3 years of YSAR

Did I really start YSAR 3 years ago? Well, this is post #156, so I guess that’s a fact. If you’re a YSAR visitor I want to say thank you for reading. It’s been fun sharing my thoughts with you. This week I want to reflect on a few related posts from the last year to create some kind of synthesis.

Last October, I wrote a pair of posts about the fastest line through a 90° corner. Does the line change if the car is a Miata vs. Corvette? Does it change if you’re on a race track vs. ice? I “solved” this problem using a simple-ish formula with circles of various radii. Turns out, if you’re driving a Miata, yes, the line matters a lot. Take a late apex most of the time, but make it a little earlier in faster corners. If you’re driving a Corvette, the line doesn’t matter that much. Driving on ice? The typical l apex is the worst line and early apex is the best. Huh? When you have no grip, and therefore no acceleration, you might as well keep your speed as long as possible.

In December, I wrote about the theoretical fastest path around a pair of cones. Should you drive a tight radius to minimize distance or large radius to maximize speed? Once again, I put some math to this problem and it shows that a tight radius wins every time. The only exception to this is if you’re not allowed to use your brakes.

These theoretical posts above were based on the typical tire model that states “there’s only so much traction, and you can use it for cornering, braking, or some mixture”. A little later I wrote about how the traction circle isn’t a circle (or a diamond, which is another valid way of drawing it). That is, grip can’t simply be divided into braking and cornering. There’s more traction available for cornering than braking, and mixing them a little is better than not mixing them. So trail-braking and early throttle application are techniques one should try to master because they maximize the use of your tires.

So the theory isn’t quite correct, but does the model hold up well enough to be useful? Is the shortest path around a pair of cones the fastest? I got to test this on a skid pad in August, and the answer is yes, the shortest path is fastest in the real world. If there is a long straight afterwards, then a late apex will win. When optimizing corners, it really comes down to a question of the straight that follows the corner. Is it worth throwing away some positional advantage to gain a speed advantage? Shortest path or largest radius?

So what’s my point in reviewing these posts? They were a great learning experience for me. I enjoy bouncing back and forth between theory and practice. When it comes to performance driving, the learning is as rewarding as the driving. As we start heading into the racing off-season, I urge you to read a book or two on racing. Brush up on the theory and see if you can put it into practice. For my homework, I just purchased Keith Code’s “Soft Science of Roadracing Motorcycles”.

Advertisements

The Ideal Student

As I was driving home from a particularly good coaching day at Thunderhill, I started thinking about what makes the ideal driving student.

  1. Safe. Safety is always the #1 priority doing something as dangerous as high speed driving. Safety doesn’t just mean being able to control the car. A safe driver has enough mental capacity to do other things while driving. Like watching out for other cars and listening to the coach. A safe driver is also prepared. Rushing creates errors both on and off the track. Arriving early to the exercises and grid is always appreciated.
  2. Humble. The best students don’t have inflated opinions of themselves. They want to improve and want me to help them. Those that already know all the answers, or display that attitude, rarely learn as much as those assume they know little.
  3. Enthusiastic. My favorite students are fun to work with. Positive and negative attitudes are infectious. When the student is having fun, I’m having fun. What’s more fun than seeing a student take pride in learning a new skill?
  4. Brave. I like students who overcome their fears. Bravery isn’t the absence of fear, it’s performing while scared (or at least anxious). Driving a car fast is dangerous, and that danger creates some fear. Sharing in a student’s triumph over fear is truly special. Every time.
  5. Inquisitive. I don’t simply mean asking questions. Anyone can ask “what’s your line through turn 3?” What I prefer is to hear comments that show they’re asking and answering their own questions. Ultimately, every driver has to teach themselves. If I hear “sorry, I messed up that corner, but I was trying to carry more momentum” I know they were thinking.
  6. Simmer. I can always tell when a student has had a lot of simulator experience. Their bodies are sometimes ahead of their minds. That is, the sim has trained some of their muscle memory but their brains haven’t yet connected the virtual and real worlds. It’s fun when the students are surprised by their own abilities.

So I guess that begs the question of the worst student… well, I wouldn’t want to be in the passenger seat with this guy.

https://vid.me/xlOb#2m30s

Data Analysis: Thompson

Last week I wrote about the ChumpCar race at Thompson. This week I want to take a look at the telemetry we recorded with an Aim Solo. As a reminder, this is what the track looks like.

The first thing I want to explore was the difference between my practice and the race. My fastest race lap was more than 2 seconds faster than my practice. Why was that?

The top panel is speed. The next two are G-forces. The bottom is time difference. I don’t tend to find the G-force graphs very useful.The black line is the “reference” or fast lap. The red line is the practice lap.

If you look at the speed graph, you can see that I’m pretty much faster everywhere. I’m braking deeper into the corners, and I sometimes have lower speeds in the middle of the corner (actually slightly before the apex). I knew when I was driving my fast lap that I had pushed too hard at point A and was understeering badly. This turned out to cost me about a tenth, so it wasn’t that bad. In order to get into the 1:26s, I’ll have to find more speed. So let’s look at the other drivers to see what I can learn from them.

I brake later and carry that speed much farther into the corner. By the time I’m on the gas, I’ve gained over a second on the green and red drivers. But I’m much later getting on the gas, and that advantage starts to decay. If there was a long straight following T1/T2, they would catch and pass me. However, on this track, my strategy would allow me to pass them going into T1 and block them on the way to T3. I brake later and keep more speed through T3.

Despite my more aggressive driving style, I’m no faster than the blue driver, and the red driver has mostly caught me by the time we fully exit T4. At some point I’m going to gap them by 2 seconds, but it isn’t through these slow corners. T5 has a funny data error; ignore that. I brake much, much later in T5 and carry the same exit speed. So I put some distance between myself and the red/green drivers. But the blue driver has a similar line to mine and we’re still neck-n-neck as we head toward the bowl.

In T6, you can see that I have the lowest mid-corner speed. I took a lot of different lines through here and this one wasn’t that good! By the time we are running down the straight between T6 and T7, I’m not much ahead of the red or blue drivers. What happens next is critical.

The red and green drivers brake much earlier. They have a somewhat constant rate of deceleration all the way to T8 and then pick up throttle from there. Without a throttle or brake trace it’s hard to say if they are coasting, but my bet is that after some initial braking, they are scrubbing speed by coasting. Coasting is generally not the fast way around the track, but it does have its uses. In contrast, the blue and black drivers have a roller-coaster speed trace. They brake much later on the way to T7 and speed up between T7 and T8.

The blue driver is losing nearly 2 seconds by overspeeding between T7 and T8. He ends up turning T8 into a braking zone. His speed never recovers and he ends up with the slowest speed down the main straight. This is a Miata, not an M3. The throttle is a suggestion to go faster. You have to keep as much speed as possible, especially in the fast corners.

Thinking back to last week, I was really excited about this series of corners. It’s a Type II followed shortly after by a Type I. That is, you brake way late for T7 to keep as much speed as possible. Then you throw away the exit to set up for T8 with just enough speed that you’re full throttle from just before the apex until the end of the main straight. That’s what I did, and it’s 2 seconds faster than the other drivers. That said, even if the others had taken my same line, they are unnecessarily lifting or braking to set up T9. That section is really about confidence more than technique. They’ll get faster as they get more laps.

3rd Gear Experiment

Late in the day we started losing the clutch. So Mario decided to do an entire stint without using 2nd gear. The results are pretty surprising.

The black line is his reference lap (fastest 2nd gear lap). There are two 2nd gear corners: T1 and T4. Once you hit T5, it’s 3rd and 4th all the way. So let’s look at the time and speed through T4. Amazingly, the loss is only 0.2-0.3 sec. In T1, he consciously holds more speed in 3rd gear. While he doesn’t get as much drive leaving the corner, he loses no time to shifting. The net effect is that he has lost no time by the time he hits T3. He does lose a couple tenths in T4 though. It’s a really tight corner, and 2nd gear puts the power down much better than 3rd. But it’s just a couple tenths. OK, so I’m driving in 3rd next year, at least in T1.

Race Report: Thompson

This weekend, I raced in a 12 hour ChumpCar race at Thompson Speedway Motorsports Park with my twin brother’s car on team “Occam’s Racer”.

With the exception of Thunderhill, every time I’ve driven a track, I’ve practiced it first in simulation. The Thompson Speedway road course isn’t in any simulation. So that changed how I approached the learning process. In simulation, my first sessions are driven very hard. My primary goal is to catalog the dangerous parts of the course. A violent wreck tends to stick in my mind. Obviously, that isn’t my plan in real life! What follows below is sort of a journal of how I prepared, practiced, and drove in the race.

Part 1: Pre-race Preparation

I’m watching about 7 videos. Some are races and others HPDE sessions. A few are really good and worth watching multiple times. I’ve got a track map that I’m marking up. The first priority is always to find the flag stations. For the most part, these are line-of-sight, low, and near the middle  of corners. The exceptions are T3, which is elevated, T5, which is on the backside of the turn after the bridge, and T10, which becomes visible on the inside as you approach the exit. Making a mental note of the flag stations will be the first thing I do when I go live. Here are my initial thoughts on the turns.

T1/T2 – Reminds me a bit of Summit Point. Starts very slow and then increases in radius as a single corner. 2nd gear shifts to 3rd in T2. Get the braking and turning done early and turn this into one long opening corner.

T3 – Still Summit-like with a slightly early apex 90° that goes a little off camber and blind at the exit. Reminds me a little of Willow Springs T1.

T4 – Tightest corner on track. 2nd gear 180° hairpin. HPDE line is late apex, but I think going around the inside will work better in a race (more defensive, keep speed longer, following straight isn’t very long).

T5 – Another 180° but slightly faster. Seems that one could go late apex, double apex, or tight. I feel like double apex is the default. There is some straight following, but not enough to prioritize exit speed over position.

T6 – An even larger 180°. A single late apex is expected as there is some straight afterward, but I suspect going around the inside is going to happen more often than not in a race. The track out gets close to the wall only at the very end.

T7/T8/T9 – The most interesting sequence of the track. T9 is critical because it sets up the main straight. If the car isn’t at the limit in T9, it means that T8 or T7 were messed up. The approach to T7 is to let the car create space as it passes the wall on the left. Trail-brake around and completely sacrifice the exit. The key is to get the car parallel with the curbing for the T8 entry. Straighten out T8 as much as possible and get on gas as early as possible. Start building speed before the T8 apex and keep building it. Track all the way out and stay out to set up T9. Diving too early into T9 will require lifting at the exit, which will kill the top speed. This whole section requires some discipline to get right. I’m really looking forward to this sequence.

T10 – Not really a turn, but the start of the main straight.

Part 2: Friday Practice

I just did 12 practice laps. I feel like that was enough to get a good idea of how to race it. In person, the track feels really different from what I saw on video. There’s more elevation than it would appear and there are several changes in camber. Also, the surface doesn’t have as much grip as I expected. My first 4 lap times were: 2:10 (out lap), 1:35, 1:32, 1:30, 1:29. It turned out that 1:30.09 was my median race lap, so it only took me 3-4 laps to get up to race speed.

Here’s my post-practice download.

T1-T2 – Brake between the 4 and 3 marker. There’s positive camber here, so grip is good. It turns downhill halfway through, so you really can’t hold trail-braking around or the car will over-rotate. Better to get shit done (braking, turning) early and then add throttle to stabilize the rear.

T3 – I didn’t do this corner correctly even once. It’s just a 90° left. Oddly, I sometimes thought it was T5, and went in thinking I was going to double apex it. I’m mad at myself for not finding the right reference points for this. On race day, I’ll have to make sure I focus on this corner.

T4 – It’s as tight as expected. It drops in elevation toward the end similar to T1, and it’s also important to get shit done early. There’s some throttle modulation required to maximize traction.

T5 – Indeed, this is a double apex. I saw a lot of cars throwing away time trying to enter too wide. The angle at the 2nd apex is critical to get right as the wrong angle could send you off track. I may try to go a little deeper tomorrow to get a better run into the 2nd apex.

T6 – A bit of a puzzle. It definitely felt faster going wide at the entry. But I also found I could put some distance on other cars going in tight. Maybe it doesn’t matter.

T7 – This is one of the best examples of a Type II corner I’ve ever experienced. I brake very late and totally sacrifice the exit. It’s a good passing zone as timid drivers will dawdle on the right side. Brake AFTER the ripples in the track.

T8 – I’m on the gas before the apex, and I feel that I should be able to commit to 100% throttle even sooner. Something to work on.

T9 – Message to self: 100% throttle the whole damn time. Commit to it.

Part 3: Mental Imagery

Mental Imagery (MI) is Ross Bentley’s thing. I generally don’t do that much (sorry Ross) because I find that simulation fits me better. But of course I don’t have Thompson Speedway in any sim. So time to do some MI. To get myself in the mood, I’m going to queue up some helmet cam footage I took during the practice session (helmet cameras are not allowed during the race but I didn’t hear anything about no cameras in the practice session…).

Part 4: Endurance Race

The race did not start well. On the opening lap, my brother Mario said “just let that guy by” as we watched an overzealous BMW e30 bullying his way through the first couple corners. About a minute later, our car comes into the pit dragging part of its diffuser. The driver, Patrick, had to dodge the BMW when it spun in front of him, and the subsequent trip into the grass yanked on the leading edge of the diffuser. Then the pit marshal came over and told us our transponder wasn’t working. So we had to make 2 minor fixes before completing even a single lap. On the plus side, the demeanor of the team became more relaxed as we had already lost the race. Now we could drive unencumbered with the thought of winning our class.

Patrick put down a lot of solid laps with a few times in the 1:29s. I got in next and started a string of 1:30s and a few 1:29s. I thought to myself “I think I’ve got a high 1:28 in me”, but it wasn’t until the end of the session that I did any. For most of the race, I was content to hook up with a car making a decent pace. At one point this was the Ztech Racing Honda Accord. Later it was an RX-7 that had the misfortune of smashing through a tire wall during the first hour. The corners where I made up the most time were the double apex T5 and the T7-8-9 combo. I often got enough speed out of the last section that I could stay in the draft of the higher HP cars down the main straight.

It took me a while to figure out how to optimize the 2nd gear corners and also the bowl (T6). But once I did, my lap times were dropping even though I was still driving conservatively. I got a couple 1:28s hit traffic, and then did more. After one really good T1 I decided to go for a flyer and see what I could do. Although I scrubbed too much speed understeering through the last corner and had no cars to draft, I produced a 1:27.29. I’m sure I could do a high 1:26 given another session, but I was running low on fuel and wanted to do some short-shifting laps to see what I could do without ever using 2nd gear. 1:29.59. I should have done that experiment earlier in the day because I probably would have driven it that way instead. Yes, it’s a little slower, but it’s easier on me and the car not to shift so much.

When we refueled the car, it only took 10 gallons. So despite the fuel gauge reading empty, there was still a few gallons left. In a 12 hour race, ideally we would run 6 two-hour stints. So only 5 pit stops. But due to the fuel gauge and cockpit heat (no cool suit), we ended up doing 8 pit stops. With the 5 minute pit stop rule, we lost 15 minutes or 10 laps. And the problems at the start of the race cost us another 4 laps. Ah, to have those 14 laps… Alas, we ended up in 11th place (9th if you discount the EC class cars that aren’t playing by the same rules).

Part 5: Post-race Thoughts

Car: The car was great. There was 1 Miata that was faster and 4 that were slower. The Occam’s Racer Miata was a Spec Miata in a former life and is still carrying at least 100 lbs of extra weight. So there’s some free performance improvements expected down the line. We ran Dunlop ZIIs in 205/50/15 at 40 psi hot. That’s maybe too high for such a twisty track. ZIIs are also not the fastest of the 200 treadwear tires. Not sure if we could get RE-71Rs to last 12 hours though. If we wanted to spend money, we could definitely get more power out of the motor, which is currently just a junkyard replacement. With Occam’s Racer not due on track again until Spring 2018, there’s some time to sort things out.

Track: Thompson Speedway is a really great track. With the exception of the main straight, there is very little breathing room between corners. That makes passing more difficult and more fun. The various small changes in elevation and camber are a challenge to optimize. I can’t wait to return. My small complaint about the track is that it’s a little bit one-note. Long corner, short straight, repeat. There aren’t many puzzles to solve. So the thinker in me was happy when there was traffic to work around. The event drew 38 cars, which is enough to make things interesting, but I would have preferred 50-60.

Racecast: We need to sort out the Racecast camera. We were supposed to be streaming live the whole day but I don’t think it worked at all. Unfortunately, no race footage from our car.

Event: The ChumpCar staff did a great job organizing the event. I wish they had run a few more warmup laps. They didn’t notice our transponder wasn’t pinging until after the race was underway. The Friday practice was not well organized. From the long line-up outside the gate to the practice session that started over 90 minutes late, it was a shit-show most of the day. Fortunately, they have good onsite food.

My Driving: For the most part, I was happy with my driving. I was particularly pleased that I was up to race pace on my 4th practice lap. I did my homework preparing ahead of time and it showed. In the race, I didn’t make contact with any cars, go 2 wheels (or more) off, or lose control of the vehicle. I raced within my abilities and managed to set fast time on the team by 2 seconds. Time to put another track decal on my windshield.

Disaster training

When we talk about driver education, the conversation often goes to line, brake release, etc. Even though we say safety is the #1 priority, how often do we talk about it? How often do we demonstrate it? Safety doesn’t just mean having the proper gear in the car or on your person, but the training necessary to deal with bad stuff that happens on a race track. For example, has anyone ever shown you the proper way to use a fire extinguisher? What driving organization teaches people how to put out a car fire? Is it even a good idea to teach that or does that increase your chance of insurance liability? What about first aid? The Motorsports Academy Level 1 course I recently took had zero content on disaster response. When novice drivers go to HPDE events, surely they expect their coaches to know what to do in a crisis. Like any performance activity, we don’t rise to the occasion, we fall back on our training. For many of us, that disaster training is shamefully inadequate.

In the video below, the car catches on fire due to an exhaust leak and dry grass. The car is equipped with a fire extinguisher, but it hardly matters because it isn’t used properly. Also, the car wasn’t shut off, so the fuel pump keeps going. This isn’t a race, so full safety gear was not required. This kind of thing could happen at any track day. Are you prepared? Honestly, I’m not, and this is a wake-up call for me to do some homework on the topic.

FWD Drifting: Part 2 and Cones in Practice

Last week I talked about some of the tuning and techniques for drifting FWD cars. Some readers may be asking “why bother?” Well, because it’s a driving skill. And if you can drift a FWD car, it will help you drift a RWD car. Inducing oversteer by dynamically changing the balance of the car is important regardless of which wheels are providing power. I shot the video below on the skid pad at Thunderhill between coaching sessions.

The first part shows an exterior view of some switchbacks. It’s sort of comical how slow I’m going and how little my car looks like a racecar. But even at slow speeds it will slide around corners. In the second part, the camera is inside the car. You can see that I don’t use the hand brake. The car oversteers by changing the balance of the car, not locking the rear wheels. It’s also set up with a lot less grip in the ear. The car has RE-71R tires on front at 26 PSI (cold) and Hankook runflats on the rear at 38 PSI (cold).

The next series of shots are what I’m calling point to point. It’s just going around two cones but with different turn radii. I start with a large radius and progressively shorten it. Which one do you think takes the least time? Back in December, I posted on this topic. See Cones in Theory. If you don’t want to read that whole post, here’s the short version: I make the statement that path A takes less time than B, C, or D. That’s the experiment I’m performing in the point to point videos above.

I timed the various runs and indeed, the tiny radius is the fastest (path A). It’s also in a very bad spot in the power band. I’m driving in 2nd gear the whole time, and there just isn’t as much power when driving the tighter radii. But it didn’t change the outcome. Path A is the fastest way around a brace of cones.

FWD Drifting: Part 1

There are a lot of people out there who think FWD is not fast and not fun. Well, as an owner of both FWD and RWD track cars, I can say that I have just as much fun in a FWD Yaris as a RWD Miata. One of the big reasons to hate on FWD is that you can’t drift it. Drifting is cool. Delicately balancing a car as it swings around a corner looks and feels awesome. It’s not always the fastest way around a track, but it definitely looks the coolest. I’m not a drifter and I don’t really understand drifting competitions, but I sure do understand the fun. In the next couple posts I’m going to share my thoughts on how to drift FWD in theory and practice.

First off, drifting is simply oversteer. That term simply means that the front wheels have more grip than the rears. A better way to think about it is that the rear tires have little grip. There are several ways to create oversteer. Before getting to that, let’s talk about tuning. Most cars come from the factory with understeer built into the design. After all, it’s safer if cars don’t spin. To get your car to oversteer, you may need to do one or more of the following.

  • Turn off traction and stability control if you have it. This nanny can totally kill all efforts to drift.
  • Use very lopsided tire pressures. Fronts should be a little lower than usual and rears should be near maximum.
  • Use different tires front and rear. Fronts can be a typical 200 treadwear sport tire while the rears should be hard and skinny all seasons.
  • Tighten up the rear suspension with higher rate springs or/and anti-roll bar.
  • Increase the toe in the rear. Normally, cars are aligned to be toe-in (negative) in the rear for safety, but toe-out (positive) helps oversteer.

So now that you car can oversteer, how are you going to go about it? In powerful RWD cars you can mash the throttle and the tires will spin too fast to make good grip with the pavement. Hey, anyone can stomp on a pedal, that’s cheating. How else can you create oversteer?

  • Lift – The simplest way to lose rear grip is to quickly lift off the throttle. This deceleration will shift the weight of the car forward on to the front tires. Since FWD cars naturally have more weight on the front, simply lifting off the throttle can induce oversteer. Try going around a circle at the limit of traction and then lift. You will oversteer.
  • Brake – You can very quickly transfer weight forward by braking. Really, there isn’t much difference between lifting and braking in theory. In practice, they can feel very different because the brakes act on all 4 wheels while lifting works only on the drive wheels.
  • Scandinavian Flick – If you want big drifts, lifting and braking may not be aggressive enough. Not only do you want to transfer weight forward, but also sideways. To do this, you steer in the opposite direction to load the weight on what will become the inside of the turn. Then turn as normal and the shift of the weight from inside to outside will help break traction.
  • Decrease radius – The speed you go through a turn is determined by the turn radius. Larger radii have higher speeds. So if you turn the wheel tighter it has the effect of slowing you down. This transfers weight forward, which helps oversteer. But if you’re scubbing the front tires because you turned the wheel too much at the start, turning it more won’t do anything.
  • Hand brake – Grabbing the hand brake shifts the weight forward and only applies rear brakes. It’s almost like it was made for drifting. That said, I generally don’t do it. It seems like cheating to me. Also, you could flat spot tires by locking them up. You can get all the oversteer you want without the hand brake.
  • Clutch pop – Another technique I don’t do. Seems like it puts unnecessary wear on the car. On the other hand, try coasting around a corner and then pop the clutch. It can spin you (more so in RWD). If nothing else, this is a reminder not to pop the clutch.

Once the back end of the car is coming around and you’re pointed farther into the corner than necessary (that’s oversteer), what next? First, you have to do some counter-steering to prevent the car from spinning. Frankly, this takes a lot of practice to know exactly how much. Second, don’t hit the throttle. This will transfer weight to the rear wheels and kill the drift. This is where FWD and RWD are really different. You set up the oversteer the same way: by moving weight to the front and side. However, throttle improves RWD oversteer and kills it in FWD. So you have to be patient and wait until the oversteer is mostly over before adding throttle.

Here’s my thought process as I grab some FWD drift.

  1. Drive in an arc to keep the suspension loaded on one side
  2. Turn the opposite direction in a decreasing radius
  3. Snap off the throttle
  4. Drift initiated
  5. Countersteer
  6. Wait for it
  7. Keep waiting
  8. Patience
  9. Throttle it out and recover

Tune in next week for a demonstration…