidewall twisting over on itself. Any chance people could post some of these as I'd love to be able to see these again.

[WhiteBlue]

A simple durability problem would not see tyre failures on lap one or two. It was clearly stated by Michelin and confirmed by team results that one pass around the banked corner destroyed the tyre. At least the manufacturer did not think that a single pass was safe.

feni_remmen's observations of massive tyre deformations beyond reasonable limits are typical for resonance catastrophes.

[youtube]http://www.youtube.com/watch?v=j-zczJXSxnw[/youtube]

Just watch the classical case of the Takoma Narrows bridge. You see crasy deformations befor the thing fails. This is resonance 101. Every mechanical engineering student has probably seen it. Any mechanical system that has a spring and damper rate can resonate. Certainly a tyre can. It is not unreasonable to think that Michelin went into the uncharted with their 2005 F1 tyre and encountered what no other tyre manufacturer did before.

[flynfrog]

so your proof is the Tacoma Narrows bridge?

From what I remember the banking was loading the outer shoulder of the tire much more than expected. Hence over heating and pop. It was either that or the Illuminati and the grays spraying chem trails causing the break down of tire rubber.

[strad]

You're gonna compare Galloping Gertie to an race tire?

[Jersey Tom]

Just watch the classical case of the Takoma Narrows bridge. You see crasy deformations befor the thing fails. This is resonance 101. Every mechanical engineering student has probably seen it. Any mechanical system that has a spring and damper rate can resonate. Certainly a tyre can. It is not unreasonable to think that Michelin went into the uncharted with their 2005 F1 tyre and encountered what no other tyre manufacturer did before.

OK, and I can post a picture of a tensile test specimen breaking under load and say this is Materials Science 101, every mechanical engineering student has probably seen it. Anything can be overloaded, certainly a tire can. Blah blah blah.

As for "Michelin going into the uncharted" ... F1 tires are fundamentally pretty simple and lean creatures. In some aspects probably more simple than a consumer tire. Getting the specific materials and shape tuned properly is really the challenge, and incidentally I think that is one thing Michelin was doing quite well.

I just don't see what it is that would point to some "resonant" behavior leading to the rapid failure. Again, not that it doesn't necessarily exist but I haven't seen such a thing.

To me, the most plausible scenario is that the tires were developed to a non-representative load case, with little margin for overload. Why they didn't have an appropriate load trace is open to discussion or speculation.

[FW17]

A simple durability problem would not see tyre failures on lap one or two. It was clearly stated by Michelin and confirmed by team results that one pass around the banked corner destroyed the tyre. At least the manufacturer did not think that a single pass was safe.

Heard it as 12 laps as the safe limit, an option was available for Michelin runners to change tyres every 12 laps which they did not take.

Let us not fault only Michelin, Goodyear had a similar problem with NASCAR in 2008. Wonder if it is resonance problem there too?

[hardingfv32]

I found a good 700 page paper on "The Pneumatic Tire" that answers a lot of my testing questions. There is a chapter on Waves in Rotating Tires. Does that relate to the latest discussions?

http://www.nhtsa.gov/staticfiles/saferc ... 10-561.pdf

Brian

[strad]

Sorry but I live in the area and the story is very well known locally..In short the bridge deck was like a wing and under windy conditions it would generate lift and rock and sway...It did it from the opening until the collapse..They cut corners and left off these side rails that would have prevented it.
BTW It's Tacoma

[strad]

I found a good 700 page paper on "The Pneumatic Tire" that answers a lot of my testing questions. There is a chapter on Waves in Rotating Tires. Does that relate to the latest discussions?

http://www.nhtsa.gov/staticfiles/saferc ... 10-561.pdf

Brian

Interesting...thank you

[Jersey Tom]

Let us not fault only Michelin, Goodyear had a similar problem with NASCAR in 2008.

Problems? Yes. Similar? No, not remotely.

[bill shoe]

To me, the most plausible scenario is that the tires were developed to a non-representative load case, with little margin for overload. Why they didn't have an appropriate load trace is open to discussion or speculation.

Partial hypothesis-

The cars were not at their grip limit through the banked oval turn. They were very highly loaded on the banking, yes, but they were not at their grip limit like an Indycar. Therefore this banked turn would tend to be overlooked by F1 tire designers because there was little or no quickness to be gained there.

This might lead to ignoring or downplaying the oval banking in terms of the overall load trace. All the other corners were pretty slow/flat. I think the average load and peak load over the lap would be radically different if you analyzed with and without the banked turn.

Tire failures are usually like airplane crashes or scuba accidents-- multiple things usually go wrong before a failure occurs. The teams may also have been running pressure that was insufficient from a durability perspective in order to optimize performance through all the slow turns.

[WhiteBlue]

It is amazing how people can simply ignore the most basic points.

1. F1 tyre side walls do not derive their strength from rubber. Like any other composite the fibres are the load bearing part.

2. You do not get failures in the first or second lap if you misjudge your loads by a few percentages. Typically tyre failures that aren't related to piecing by sharps will occur towards the end of the design life. Many dynamically loaded components have a typical time to failure. Only massively overloaded parts fail very quickly. Resonance is the classical case were engineers see excessive and unexpected over loads. I have yet to read a single explanation why Michelin missed something like 30-50% of the relevant loads. The over loading must have been of that magnitude in order to reduce the design life to something like 10% of the plan. I will bet that the design life specification for the side wall construction called for at least one race distance.

[strad]

More BS