Tire Degradation & Data

[Locust]

I love this thread. Thanks for all the data! Wish we had threads like this on any of the Evo forums.

[glamcem]

Originally Posted by PST

We've run nine hard, hot sessions on our RE71Rs (245/40R17) over the past month or so. Here's a graph showing lateral acceleration and speed:



Blue- Yesterday’s second morning session. 1:27.2
Orange- July 30th morning session. 1:26.4. RE71Rs & Tarmac 2s
Green- July 18th morning session. 1:28.1. RE71Rs & stock suspension

Track temps were within a few degrees of each other, so we can consider that a constant. Given the setup changes, the Blue session should have been noticeably faster than the other two sessions, but it wasn't. Why? We simply were not developing the lateral load that we had been in previous sessions. The RE71Rs are still soft, and have about 4/32nds left (rough average). But the grip just isn't there.



Here's the IR tire temps of those same laps. The blue session was as warm as the orange session, with the exception of the big sweeper at the end of the lap where the orange temps got much hotter. I can't help but wonder if we wouldn't see the same thing on the left tires through that shorter sweeper a few turns earlier where the grip was noticeably higher.

You'd think the tire temps would get hotter as they dropped off, but it appears here that the temps aren't getting as high. It looks like since there's more sliding friction and less static friction, the specific µ is lower, the force being imparted on the wheels is lower (evidenced by the lower lateral acceleration), so less work is being done. That plateau through that last corner might be a visual representation of the limit of the tires.

Anyone have any input?

what makes you think that you were not generating enough G load or Blue line should be faster? you know not all the setup changes yield gains, and it sometimes hurts.. and I am looking at two graphs I would interpret (as a data analyst, lol) as they’re pretty consistent and the only big difference between orange and the blue lines is where you see the orange line jumps ahead of the blue is about 8000 mark, which makes me think it’s probably because of a late brake/ trail braking since it looks like it carried out more speed throughout that specific corner (area under the orange and blue line is your difference) if you look at the IR temp graph (on a side note I don’t believe both IR or pyrometer readings since it’s almost impossible to read the temps properly, unless you have a racing team with specific equipment, even then those wouldn’t be 100% accurate) this also backs my suggestion since the sliding through that specific corner generates more heat ..

You do know that .8 secs of a difference can be caused by anything else correct ? such as 1-2 degree difference in ambient temps or driver confidence (or lack of confidence) combined with the possible reason I mentioned above


FYI, I personally never experienced that an Extreme performance tires heat cycle out (even some R comps don't even heat cycle out such as Nitto NT01 ), the tires (200 TW or above) typically feels greasy after couple laps because it's easier to overheat them not because they heat cycle out.. for that reason you can also destroy them easily, typically the good indicator for street tires and some streetable r romps is the treadwear not the sessions you put on them
I would personally keep using them to the cords they typically perform the best right before you see them corded but that's just me

[glamcem]

Originally Posted by Alan

I have 15 days of 3-4 sessions (20-30 mins) each on a set of Hankook RS3s and they are an absolute mess. I have four more days scheduled this year and will try to make it work. Finding a way to keep tires from heat cycling out (or at least slowing it), especially for us budget track day folks, would conquer the market. I guess the answer is cancel one weekend each year and use the money for a second set of tires.

This board should find a way to organize a tire exchange so non-track folks can use our heat cycled yet still plenty of tread track tires. They still work pretty well on the street.

that's 60 heat cycles, I'd be happy if I can get 25-30 HCs from my Z214s
like I mentioned in my above post heat cycle is not probably your concern, if you want the best of both worlds I'd simply get another set of wheels, wrap them with NT01s and use them during the track season on both track (and drive to and from track) since they typically last about ~10,000miles (100TW) without heat cycling out, assuming you don't put crazy amount of miles when DDin during the summer... keep in mind this is an R comp tire so you probably don't ever heat cycle out your RS3s

[GSpeed]

First and foremost, thank you for the detailed response. I love this kind of feedback, and challenging me to take a harder look at the data forces me to make sure I'm interpreting this stuff correctly. I'd like to go through and address your points individually.

Originally Posted by glamcem

what makes you think that you were not generating enough G load or Blue line should be faster? you know not all the setup changes yield gains, and it sometimes hurts..

The time difference has to show up somewhere, and that's the most likely place. Longitudinal acceleration is very similar between the two sessions. Acceleration values are similar, and I have no other reason to assume the engine is producing less power. Braking values are slightly higher on the July 31st session, which I could also attribute to a higher load capacity of the tires. Temperatures between the sessions were almost identical.

Originally Posted by glamcem

I am looking at two graphs I would interpret (as a data analyst, lol) as they’re pretty consistent and the only big difference between orange and the blue lines is where you see the orange line jumps ahead of the blue is about 8000 mark, which makes me think it’s probably because of a late brake/ trail braking since it looks like it carried out more speed throughout that specific corner (area under the orange and blue line is your difference)

I didn't do a good job establishing this in the original post, but that temperature trend does seem to be consistent across the session. Here's a plot showing front IR temps from both sessions. Each session has four laps shown, all within 1.5% of each other, timewise. The driver is very experienced and has traditionally been very consistent.

Originally Posted by glamcem

if you look at the IR temp graph (on a side note I don’t believe both IR or pyrometer readings since it’s almost impossible to read the temps properly, unless you have a racing team with specific equipment, even then those wouldn’t be 100% accurate) this also backs my suggestion since the sliding through that specific corner generates more heat ..

I would tend to agree with you on the IR consistency thing. I would not be comfortable making decisions comparing this data to another sensor setup on another car due to leakage current, etc. but with seven sensors on this car, they do appear to be consistent to each other. I'm most interested in comparing my data to previous sessions.

As far as the sliding thing goes, do you have a source on that? Not that I don't trust you, but I'd like to read more about that. It's something I've been thinking about a lot this past week and I'd like to understand the science behind it. Like I posted earlier, I would think that since there's less work being done in the form of less force being generated, there's less energy being converted to heat. But that may not be a proper assessment.

Originally Posted by glamcem

You do know that .8 secs of a difference can be caused by anything else correct ? such as 1-2 degree difference in ambient temps or driver confidence (or lack of confidence) combined with the possible reason I mentioned above

Yep. The driver/owner already made the decision to try another set of tires and see how that affects the car. As the engineer, I'm trying to look back through the data and see if I can conclusively prove that the tires are indeed dropping off, because I want to be able to see this in the data in the future in different situations.

Originally Posted by glamcem

FYI, I personally never experienced that an Extreme performance tires heat cycle out (even some R comps don't even heat cycle out such as Nitto NT01 ), the tires (200 TW or above) typically feels greasy after couple laps because it's easier to overheat them not because they heat cycle out.. for that reason you can also destroy them easily, typically the good indicator for street tires and some streetable r romps is the treadwear not the sessions you put on them
I would personally keep using them to the cords they typically perform the best right before you see them corded but that's just me

Yep, as mentioned before, the owner/driver already made the decision to replace the tires. They've got enough rubber left they'll probably last for quite a while for someone else, and less than a second of difference is indeed slight. But he wants to continue to test on fresh tires so we have a useful setup base. He's very experienced, and I'm going to trust his judgement on the feel of the tires. If he says they don't grip as well cold as they used to, I'm not going to second guess him.

That being said, I do feel strongly we're overheating these tires. You may be right that it's not a "heat cycle" in the traditional sense, but more simply overheating and degrading the rubber. But in Texas, there's no way around that. People are going to overheat tires, and we need to know how and why that happens.

[stugray]

Sounds like someone needs to invent a RF temp sensing circuit board that you stick to the inside of the tire.
It would log temperature cycles and extreme temps that could be read out with an RFID reader.
Hmmmm....
I have three sets of hand-me-down hoosiers for the race car and no way to know which ones have been abused.

[GSpeed]

Originally Posted by stugray

Sounds like someone needs to invent a RF temp sensing circuit board that you stick to the inside of the tire.
It would log temperature cycles and extreme temps that could be read out with an RFID reader.
Hmmmm....
I have three sets of hand-me-down hoosiers for the race car and no way to know which ones have been abused.

IR sensors would work well for that. I've been trying to sell people some for a while now, but no one's been really interested.

[stugray]

Originally Posted by PST

IR sensors would work well for that. I've been trying to sell people some for a while now, but no one's been really interested.

I have been looking into this for years.
I have even found the sensors for $20 ea. and built my own datalogging system for the racecar.
My next piece I intend to build is a better thermal logging system that has 4X IR sensors and at least 8X Thermocouple sensors.
My current system logs all engine parameters, displays it to the driver in realtime (if they choose to look), and is downloaded via WiFi to a laptop as soon as the car pulls into the paddock.


I told people about the thermal imaging of tires idea many years ago and got shot down because "it would never work" and "the driver could never use that much data".
I tried to explain that the data was not for the real-time use of the driver, but for post-race analysis.


Now if you watch F1 you will see that the techs in the pits are watching the tires using IR cameras transmitted from the car in real-time.
They can tell instantly if you flat-spot a tire or have alignment issues from contact.


My idea above about the RF tire temp datalogger was more for maintaining inventory than for diagnostics.
It would be nice to know "Oh, this tire looks like crap but has only had 2 heat cycles, while this other one looks great but has had 14 cycles".

[glamcem]

Originally Posted by PST

The time difference has to show up somewhere, and that's the most likely place. Longitudinal acceleration is very similar between the two sessions. Acceleration values are similar, and I have no other reason to assume the engine is producing less power. Braking values are slightly higher on the July 31st session, which I could also attribute to a higher load capacity of the tires. Temperatures between the sessions were almost identical.

Ok, I have to say that I didn't pay attention to the dates ..if these readings are taken on different days, temp/humidity differences alone would make up for the difference.. I see ~2 secs difference (in a 2 min lap) between my morning and afternoon sessions on a hot summer days .. of course my car is boosted so the fluctuations will also be higher but 7/10ths of a second difference is pretty normal so your driver is pretty consistent, indeed

I didn't do a good job establishing this in the original post, but that temperature trend does seem to be consistent across the session. Here's a plot showing front IR temps from both sessions. Each session has four laps shown, all within 1.5% of each other, timewise. The driver is very experienced and has traditionally been very consistent.

I should've said those IR temps look consistent in their inaccurate reading range in other words you can still use them as a reference to compare two different readings but I am pretty sure they are far from being accurate ... my method of estimating the tire temps is simply looking at cold and hot pressures right after the session to see if the car hits the target pressures, if the tires are within the suggested camber specs and PSIs you're good..

I would tend to agree with you on the IR consistency thing. I would not be comfortable making decisions comparing this data to another sensor setup on another car due to leakage current, etc. but with seven sensors on this car, they do appear to be consistent to each other. I'm most interested in comparing my data to previous sessions.

that still doesn't tell us anything about the target pressures, lowering the tire pressures will get you to the range quicker but will also add more heat whereas with the higher cold PSIs it may sound like it's easier to get those hot PSIs since we're higher to begin with right? but in reality it's now less likely to inflate since there are more air in the tire..

As far as the sliding thing goes, do you have a source on that? Not that I don't trust you, but I'd like to read more about that. It's something I've been thinking about a lot this past week and I'd like to understand the science behind it. Like I posted earlier, I would think that since there's less work being done in the form of less force being generated, there's less energy being converted to heat. But that may not be a proper assessment.

I don't have a source in hand but it's not really hard to guess that sliding/drifting will definitely increase the heat which will ultimately cause premature tire wear..you can search on Google and I am pretty sure it;s been discussed ..that's in fact the main reason it's easier to over heat street tires vs track tires, because of the harder compound on street tires they will slide easier and feel greasy because of that after a lap or two

Yep. The driver/owner already made the decision to try another set of tires and see how that affects the car. As the engineer, I'm trying to look back through the data and see if I can conclusively prove that the tires are indeed dropping off, because I want to be able to see this in the data in the future in different situations.

I should say there seems to be nothing wrong with the tires, at least it cannot be seen with the given data, you might be getting closer to the limits of the tires though, which is a completely different thing

Yep, as mentioned before, the owner/driver already made the decision to replace the tires. They've got enough rubber left they'll probably last for quite a while for someone else, and less than a second of difference is indeed slight. But he wants to continue to test on fresh tires so we have a useful setup base. He's very experienced, and I'm going to trust his judgement on the feel of the tires. If he says they don't grip as well cold as they used to, I'm not going to second guess him.

I think I should've been more clear on how I read the first graph it's pretty clear that 7500 mark on the X axis is where the braking zone starts so it slightly has more overall speed Compared to blue and green lines.. another thing to mention that the Orange and Green lines are almost identical on braking zones and corners except the 2500-4000 mark ..the difference would be more on a track with more elevation changes IMHO

That being said, I do feel strongly we're overheating these tires. You may be right that it's not a "heat cycle" in the traditional sense, but more simply overheating and degrading the rubber. But in Texas, there's no way around that. People are going to overheat tires, and we need to know how and why that happens.

I would suggest you to try NT01s (235/40/17s) especially if the car is still NA and slight strech with a 9" wheel.. I was very happy with my previous setup and wanted to try Z214s this time but I will most likely go back to that NT01s ..overheating also shouldn't be an issue with them,

[glamcem]

Originally Posted by stugray

I have been looking into this for years.
I have even found the sensors for $20 ea. and built my own datalogging system for the racecar.
My next piece I intend to build is a better thermal logging system that has 4X IR sensors and at least 8X Thermocouple sensors.
My current system logs all engine parameters, displays it to the driver in realtime (if they choose to look), and is downloaded via WiFi to a laptop as soon as the car pulls into the paddock.
.

don't get me wrong but I am still not convinced with the accuracy of the those $20 sensors, I think a better investment would be a TPMS sensors that can read and display the tire pressures (like the one that comes with Corvettes) ..

[glamcem]

@PST, I forgot to ask how you got these logs? Are those ODB2 readings ? (assuming they are not if they're not merged)...

that's another thing to mention, GPS readings (external or not) are also not that accurate especially when cornering since GPS readings are extremely insufficient for calculating the distance (delta in the corner), that's the reason why we see more fluctuations in braking/corner zones as opposed to straights, go watch some videos with overlays/dashboard and you'll see what I am talking about..

However, on the straights they do a better job than the OBD(Ecu) readings since overall tire size cannot alter the readings and it uses difference and time to get MPH.. from now on I will only use the ODB for engine data, engine speed and actual speed and use the GPS readings as a point of reference

[GSpeed]

Originally Posted by glamcem

Ok, I have to say that I didn't pay attention to the dates ..if these readings are taken on different days, temp/humidity differences alone would make up for the difference.. I see ~2 secs difference (in a 2 min lap) between my morning and afternoon sessions on a hot summer days .. of course my car is boosted so the fluctuations will also be higher but 7/10ths of a second difference is pretty normal so your driver is pretty consistent, indeed

Temperatures were measured across the track surface continually, and averaged across each session. The temperatures matched exactly, to within measurement error. All other environmental conditions were close enough we could not measure a difference.

Originally Posted by glamcem

I should've said those IR temps look consistent in their inaccurate reading range in other words you can still use them as a reference to compare two different readings but I am pretty sure they are far from being accurate ... my method of estimating the tire temps is simply looking at cold and hot pressures right after the session to see if the car hits the target pressures, if the tires are within the suggested camber specs and PSIs you're good..

I'm not sure I follow you here... You're saying you think it's more accurate to measure tire temp by going off of pressure readings before and after the session, than to measure tire temps realtime with industrial IR sensors?

Originally Posted by glamcem

I don't have a source in hand but it's not really hard to guess that sliding/drifting will definitely increase the heat which will ultimately cause premature tire wear..you can search on Google and I am pretty sure it;s been discussed ..that's in fact the main reason it's easier to over heat street tires vs track tires, because of the harder compound on street tires they will slide easier and feel greasy because of that after a lap or two

Keep in mind we're not talking about drifting here. We're talking about a slightly higher amount of slip angle, beyond the peak efficiency of the tire. There's a HUUUUUUUGE difference in drifting around a corner, and a few degrees extra of slip angle. I'm sure you understand the challenges and costs of measuring slip angle realtime, and why that's beyond the scope of this project. We're not testing tires here.

That being said, I'm still researching this behavior I'm seeing. I imagine slip angle vs. temperature is proprietary data that we're not going to find, but I'd like to know if this trend is backed by scientific testing anywhere. I'm not convinced the higher amount of slip is reason enough for the temperature difference. There's no additional energy being turned into noise, or smoke, or anything like that.

Originally Posted by glamcem

I would suggest you to try NT01s (235/40/17s) especially if the car is still NA and slight strech with a 9" wheel.. I was very happy with my previous setup and wanted to try Z214s this time but I will most likely go back to that NT01s ..overheating also shouldn't be an issue with them,

I appreciate the suggestion, but that would significantly cloud our comparisons. The RE71R is doing fine for our purposes so far, and changing compounds would more or less invalidate any of our previous tests.

Jake

[GSpeed]

Originally Posted by glamcem

@PST, I forgot to ask how you got these logs? Are those ODB2 readings ? (assuming they are not if they're not merged)...

that's another thing to mention, GPS readings (external or not) are also not that accurate especially when cornering since GPS readings are extremely insufficient for calculating the distance (delta in the corner), that's the reason why we see more fluctuations in braking/corner zones as opposed to straights, go watch some videos with overlays/dashboard and you'll see what I am talking about..

However, on the straights they do a better job than the OBD(Ecu) readings since overall tire size cannot alter the readings and it uses difference and time to get MPH.. from now on I will only use the ODB for engine data, engine speed and actual speed and use the GPS readings as a point of reference

I'm familiar with the concept you're talking about (accurate position data vs. interpolation), but these trends show up regardless of whether the data is aligned by GPS position (10 Hz) or wheel speed integration. All data was taken with an Aim MXL2. Keep in mind GPS antennae in cell phones are usually 1 Hz, so there's a significant difference in what we're using and something like Harry's Lap Timer.

Jake

[glamcem]

Originally Posted by PST

Temperatures were measured across the track surface continually, and averaged across each session. The temperatures matched exactly, to within measurement error. All other environmental conditions were close enough we could not measure a difference.

I understand but still doesn't give us enough information since there are many other factors (engine power, pad/rotor life, etc) that affect the lap times and tires alone cannot be held responsible for the overall lap times, especially when they don't heat cycle out and when the tests haven't been done consecutively

I'm not sure I follow you here... You're saying you think it's more accurate to measure tire temp by going off of pressure readings before and after the session, than to measure tire temps realtime with industrial IR sensors?

yes, take a look at this video, and listen closely after 3:38 mark
[ame="http://youtu.be/YhLzi5f49qE?t=3m38s"]https://youtu.be/YhLzi5f49qE?t=3m38s[/ame]

Keep in mind we're not talking about drifting here. We're talking about a slightly higher amount of slip angle, beyond the peak efficiency of the tire. There's a HUUUUUUUGE difference in drifting around a corner, and a few degrees extra of slip angle. I'm sure you understand the challenges and costs of measuring slip angle realtime, and why that's beyond the scope of this project. We're not testing tires here.

That being said, I'm still researching this behavior I'm seeing. I imagine slip angle vs. temperature is proprietary data that we're not going to find, but I'd like to know if this trend is backed by scientific testing anywhere. I'm not convinced the higher amount of slip is reason enough for the temperature difference. There's no additional energy being turned into noise, or smoke, or anything like that.

Jake I understand your points and agree with most of the things you mentioned here but I have to repeat; that sudden jump on IR temp readings for that specific corner where you see higher corner speed also backs my claim.. it looks like that relationship is pretty obvious, I am not judging the inaccuracy/accuracy of the sensors I am just comparing data horizontally

I appreciate the suggestion, but that would significantly cloud our comparisons. The RE71R is doing fine for our purposes so far, and changing compounds would more or less invalidate any of our previous tests.

it looks like RE71Rs are great tires along with the Direzza ZII and RS3s many people so I agree with you on that

, I was under the impression that you were not satisfied with the results hence the word "degradation" and since you said: " Blue session should have been noticeably faster than the other two sessionsWe simply were not developing the lateral load that we had been in previous sessions. The RE71Rs are still soft, and have about 4/32nds left (rough average). But the grip just isn't there." that is the thing though, we cannot tell the grip wasn't there by looking at the data in hand, especially when almost all the lines are identical..you might as well be reaching the limits of the RE71s since all those lines almost identical except a few places...and average corner speed hike on that 7500-8000 mark seems to be responsible for that (or combination with other factors I mentioned above)

In addition to that, we would expect the tires that don't heat cycle out perform better when tread is less , that's the exact reason why NT01s and many similar tires perform the best right before you see them corded or why people have their Toyo RA1s shaved.. now that totally is a different topic of course

Jake

[glamcem]

Originally Posted by PST

I'm familiar with the concept you're talking about (accurate position data vs. interpolation), but these trends show up regardless of whether the data is aligned by GPS position (10 Hz) or wheel speed integration. All data was taken with an Aim MXL2. Keep in mind GPS antennae in cell phones are usually 1 Hz, so there's a significant difference in what we're using and something like Harry's Lap Timer.

Jake

yes, I am aware of that but I am still not satisfied with my 10 hz external GPS receiver, both Qstarz and Dual XGPS 160 fail to give that kind of precision at corners..

I don't have any personal experience with Aim MXL2 but it seems like a solid option, no wonder why so many teams and TA cars I've seen have been using them.. it just costs too much