Tyre compound behaviour

[hollus]

I've fixed the image links for you. Now they show the images.
Please note (open the post for edit to see details) how I added a different link, to the .jpeg image itself. I left your original (URL only) links in place, but the urls are unnecessary. Also not the different tags, img vs url.

[Mike_s]

Thank Hollus,
Apologies to all for my incompetence

[Mike_s]

https://imgur.com/VVgHxb0
In the above pic, I have plotted complex modulus against frequency for 3 different temperatures. We can use a technique called time-temperature superposition to shift the curves along the x-axis to create what is called a 'master curve'. This enables the determination of behaviour at different frequencies and, once the master curve is compiled, it can then be shifted on the y-axis to determine behaviour at different temperatures.
The visco-elastic behaviour of the sample can be elucidated over a wide range of temperature and frequency of loading. This information can then be used to look at what tyre guys call 'hysteresis' and use this information to look at tyre warm-up and when the tyre stiffness will be too high for the tyre to 'envelope' the surface texture of the circuit, or too low to provide sufficient grip.
It is worth noting that the measurements I have done are within what is called the linear visco-elastic range; this means that it is looking at strains that are not damaging the structure of the specimen. The same instrument can be used to stress the material until it starts to exhibit non-linear behaviour, which would manifest on track as graining.

[Mike_s]

We quite often see drivers having to cool their tyres down after a 'push lap'. This suggestes that something in the tyre structure is changing when the tyre gets too hot. I used a technique called 'Differential Scanning Calorimetry' (DSC) to look into this;
DSC looks at heat flow in a material - think of a kettle; energy is put into the water and the temperature goes up in relation to the energy input. However, at 100 deg C energy input does not increase the termperature, but is used to create a 'phase change' in the water as it goes from liquid to vapour.
The DSC instrument is able to detect phase changes very precisely.
The sample was heated from -80 to +250 deg C and cooled and then the process repeated again, the results were quite interesting;

At ~160 deg C there was a distinct phase change in the sample, but on the second cycle the phase change did not occur. However, if the sample was left to 'recover' for a period of time, the phase change came back.
The DSC does not provide any information about precisely what is changing, and I have not combined this with rheological measurements, but it was interesting that I was able to detect something which we observe on the track.
We don't see too much blistering this season, but I was able to look at decomposition temperature using the DSC.
Decomposition starts at ~260 deg C. This seems like a pretty high temperature for the compound to reach, but given the temperatures of the brakes under hard decelleration, it isn't out of the question.

[Mike_s]

I was hoping to see a phase change for the resin in the compound melting, but it wasn't evident in the sample I tested...

[Mike_s]

That is more or less it for the lab work that I conducted. My objective was to see if useful information could be obtained from a specimen of discarded rubber, I believe that I demonstrated that it can.
I would have liked to conduct more work to see the effect of heat cycles on the specimen and to look at how the rheology changes after heating the specimen to 160 deg C and how long it takes for the properties to recover. Looking at when the compound goes from linear to non-linear behaviour woudl also be informative to see how this translates from lab to track.
It would also be interesting to look into whether different compounds exhibit similar visco-elastic behaviour, or whether there are differences in visco-elasticity between the compounds that would inform on improved tyre warm-up procedures; clearly to get the tyres into a working range more quickly it would be helpful to maximise the viscous behaviour, therefore longer loading times should help get more temperature into the tyres. The work was conducted with a commercial testing laboratory and the scoping work to learn how to mount the specimen onto the rheometer and generate the initial work was quite time consuming. That said, it cost me around 2000 Euros, which was quite a lot for me, but I'm guessing it is peanuts for an F1 team... Moreover, once the methodology is developed, the actual rheological measurements can be done relatively cheaply.
At present the teams are having to use a surrogate metric for tyre performance modelling, but I think that the rheological measurements I have done could be really useful in understanding how to manage tyre performance. I should be pretty straightforward to look at how heat cycles change behaviour and, in turn, how best to manage the tyres.
With regulations changing in the next years it might be more important to understand the compound behaviour too.
My marble weighed about 1.5g; The rheological measurements used about 0,5g, the DSC measurements used about 25mg, so not a lot of material is needed to conduct the testing.
Finally, it seems that Pirelli stipulate that teams may not conduct destructive testing on the tyres (except by using them!). Rheology is not destructive testing and DSC doesn't need to be destructive, unless you wish to look into thermal degradation temperatures. I used a marble, which is most certainly in the public domain, but the tyre technicians also clean the tyres up after quali runs etc, the material scraped from the tyres would give greater certainty about provenance of the specimens, also the heat cycles already experienced by the samples.
If anyone has any questions about the work, or interpretation of the data, I would be pleased try and help.
Mike