Project update for May 10, 2011: So, back to autocross performance of this 2011 GT. To say we have been struggling with the car in STX this season is an understatement: We've been getting beaten badly at both Regional and National SCCA events. The car started with off-the-shelf AST coilovers, our camber plates, 18x9" wheels, a 265mm street tire, and we just started autocrossing it. We've made a lot of updates since then but the results haven't been improving much. Sure, its faster, just not remarkably so. From day one we've been fighting to keep the rear tire spin under control - it was always a real motherf*cker to drive in a parking lot. As in - harder than anything I've autocrossed in the past decade. Rear traction under acceleration always severely limited every aspect of an autocross run. We also fought some other issues, but it always drove like it had 600 whp and 225mm mud and snow tires. It didn't matter how smooth you were on the throttle - it was always in a constant state of power oversteer, but we just came to accept that.
Costas, Amy and I got used to dancing that sideways dance, tip-toeing on the throttle at every corner exit, for months. On a road course (where you are in higher gears, and usually either WOT or almost completely
off throttle most of the time) it was much easier to deal with, but in an autocross situation - where we are using part throttle
so much more often than full throttle - we're slipping and sliding around like a greased pig. Also, on wet road courses it was also VERY difficult to control the back end of the car, and I had lots of oversteer issues in slow corners at both MSR-H and HHR in the wet.
April 29, 2011 - Riverside Annex Test-N-Tune
Come to find out, we needed another set of drivers that
hadn't been driving it since day one to give us some much needed perspective. Another set of eyes/hands/feet to see the issues clearly. We got that and more at
this private test day on April 29th. Todd F set up this event with AST-USA and we got the invite, as did 5 autocross cars. This was an intense day of testing, on concrete, with AST's shock valving trailer on site to make valving changes to test and verify on the spot. Couldn't have asked for a more ideal situation, and we're thankful Todd and AST let us join them. We brought lots of tools: DL1 data logger, Chase cam video aimed at the front tire, timers, durometer, pyrometer, and more. This degree of testing on this car was long overdue, and much information was gained - I'm still sorting through the data. I'll give a highlight of what we learned in this post and
follow up with more data and video as time permits.
The Course
The site was the Riverside Annex, and the area was the 4-10 split, and the surface is concrete. The course laid out was typical for our Test-N-Tune events. It was 30 seconds in length, with a constant radius sweeper on one end, a decreasing radius on the other, with one long side of slaloms (60' spacing) and big offsets on the other side's length. The Mustang was only at WOT for maybe 2-3 seconds per run, so it was not a power course. To simplify testing it was run as a continuous circuit, with a start/finish line where we could mark consecutive laps, and each driver usually made 3-4 laps at a time with a running start. This minimized starting/launching variables, minimized driver learning curves, and made for quicker testing. So the driver would make a half lap before the start, then lap/lap/lap, then come in to look at times, adjust pressures, check data, and we'd shag any cones knocked down.
We used to hold 2-3 of these private Vorshlag autocross test-n-tune events for our own cars and VTPP testers every year, but haven't held one in 2 years, and it has showed. Our DSP run last year in the E46 was a train wreck, as was our untested E30 autocross set-up at the 2010 GRM challenge. Its hard to learn enough in a 4 run autocrosses to make speedy set-up progress or test things back to back, but when you can make 70 runs in a day (like we typically do at our test-n-tune events), you can learn and test a lot of variables much more rapidly. We had about 60 runs logged in the Mustang at this event, and we learned an enormous amount in only 6 hours.
So we got down to the site at around 9:30 am on that Friday, quickly unleaded the trailer, and helped set-up then mark the course - half of which was already up when we got there. We had 3 sets of wheels and 4 set of tires we had on hand to test (we ran out of time and didn't get the Hankook RS3s mounted or tested that day), plus we wanted to try some new shock settings and maybe even new base valving. As anyone running in Street Touring knows, there are a LOT of tires in the 140-200 treadwear range to test, and for this particular class & car there are at least 7 different potentially competitive tires we could choose from. We have very limited outside test data to rely upon for this somewhat unique STX car (heavier and more powerful than anything else run in ST so far) and max tire size (265mm), as we've been running in other classes (on Hoosier Rs) or on 245mm tires in STU for other cars in the past few years. Also, the ST tire tests done in GRM the past few years have been on smaller and lighter cars (Miatas and Civics), and the tires that are fast on those don't always "Scale up" to the bigger and more powerful cars like our 3600 pound mustang with 400 whp.
Baseline Set-up Testing
Before we could start putting the various tire brands to the test, we needed to play with a few things on the baseline set-up, and make the big shock adjustments. Before that, we had to get the driver familiar with the course, the site, and the existing set-up. A quick note about the existing setup: It has been a constant blur of changes, with never more than a few runs without some variables changed radically. That's pretty much a nightmare. We have never been to an autocross event where at least 3 or more major things had not been changed. Costas would jump in the car to make his run and if we were lucky, Amy and I might get a few miles around the block to test the latest round of changes. At the last event Costas drove the car at he had to quickly learn to drive with these changes: new Tires/Pads/Wheels/Front Swaybar/ECUmap/Coldair/etc. Whew! We were all very much looking forward to getting more acquainted with the car with a lot of runs today.
We came with 2 drivers, myself and Costas, but I had just run a
LeMons endurance race the weekend before where I think I fractured a rib (3 weeks later it is still very painful). Long story short after one half of a lap on the 30 second autocross test course I was doubled over in pain and out of the car for the day. The race seat was digging into the injured rib on left turns, so this meant that one more variable would be removed from our testing - the driver.
Costas took over as the main test driver for the day and I was on stop watch/tire pressure/durometer/camera duty. This would actually speed up our testing, as one driver could minimize the driver learning curve more quickly than two, and he's already shown to be super consistent, and we've typically been no more than .1 sec apart at most events.
So Costas ran half a dozen warm up laps to familiarize himself with the car and course, and his times quickly stabilized to within .1 sec between most runs. We then started "Baseline testing" the shock settings and tire pressures on the Yokohama AD08 tires, which are a very known quantity for us having run them on another ST car in 2009 and at most events this year on the Mustang. We also had Brian Hanchey and Stuart M of AST-USA on hand to closely watch the car as Costas got acclimated to the course and the Yokohamas, and Brian noticed something weird going on straight away. He made some shock adjustments and the car was a tick quicker, and we made some tire pressure adjustments (trying some radically new pressures) and found a few tenths more, but it was still the usual handful to drive.
Hanchey then hopped in and took a few laps, to see how it felt first hand (invaluable), fought it the whole way, then hopped out and said "How in the F*** are you
DRIVING it like this?!!" We then asked Stuart to make some laps and he felt the same way, and has a 2006 GT that he daily drives and tracks as well. Later that day, former F Stock champion Casey Weiss took a few runs in it also; he's been running in S197 cars the past 3 years in F Stock. Same observation - the throttle is
WAY too sensitive, car is too hard to drive. Hanchey has two decades of experience, as does Casey and Stuart, and they were all 2 seconds off the pace even after a few laps. That's not a knock on their driving - no, its showing that this mess of a set-up is that hard to get used to. I've co-driven with Weiss and Hanchey many times over the past 20+ years, and they can usually hop into any car and go fast as Hell within 1-2 laps. But when proven, fast drivers like these hop in the car, and it still takes a significant learning curve. Luckily we've already been "learning" this for months. Given another dozen laps they would have been right on pace. Even after 5+ autocross events in the car and dozens of laps, its still
difficult. Costas was literally making throttle changes with his toes thru thin racing shoes - not the typical ankle movements - its that sensitive.
Left: Costas during one of our many tire changes for the day. Right: Hanchey working with some F Stock racers on ASTs in a 2007 Shelby
They all felt that because of the very abrupt power application the car was nearly undrivable, and certainly holding us back. Hanchey was convinced the car had serious throttle tip-in/mapping errors. We've been fighting it all year, just thought it was supposed to be this hard with 400 whp through 265mm street tires. From an outside perspective - it was obvious. This was a 30 second test course run entirely in 2nd gear, nowhere near redline but always in the meat of the powerband on every corner or transition exit. Hanchey said to "try a run in 3rd gear" to prove his point. I told him that was crazy, but we humored him. Costas made 3 laps in third gear. It was ~2000-2500 rpm lower than in 2nd gear at every spot on course. He came in and said "No way, that felt SLOW",
but it was .3 sec faster - best run of the day so far. WTF?! Yes, there is a problem with power application and this proved it.
Now it wasn't faster on every lap. He took 3 laps like that to get used to it, and the first lap was slower, the second lap was even, and the last was a bit quicker then all previous laps. It was the change in power application, but also a *large* change in lines he took through the cones. Very different line could be utilized with the less sensitive throttle. More of a "momentum" line than a point and shoot line. But in the end, it was faster cutting back on the power by running the higher gear through the course.
Why is that? Let's explain - modern fly-by-wire throttle systems have the ability to have the throttle response (correlation between % pedal opening and % throttle body opening) tweaked, by both the OEMs and aftermarket tuners. The "trick" is to make the throttle mapping much more "falling rate" or digressive, which make the cars "feel faster". This means - the first small percentage of pedal opening makes for a LARGE percentage of throttle opening. So when you barely give the car a little gas pedal, its opening the throttle very aggressively, and it
FEELS FAST. Which makes it very difficult to delicately adjust throttle, which is all you do in an autocross with a powerful car. That's been the case all along with this car, both with the stock tune and the aftermarket tune we added via the SCT Tuner. This is actually the 2nd throttle mapping we've tested and its still MUCH too aggressive.
Throttle Mapping Issue = Found!
What we actually need, to make the car easier to control at lower speed events like autocrosses, is
a rising rate throttle map as described and charted on page 36 of Neil Robert's excellent
"Think FAST" book. And I quote,
"A rising rate throttle... provides fine control over the lower power end of the throttle range. That helps you blend cornering into forward acceleration smoothly and early". An example of this from the 1960s was the uber-powerful and light Can Am cars, another quote, "Jim Hall said that a rising rate throttle linkage was an absolute must on the big block Chevy (1000+ hp) Chevy-engined Chaparral Can Am cars to make them anywhere close to being drivable." We don't have a 1000 hp big block underhood but we
are trying to take 400 whp V8 and power it through somewhat narrow street tires, and do so in a parking lot, in lower gears, with abrupt power transitions.
So the throttle mapping is
a big thing that needs to be changed, and throughout the day I kept asking Costas to make a run or two in 3rd gear, every hour or so. And every single time, on every set of tires, and every permutation of shocks,
the car was as fast or faster in 3rd, a higher than optimal gear, because it was so much more drivable. Matching those times in 2nd gear took extreme control and wasn't nearly as repeatable. The shop that is providing our custom SCT tunes for the Mustang was called and we described the issue in great detail; we were already on a supposedly less aggressive throttle map, but now a third custom tune was created, and Paul M and I just loaded it to the Mustang this evening. We'll be testing this in anger this coming weekend at a SW Texas Divisional series 2-day event.
The next big revelation was in shock revalving. We had already had AST-USA revalve the fronts but the rears were still pretty much "off the shelf". Hanchey had a suggestion, we asked them to try the valving change he thought would help, they revalved the shocks on the spot, and it made a huge difference in times right off. It has us thinking about some even more radical shock ideas to test in the near future- which, for business reasons, I'm going to test before we explain these plans.
After this significant jump in times was from the shocks verified on several more laps with the Yokohamas we moved on to the Toyo R1Rs, then the Dunlops, each time re-verifying the shock settings front and back as well as the tire pressures that worked best on the Yokes. So which tire won? Well, it was pretty close between two of them, and I don't want to spill everything just yet (need time to crunch numbers and edit video), but
the Toyos are for sale...
Fresh set of 265/35/18 Toyo R1Rs ..... now FOR SALE!
These R1R tires are still
full of tread and have less than 20 total autocross runs on them, all in the last 6 weeks. Still very fresh, great durometer numbers, so how about $600 + shipping? Probably make great track tires for somebody (and I will keep them for just that if I only get low-ball offers). They won't be going back on this Mustang for STX use, that's for sure. We do really like the Dunlops, even these 3 year old former street tires we tested with. I bet you will see a fresh set on our car soon, for testing head to head with the Hankooks and Yokohamas, and possibly in some unusual sizes. We learned some things that day that we want to verify with even more testing.
We didn't have time to run into town and have the Hankooks mounted - the 265/35/18 Hankook RS3. So we've already mounted those and they will be run this weekend at the SW Divisional here at TMS. But as the day was winding down we did re-mount the first set of ours - our control set, the Yokohamas. We noticed only a slight uptick against the original times, so the temperature creep didn't account for more than .1 sec of our overall gains, from beginning of day to end.
OK, so what did we learn? How much time did we gain, if any? From the beginning of the day, starting with shock and tire pressure settings we'd refined so far this season, to the end of the day, we found 1.1 seconds on a 30 second course. That's.... a lot. This was between testing some unusual tire pressures, the various tire testing, and most of all the shock changes (valving changes uncovered almost a second alone - you can't tell me "shocks don't matter much"). And this is still with 2nd gear power application
almost impossible, and the car quicker in 3rd gear. Will this latest round of improvements + a new throttle map translate to a more competitive car on a full SCCA sized course? We hope so.
THere's a couple of stills from the "tire cam" videos (that I still need to edit/upload) showing sidewall deflection; the Toyo was compared to the Dunlop & Yokohama. Costas also managed to replicate that unusual "limp mode"
error icon on the dash that I ran into event last month. It happened 3 times during the day, but he wasn't allowing any more wheelspin than normal, just the same massive amount we always see in 2nd gear. He'd notice an abrupt lack of power, abort the run and come in to show the icon; I snapped a picture of it. To clear the error you have to turn the key off and restart the car. Odd.
More soon,