As described in another thread, I put on a set of Michelin Pilot Super Sports last week and I used a DE day this week to find their optimum pressures for track work. I thought the full procedure might be interesting to those tempted to "just bleed a little air" when they feel like their tires are getting 'greasy' on a track day.

First, the participants:

This was my third track day. I haven't used my competition license in a little over 17 years, so I necessarily qualify as a novice under the rules. It's become something of an in-joke at the drivers meetings, but knowing my experience the chief instructor assigns me an appropriate 'instructor' each time. Depending on what interest I've expressed. When I first signed up for a track day after buying this first Porsche we've owned, I said I wanted to learn their handling nature. He put me with a local autocross champion who drives a 1976 Targa with monoball suspension, corner-balanced and wearing race rubber, roll cage, and so forth. I think he figured that if anybody could talk about Porsche-specific hints for track driving, it would be a guy used to hustling that combination around Willow Springs. Great guy, fellow engineer working at China Lake NAS, and we got along fine. Made friends and we get together regularly at the local Porsche club. He was at this track day on Monday, parked in the next pit, and helping with the MPSS evaluation.

This time, the chief instructor knew I was planning to run these tire tests and gave me a different 'instructor' with a lot of kidding about using up their instructor time and "Who is going to be giving who tips?" But again, he made a good choice. This guy is a German who grew up driving the Nordschleife, like that woman that visits on Top Gear. He still keeps a Porsche there and one here in Santa Barbara. He flies back and forth regularly and competes in the endurance races on the Nurburgring. His car at the track day was his local Porsche, a gen two GT3. The one based on the 2004 996 that is. It was shod with Michelin Pilot Sport Cup tires, as well as four-point belts and other items likely in a GT3 tracked most of its life. We all had racing pressure gauges. (If you care, mine is a liquid-filled unit with barometric-compensation from Longacre Racing.) Christian carries a contact pyrometer to events, but I couldn't find mine from the old days and had only an infrared unit along. He was going to be the one hopping out to take measurements, so we settled on using his equipment for all measurements to reduce the chance of error.

So we had three very experienced competition drivers, racing quality equipment on hand, and two of us were engineers. No actual race engineers, but you take what you can get. Now the big issue is that pit pressures are not the same as the on-track working pressures and they're a far cry from the cold setting before you fire up the car in the morning. Dealing with that requires some very specific goals or you get lost in a wilderness of pressures and temperatures and numbers that eventually seem unrelated in any fashion the human mind can grasp.

In ye olden days, we chose tire pressures as simply a way to fine tune the suspension. Tires are the pneumatic component of a race car's suspension. That's really still true, but my comments in an early thread were too simplistic for modern tires. I've been looking into modern developments. Turns out the recommendation of a specific pressure by Michelin UK is not as unlikely as it sounded to an old timer. I still may not agree, but it isn't as implausible as it sounded. They design modern tires to provide what is termed "contact patch management" or "contact shape control" or what have you. Basically, they computer model the behavior of the tire components at different pressures and with different levels and combinations of side load and linear thrust, and they have in mind a specific gage pressure level in order to predict the tire's behavior under those different loads. Concurrently, they select a tread compound, or compounds plural in a very high-end tire like the PSS. The compound choices dictate the working range of temps for the tire, but that sweet spot has become very broad. The Pilot Cup line has a recommended working pressure front/back of 32/36, amazingly specific to my intuition, but compensates with a startling "sweet spot" that stretches from 160F to 220F. Sweet 'spot'? That is as broad as a cake compared to the Hershey Kisses we had to work with. Sixty degrees of 'sweet'? Where do I target within that range? Because you can't convince me the grip is the same over that whole range. But more later.

I remain skeptical of translating all that to a single target 'hot' pressure for consumers of their tires. The 'hot' or 'working" pressure always varies from one track segment to the next as anyone with our old-fashioned data recorders or modern telemetry can attest. (Just watch the TPMS readings from a 997.2 for confirmation of this.) Picking one single value really means they are spec'ing a "pit pressure." That is, a value that will be seen when measured after leaving the track and letting the tread heat blend with the carcass heat for a few moments before applying the gauge. I am somewhat critical of their doing it this way, but customer relations never have been my strong point. If a single value improves that working relationship with their competition customers, then we could at least take advantage of what it tells us about Michelin's thinking. And we needed to read their minds a little for this effort, because they give us such numbers for the Cup tires, but not for the Super Sport.

We needed goals though. Clearly, modern designs do not respond well to the old shade-tree technique of "let's try this and then see what the lap time does." Worse yet, we didn't have the track to ourselves. The two youngsters (about 40 and 46) might have been able to produce consistent lap times within a tenth or so if the track were empty, but it was not. Certainly, at my age I'm not that good these days. My lap times were going up and down in whole seconds, for all love, not tenths. So we needed something else to measure besides lap times.

Elaborating the notes in my logbook, our thinking went like this. Michelin predict a rise from 25/28 cold (i.e. 68F) to that 32/36 'hot' pressure. That is seven psi in front and eight in back. The pressure increases come from temperature increases of course and that 7/8 psi rise equates to temperature rises of 100/114 for the carcass itself (which dictates the gage pressure. Tread temps are too transient.) Since the reference 'cold' temp is 68F, they are foreseeing 168/182 as carcass temps as the car sits in the pit after lapping long enough to stabilize the amount of heat in the tires. Let's drop this false precision. 170F in front and 180F in back. And even that is precise beyond track realism, so let's settle for "about 175 F". Since the temperature range for optimum performance of the Cup compound is 160F-220F, they are looking for a carcass temperature about one quarter of the way into that range. That's plausible, because during a lap the tread temps could be cooler than the carcass when the air flow and track surface temps permit, but only for short distances. Most of the time, segments of the tread are rising temporarily well above the carcass temp and then falling back to that steady value as air flow cools the outer surface of the tire. A target temp one fourth into that broad sweet spot means the tread can get only 15 degrees cooler than the carcass, but it can be as much as 45 degrees hotter in some corners and still perform well. A good place to put the target temp in other words, so we seemed to be following Michelin's thinking.

Now for the Pilot Super Sport. Where do we want to end up? We do know relationships we want between different parts of the tread, but more on that later. First, we need a target temp. Something like that "about 175 F" for the Pilot Cup. Where would it lie? Well, the Cup tires are essentially 'streetable' competition tires by design. Conversely, the Pilot Super Sports are 'competitive' street tires. We first thought 'trackable', but that's silly. Nearly any modern tire fits that description. I've taken an Acura sedan with MXV tires onto a track without dramatic incident. They are 'trackable' tires, albeit stone slugs. The PSS line clearly is designed for cars like Ferrari, Porsche, Koenigsegg and so on. They are meant to be street tires that are competitive on track days, though not outright racing. The important adjective there is street and such tires have need for cooler compounds for several reasons.

Street tires need a compound (or plural) that will reach working grip levels at lower temperatures. We don't have time for a lap to "put heat into the tires" when a child runs into the street. How cold might the PSS range be? Darned if we could decide, but one indication comes from my TPMS system which is pretty accurate. (Not precise at all, but accuracy is a different virtue and useful for this question.) In a drive downtown, the cold setting of 34/37 was going up to 39/42 pretty predictably. This is partially my driving style of course, but a five psi rise is a starting point for consideration. Five psi requires a temperature rise of 70 degrees F. Add that to the 68 F baseline for setting the tire pressures, round off to a reasonable precision and you have tires working at 140 F for the street with an aggressive driver. Alright, fair enough. Clearly, the compounds for a street tire have to be chosen to be at least that 'cold'. Add in drivers less aggressive and the tires on a dry day may be only 120 F just when a panic stop is needed. Now consider a rainy day with the road surface at 50 F and the tires efficiently cooled by the surface water. What would we expect? 100F perhaps? And we still need to stop short of that kid.

All in all, that broad range of sixty degrees that the Cup tires provide won't be enough for a predominantly street tire like the MPSS. And indeed, we learn that Michelin has provided multiple compounds on the MPSS tire specifically to deal with this issue. So the range of one compound will make up for the deficiencies at the extreme ends of the range of the other compound. The colder of the two probably works effectively down to 100 F and perhaps as low as 80 F. The 'hotter' of the two compounds won't reach as high as the Cup tires, but it certainly must overlap a lot. Perhaps a range of 130 F to 190 F?

See what I mean about a blizzard of numbers? At this point, we needed something specific, a goal we could work with while an open-exhaust BMW was roaring past the pit making conversation impossible and thought problematic. "The 160's" I said. No reason except intuition. Cooler than the Cup tires intended for serious racing, but warmer than the street levels. "Give me anything close to that and let's work for the right relationship between temps at the outer edge, the middle of the tread, and the inner edge." Done. Mind you, all these lead-up paragraphs are taken from about one dense paragraph of notes in my logbook and a two-minute discussion with David and Christian. That's why I pointed out the experience level of us three. If you don't have... well, more on that later.

Before coming to the track, I set the tires at 68F to 33/36 psi. I foresaw the need to drop pressures at the track so I wanted the baseline, my last precise value, to be that one psi lower than I'd been running since mounting the tires. I had gotten permission from the chief instructor to pull off driver's left shortly past track exit and let my passenger, uh... the instructor that is... get out and take measurements. He did only two tires each time, and measured them with the contact pyrometer "inner/middle/outer" in a loud voice so I could record the values inside. (Yes, yes. Mine is a bloody touring car, not a 'serious' car like his GT3. It has a pen holder that "falls readily to hand. "Sigh. He has a four-point-belt anchor. I have a pen holder. But I whupped him in the autocross. The pen is mightier than...)

We began with a warm-up in the first session. By the second lap, the tires had risen to 39/42. The same values I see all the time on the open road, but in this case it was a six degree rise since the tires began at 33/36. That's enough heat from a slow track lap, relatively slow, to add 85 F to the carcass temps. The encouraging news was the consistent rise in all four tires. At a slow pace (i.e. freeway speeds or a little more), the car's suspension was asking for about the same work from each tire. That wouldn't last, but it was a good starting point. Then we put in three laps at perhaps 7/10 to use the old racer's jargon for "a pace that isn't me really paying that much attention but I'm not getting run over by the ice cream truck either." With true novices out there in the run group, that was about as fast as we could go without continually slowing down and awaiting a point-by. Besides, we were getting acquainted and busy talking. We'd have done more laps, but these were run groups set up for the convenience of others not tire testing, so we used short segments to heat the tires after each adjustment.

To give you an idea of what we measured each time, here is the data from that first three laps:

LF: 171/164/150 41.5; RF: 156/164/161 41.0
LR: 158/165/158 44.5; RR: 164/167/158 43.5

You can probably decode that gibberish for yourselves if you care, but one hint: by convention the inner tread reading is the inner value, so for the left front the 150 is the inside tread, but the corresponding value for the right front is 156. Just tradition as much as anything. The important point of that first reading is just confirmation of our intuition. Worked moderately hard, the pressures rose differently in each tire, and all of them were high enough to make the center of the tread work harder than the shoulders. Not an optimum condition at all. We began work.

First, we lowered the rear tires by two pounds. We fiddled around like that for a couple of sessions. We worked on one axle at a time, to reduce the blizzard effect. (Because the pressures front to rear interact of course.) Our goal was temps in the 160 range, indicating we had the compounds working for a living, but most important was to achieve a desirable relationship between the readings. I already said you don't want the center highest. Working the center of a tire the most is neither optimum for tire wear nor handling. You do want the inner tread blocks to be hotter than the outer, and you want the middle to be about the same as the inner tread or slightly cooler. That was the pattern of temps we wanted.

One early conclusion was probably predictable. The car needs more camber to use the front tires effectively. That is, the camber setting needs to be more negative. Very likely, the toe-in is inadequate as well. That was food for discussion after the first two sessions. We had no in-pit arrangement to adjust the suspension. A 997 for the street does not make that feasible without some expensive equipment. Anyone with that stuff has one of the GT cars instead. It isn't a matter of turning some set screw "a little clockwise, Joe!" We agreed to use tire pressures to compensate for the suspension for now, but you should know that's not good practice. First, you get the suspension right, then you optimize the tire pressures, and then -- if you must -- you go back once more and tweak the suspension settings. For now, I declined that "right way" because I hadn't committed to setting up the suspension on this car for track work. I wanted to consider how much change the result of our tire work screamed for.

By late afternoon, we'd reached a point where the temps had almost the right relationship and I liked the feel of the car as well. It was responsive to steering input even in the low-speed corners like two, three and the skidpad return to the front straight. If you don't know "Streets of Willow", that just means the three corners that require speeds around or below fifty, depending on the car and the type of tire mounted. At speeds from fifty up to eighty or so, the initial turn-in was by steering but the car's balance in the turn was controlled entirely by my throttle input. I could take it down to an apex or drift it out to the edge just by throttle adjustments. That isn't to everyone's liking, but it is the fastest way around a track with most cars, notably rear-engine supercars like... well, you know one. And it certainly is second nature to me after driving Formula Fords which behave that way from first gear all the way up to their top speed.

The car is quite happy as the rear tire pressures and the fronts approach each other. That was obvious quite early. I must confess to losing a page of my notes in the episode I describe elsewhere. (Basically I nearly ran out of gas and tried to get over to the track pumps before the attendant left. I didn't.) We reached a point where the pit pressure was the same front and rear and we foresaw "as soon as we put heat in the fronts, we'll be running a Porsche with more pressure up front than in back." That turned out not to be the case. (Probably our good fortune.) The fronts had been hot from the previous run, notably hotter than the rears, when I mentioned that possibility. But with the final adjustment, I was able at last to get the rear tires working as hard as the fronts. Once I began getting heat into them the pressures sorted themselves out nicely. If memory serves, we had 37 psi (hot) in the pits, both front and rear. After a couple of laps, we had 39 front and 40 rear in real time. That is, on the TPMS. Or so I remember. As I said, one page of notes has been left for some archaeologist to unearth and decipher in the misty future.

At different points, we let out two pounds here, or two pounds there, and once we tried putting back one pound in the fronts. That was... not so easy as we had hoped with the on-board compressor. Should have brought along a proper unit. The trick of course, notes or not, is to arrive at effective pressures. Ones that work all parts of the tread in the relationship the supplier proposes. We did that, but when done, how do you know what you've learned? What cold pressures will produce those working pressures next time? Well, we did a little back figuring at the track, using the temp/pressure relationship, but that way is very imprecise. What you do is you take the car home (hopefully without crashing it owing to the track pressure settings) and let it sit overnight. That's what I did.

Next morning, at 68F, I measured the cold pressures that had made the car handle so well: 33/35. The final readings with the pyrometer still said we had the front pressure too high, so 32 psi is indicated there. Perhaps 31, but the truth is I like the car more when the front/rear grip balance is achieved by the tire sizes mostly and the pressures are close together. If I return to the track for a rump session, I may try 34 in the rear and see what the temp relationship is across the tread. If the rears are happy at 34, I could put the fronts down an extra pound and run 31/34. But for now, 32/35 is my baseline for track use. That gives me the handling I like in low speed corners while retaining the stability required in three-digit corners like the Waterfall at the Streets. What about road use?

One problem arises. The on-board TPMS considers 34/40 to be the 'spec' pressure, not even the revised 34/37 the GTS supplemental manual specifies. And TPMS alarms at four psi under spec. So with cold pressures of 32/35 it screams "Tire Pressure! Tire Pressure!" no matter what the current pressure reads. (It does its own temp/pressure correlation just like we engineers kept doing all day.) Well, I say 'screams' ... actually, it politely says that in a message. But this giant 'bang' symbol appears on the display to demand your attention. That is distracting mildly on track, but it would be intolerable on the road. Besides, on the road, I can't possibly get enough heat in the tires to keep the middle tread blocks up to temp without more cold pressure and these are pricey tires to be wasting half the tread. So I added two pounds front and rear to calm down the TPMS and to promote long tire life in road use. That gave me road pressures for the new Michelin Pilot Super Sport of... 34/37. Sigh.

But -- and this is important -- I'm not relying on those engineers at Zuffenhausen to pick a pressure spec for the car, by golly. We just did it ourselves with a day of effort from three drivers and un-ending kibitzers offering advice. Seriously, we got to those values by measuring tire response to the heat inputs of fast driving. The extra two pounds for road use are realistic adjustments to operating almost constantly in the low end of the temp range. So I can use those pressures knowing the handling is fine and that no one arbitrarily bumped them up for some politician's fuel economy initiative or any other extraneous reason we might conceive. For now, with the suspension settings at stock values, 32/35 is an empirically derived setting for my car on a track day. And two pounds higher all around is the value for ordinary road use: 34/37.

I'm pretty sure that after a couple of days to consider side effects, I'll go ahead and change the suspension to the most negative camber setting provided by the stock suspension in front and move the toe-in setting halfway between the current value (whatever that turns out to be) and the limit of adjustment. Then I may need to adjust that track-day baseline of 32/35 which means another track day, darn it.

Let's go back to our self-assured choice of tire pressure settings that by wild coincidence just happen to match what Porsche prescribes for the GTS now. Surely it didn't need our spending all day to convince people that Porsche engineers are pretty smart people. We did collectively speculate that political influences might be pushing their spec up a couple of pounds, but that turns out not to be the case. 32/35 just is the right pressure to use all parts of the tread of a very wide tire effectively when the suspension is set to factory recommended values. And 34/37 is an appropriate setting to use when only the open road awaits. It isn't driver style, but simple reality, to observe that heat can't be put into the tires on a public road the way we do on even a short 'technical' track like the Streets of Willow.

All that makes that track day useful as well as entertaining for me. But if you don't know how to do everything David and Christian and I just did, then why bother to second-guess the professionals at Porsche who do this for a living? Just a thought for your consideration. Of course, fiddling with your suspension settings will add another whole dimension to the questions you need to answer unless you can find a professional to answer them for you.

Gary

 

Hi there

Excellent write up and glad you found what I've found on track.

I am also now working with 34/37 cold pressures for the road.

On the track I think I maybe drive quite a bit harder and certainly do a lot more laps so I see pressures/temperatures increase further.

I basically aim for circa 36/38 hot as like you I find 3 laps tends to add circa 7-8psi, but when you've been out for 10+ laps or so your easily adding over 10psi, hence I manage my hot pressures around 36/38 psi which seem to give best results in feel and grip of the car.


Here are some videos from my trackday this weekend:-


[youtube]N5RuD_UhWYo[/youtube]


[youtube]yRVOlO6-if0[/youtube]


[youtube]g6Vhgt4mtww[/youtube]


[youtube]etjXRHMgdHw[/youtube]

 

Being a novice at best, that was a great read. Thanks for taking the time.

 

Good write up.

As i wrote earlier this year...

++++++

I have a base carrera with 18's so my experience is somewhat different. I run at 34-35 front and 36-38 rear HOT with Michelin Super Sports.


If I get too high in rear pressures, say over 42-43, the car feels like its on ice. Once this summer I went too low and that was not good either. Flet like I was driving in mud.

++++++

after a few more DE's I think optimal for NJMP with my car and skill level is 35/38 hot.

I have GT3 control arms which allow me more negative camber than stock.

 

Gary,
Two great posts - thanks!

 

I found the same, especially when rear pressures went above 42psi, felt very greasy, especially under braking.

 

Most excellent Gary!

 

Okay, that's a skeleton of what we did. Just as I said, the handling in the afternoon was brilliant. I want more camber and maybe toe-in to get that car behavior without abusing the confidence of my tires, but 'brilliant' is enough to say for now. So here's the question: I started at 33/36 cold and ended with 33/35 cold. How the hell did that one pound make such a big difference in on-track results?

I could just evade explanation by saying "It's a hill-climbing search and sometimes they work that way" but I'd rather not. At 110 mph going over a crest you want to understand your car, or at least I do. So about now you should be asking "Huh?" Or words to that effect. I do know because of all that collective experience I mentioned, but I realized this morning that I never explained. And we have a long 'medical' day ahead so I can't take time now. But you might want to think about it. I know a lot of you have a hunch already, even though you weren't there, but I'll let the others consider it until tonight.

Gary

 

Gary,

With the hot pressures you ended up with did you have much in the way of rolling on the shoulders? On my PS2s with those pressures I observed some wear just below the end of the tread on the outside sidewall. I ran higher pressures on my PSSs than you but I left my pyrometer at home. With the camber set at -2.5/-2.0 I didn't see hardly any wear on the inside tread shoulder and a slight amount at the extreme outside edge of the tread shoulder. The pressure being too high would make sense to me on that issue.

I've already put my pyrometer in the boot for Mid-Ohio on 10/15. I've got the lazy man's versions that holds and displays all 12 readings.

Thanks for the insight and your time.

 

The answer is that the tire is affected by other things besides the static pressure setting in the pits. Suspension settings of course, but those were constant on this day. What remains is a bunch of environmental stuff that darn sure better not make that much difference in handling, and ... me.

When I began the day, we warmed up the tires with three moderate speed laps. I was driving moderately both because I'd made a significant change from factory pressures and because we were recently fielded with a bunch of novices.

Then we began with readings that said we had to lower the pressures in the rear, so I was conservative again, and in fact I remained conservative until the changes began to converge on workable temp results. In particular, I found that the car was quite willing to accept front/rear pressures that were close together. That was reassuring as to nasty surprises waiting in the high-speed frontierland, but also meant the car could be adjusted to my preference in handling style. At the same time, the spacing of novices in my run group was spreading out, so I had the chance to run harder if need be.

The last big change in the Gary component was our finding how easily the heat was being put into the front tires and how little was going into the rears. That was the situation that led to almost identical pressures front and rear in the pit. At that point, I changed driving style from conservative and consistent for data collection to remedial driving. That is, I wanted to put heat in the rear tires to remedy the problem. A quick description of turns two, three, and four (which begins the esses): I came into two at higher speed than before, delayed braking longer, and went higher (it is on an upslope) before turning down to the chute leading to three. My turn-in was brisk and I got on the power hard and much earlier than the physical apex of the turn.

Going downhill to three, I accelerated hard out of second gear and into third for perhaps two seconds, something I hadn't bothered to do when driving for consistency to collect data. That put me into the downhill braking segment for three at something around 80 mph instead of 65-ish, and I held that acceleration until the last possible instant before turning into the off-camber surface of turn three as briskly as possible and catching the car in the slide that resulted. ("Accelerate into the turn until you see God. Then pray He left you room to brake." Old aphorism.) Then I used power to control the rotation and carry over to the uphill apex and the track-out leading to four. And so forth in a similar vein. It wasn't exactly "toss and catch" but moving in that direction for sure. Work those tires was my mantra.

Doing that on the out lap created a lot of heat in all four tires, but in particular it kept the rears from lagging behind the fronts. That became my new technique for data collection runs and we began adjusting pressures again. We didn't know where the cold pressures were going. We just worked with differentials relative to the previous run. Our mental calculations of temp/pressure curves were just meant to double check ourselves and avoid doing something stupid. (For example, creating that inversion with cold pressures, where the fronts were higher than the rears could have had startling consequences. Always best to check your work to avoid some of the surprises in life.) Since we didn't know it, it really is an objective datum that our empirical technique led us back to the revised pressure spec Porsche has for the GTS. (Well, for road work that is. Or two pounds less all round for track use.)

The significant difference is that we came back to that point from having explored the variations all around it, and we got back there only because the evidence of tire heating compelled a change in driving style to get all four tires working together. The exploration became an important part of the destination. That isn't unique in life, but we'll skip along to the important point for track days:

Quote:

The way you warm up a car's tires and the way you drive thereafter is a major influence on the grip it will be able to produce.

To confirm this, watch a grid full of high end drivers some day on their parade lap. All of them are everywhere on the track, nowhere close to the racing line, but if you study them you'll see that different styles are involved. Some are relying on repeated heavy braking, some do long sine waves down the track, and some like Fernando Alonzo throw the car sideways nearly hard enough to spin.

Somehow, you want to learn a technique that brings the tires up to working temperature, or fairly close, before you start driving "in anger" as we say. If you don't do that, you'll end up with the tires at one end or the other pushed beyond their working range. And sometimes the imbalance in workload can even let the tires at the other end begin to cool off. Or at least cool down below the working range for the compound. That absurd combination has led to many of the spins after cars have to circulate behind a pace car.

That's why we used to say: "We drive the car, but we race the tires."

Gary

 

No. And that definitely was one of the things we watched for. Also graining, when I began driving the out-lap so aggressively to put heat in the rears. Though God forbid we should do that to a set of new road tires this expensive.

Gary