Rev-mashing: part 2 of 2

Are you rev-matching or rev-mashing? Let’s repeat the question from the end of last week. Where in the corner should you downshift and what are the consequences of shifting at the wrong point.

Red Zone

Lots of people start to downshift during threshold braking. As soon as they hit the brakes, they also depress the clutch and blip the throttle. There are several problems with this.

  • The car was already at high RPM and the blip just sent it even higher.
  • The car is currently going too fast to engage the lower gear.
    • If you shift to the lower gear immediately, you may destroy the engine by revving it past redline.
    • If you wait with the clutch depressed, the engine and transmission will spin down, negating any benefit of blipping the throttle. When you do engage the clutch, you’ll have to feed it in gradually to prevent locking up the tires. Congrats, you just ruined your brake bias and made the stopping distance longer. You’re also putting wear on the clutch.

Blip-shifting in the red zone is the worst possible place to shift.

Orange Zone

This is generally the best place to downshift. For really long decreasing radius corners, it’s too early though, as you’re still aggressively bleeding speed through half the corner.

Yellow Zone

In this trail-braking zone, your concentration should be on controlling the speed and angle of the car using a combination of brake pedal and steering wheel. It’s not a good time to take your hand off the wheel or dance on pedals.

Green Zone

This is the point of maximum lateral g-force. Your foot is making the transition from brake to throttle. In longer corners with extended trail-braking zones, this is a fine time to shift.

Blue Zone

You’re balancing throttle and steering as you pass the apex and track out to the exit. The over-rotation you initiated in trail-braking has to be wound out some in here. Probably better to keep both hands on the wheel.

Purple Zone

Although it seems way too late, shifting in the Purple zone is an ok place to shift. You’re not going to break any lap records doing it this way, but you’re also not going to do any damage to the car. If you drive through a corner in a gear that’s too high, it’s not that big a deal. If you’re balancing throttle and steering, as you should be, you don’t really need full power anyway. I don’t know anyone who actually downshifts after a corner. It would be like shooting free-throws underhanded: works okay, but looks too silly to be taken seriously.

Demonstrations

Here’s a popular YouTube video instructing how to heel-toe shift. The video overlays footwork and RPMs. Watch as he starts the downshift too early. The revs drop and feeds out the clutch gradually. If you have to release the clutch gradually, you’re not rev-matching, you’re engine-braking (or engine-breaking). One wonders how such a flawed example can have so many views.

Here’s the right way to do it. Notice how quickly the clutch is released as he blips each gear. You might also notice how he changes his hand position depending on which gear he’s selecting. It’s most noticeable for 2nd gear where he rotates his hand thumb-down. He’s not doing this to look cool, but I’m guessing somewhere there’s a ricer in a stanced Honda back-handing every fucking gear…

Advertisements

Rev-thrashin: part 1 of 2

Check out this awesome 1980s movie poster.

I weep every time I see this poster, partly because I grew up in the 80s, a time of particularly embarrassing fashion. But now I also weep because I just learned that Pamela Gidley recently died. She’s probably most famous for “Twin Peaks: Fire Walk With Me” but I’ll always remember her as Cherry 2000. However, the real reason I’m bringing this poster poseur to your attention is that the wrist guards are on backwards. Wearing them like that would do more harm than good. Obviously nobody at the photo shoot had any idea about skateboarding. Why am I talking about skateboarding. Oh, I don’t know…

Whether you call it rev-matching, blip shifting, or heel-toe-shifting, this technique is often categorized as an advanced skill. I agree, but not because it’s difficult to do, but because it doesn’t have much effect on lap times. It only becomes really important when you’re searching for tenths of a second and not whole seconds. That said, driving enthusiasts love to display their rev-matching skill on the street or track. Like the Thrashin’ movie poster, sometimes they do it so wrong that it looks like wrist guards on backwards.

Let’s look at an idealized corner depicted as a rainbow of colors.

  • Red – Threshold braking zone. All brake no steering.
  • Orange – Initial trail-braking zone. Ideally, this zone is really small.
  • Yellow – Trail-braking zone. The car is rotating as the mixture of brake plus steering puts more weight/grip on the front of the car.
  • Green – Nadir. This is the point where the right foot is moving from brake to gas, and the point of minimum speed. Ideally, this zone is really small, but some corners do benefit from a little coasting.
  • Blue – Throttle zone. The position and angle of the car are controlled by a mixture of steering and throttle.
  • Violet – Straight. All throttle no steering.

At what point along this path should you downshift? Your perspective may be different from reality, so if you have some video of yourself driving, observe your shift points. Ask yourself “what are the consequences to the driveline and handling of the car?” Tune in next week for some discussion.

P.S. The image above is from my upcoming book, “In Slow Out Fast (and other lies of the race track)”. This long-awaited sequel (by me, not you)  to “You Suck at Racing: a crash course for the novice driver” focuses on the intermediate driver. Due out this Summer and available from Amazon.

Budget sim rig

If you’ve been reading my blog, you know I’m an rabid proponent of simulation training. It’s the cheapest and safest way to improve your car control muscle memory. In this post, I want to talk about building a low budget sim rig in mid 2018.

Software

There are two main contenders: iRacing and Assetto Corsa. The downside of iRacing is that it has a subscription and collection fee structure. The subscription part is a monthly fee. The collection part is that you have to buy cars and tracks to add them to your collection. 1 year of iRacing will cost you $200 or more depending on how many tracks and cars you purchase. What you get is laser scanned tracks, great racing, and a really helpful community. I highly recommend it.

Assetto Corsa costs about 10% of iRacing. Most cars and tracks are developed by the community and are generally free. You can also pay for official expansion packs. Assetto Corsa is a lot more tweakable than iRacing and has AI opponents, which iRacing does not. If you want to race as a competitive eSport, stay tuned for Assetto Corsa Competizione, due out this Summer.

In addition to iRacing and Assetto Corsa, there’s also a few others worth mentioning: Automobilista, Project CARS, Project CARS 2, rFactor, rFactor 2, and DiRT Rally. I have all of these, and each has its merits. Some tracks are only available in specific simulators, so if you want to train at that location, you have to get that software. On the downside, learning how to use each platform takes some time. Here’s my recommendation: get Assetto Corsa first. After that, it depends on your goals. If you’re interested in wheel-to-wheel racing, you should be practicing against other people to learn the habits of stupid people, not smart AI. You can do that with Assetto Corsa, but iRacing is better. If you’re driving off road, you can do that with Assetto Corsa too, but DiRT Rally is better. As you start to train for specific tracks, you may need to pick up Automobilista, rFactor 2, or even the venerable rFactor. And if you want to be blown away by how gorgeous a racing game can be, Project CARS 2 has the best eye candy.

Computer Hardware

Build or buy? Like building a car, building a computer takes time to research and some technical know-how to construct. If you’re not a computer builder, just buy the cheapest decent gaming rig on Amazon. Generally speaking, you can’t get anything decent for under $500. Sure, there are things marketed as gaming PCs for less than $500, but they don’t have a suitable GPU. The single most important thing in your computer is going to be the GPU. Most computers have sufficient CPU, RAM, and storage, but the GPU is highly variable. How do you check the performance of the GPU? Passmark. Assetto Corsa and iRacing don’t require a high end video card, and you can get away with a Passmark score of 1000. rFactor 2 and the Project CARS titles need higher end cards with a score of 2500 or more. My advice is get an nVidia GTX 1050. It has a Passmark score of 4500 and can be found cheaply whether you build or buy.

A quick perusal of Amazon shows you can buy an iBUYPOWER gaming PC for $499. But it’s got a GT710 video card with Passmark score 677. That’s no gaming PC. For the same price you can get a Shinobee with the recommended GTX1050. But who the heck are iBUYPOWER and Shinobee? Good question. You might prefer buying from a more established name like Lenovo, Dell, HP, etc. Expect to pay a little more for that, but maybe it’s worth it. In any case, you’re looking for a decent video card and probably a cost of at least $500.

If you want to build your computer, go to the Passmark best value table to find your GPU. Sometimes you can find screaming deals. I upgraded a GPU at one point for just $35. Right now, I would get a Radeon R9 350 (Passmark 2265) for $80. This is actually a card made for laptops that they shoe-horned into a PCIE slot. So it requires very little power and cooling. Looking at the local computer surplus supply, I can get a Dell with 8GB RAM for $150. I’d swap out the HDD for an SSD for $70. The end result would cost about $300 and the performance would be good enough. But for $200 more I get a lot of newer parts fully assembled, a warranty, the ability to return it to Amazon, and with twice the performance. So unless you actually like building computers, buy a budget gaming PC from Amazon.

Driving Hardware

The spectrum of hardware is immense if you consider the boutique companies making high-end products. But since this is a post on building a budget system, there are 3 major manufacturers to consider: Logitech, Thrustmaster, and Fanatec. I have not used Fanatec. I’ve steered clear of them for my own purchases because I’ve heard of too many horror stories with customer support. However, their hardware is really slick and it’s very popular amongst the top simmers. If you’re not building a budget system, look seriously at Fanatec. But for a budget rig, look elsewhere.

I own both Thrustmaster and Logitech products. I like them both. A sim rig really needs 3 things (1) force feedback steering wheel (2) all three pedals (3) brake pedal with progressive resistance. Although it aids in immersion, you don’t need an H-pattern shifter or hand brake.

Most of the recent Logitech and Thrustmaster wheels have good force feedback. Thrustmaster wheels have stronger motors. How important is that? Not very. What is important is the feel of the brake pedal. Most brake pedals have linear resistance due to using a single uniform spring. The more the brake pedal travels, the more brake is applied. Uh, that’s not how real brakes work. They work on pressure. The most authentic brakes push hydraulic fluid onto a load cell, just like in your car. They also don’t fit in the budget category. The inexpensive alternative is to use stacked springs, springs with varying rates, or rubber plugs. Mixing resistance types can give a brake pedal progressive resistance.

If you like tinkering, you can find older Logitech and Thrustmaster rigs on Craigslist or racing forums. Mod the brake pedal and you’re looking at some savings. You can also increase the precision of the sensor outputs with Bodnar cables. My rig is a mixture of Thrustmaster, Logitech, PerfectPedal, and Bodnar parts. The performance is probably up there with the boutique gear that costs $2000 or more. If your goal is sim training and not sim building, just get a Logitech G29. It has progressive brake resistance out of the box.

TL;DR

Expect to pay about $900 for a sim rig. If you build instead of buy, you could cut that in half, but at the cost of your time and labor.

  • $20 Assetto Corsa
  • $500 gaming PC
  • $90 1080P monitor
  • $260 Logitech G29

 

Race Report: Watkins Glen

TL;DR check and replace hoses often.

Watkins Glen is a kind of mecca for many racing enthusiasts. Whether you’re a spectator, driver, or sim racer, it seems that the Glen is really popular. I’m not really sure why that is. I’ve driven 10 tracks in the real world and about 50 in simulation, and it’s not in the top half of either list. I won’t make you endure the simulation list, but here’s the real track list.

  1. Thunderhill West
  2. Sonoma Raceway
  3. Buttonwillow Raceway Park
  4. Laguna Seca
  5. Thunderhill East
  6. Thompson Motor Speedway
  7. Watkins Glen
  8. New Hampshire Motor Speedway
  9. Carolina Motorsports Park
  10. Willow Springs

So what do I have against WGI? Two things: (1) it’s dangerous (2) it’s not very interesting. I like technical tracks that have compromises, off-camber corners, blind apexes, and decreasing radii. I don’t like long straights. In fairness, WGI does have some interesting sections. The inner and outer loops are pretty great, but in general, the corners are just too far apart. Don’t get me wrong, it’s still a fun track. And I’ll still add a couple of days to a work trip to go racing at The Glen. Which is how I ended up racing there this weekend.

My brother had just done a bunch of work on his Miata. He upgraded the brake, clutch, header, wheels, and had a fresh motor build controlled with a Megasquirt. He also made some trick custom bodywork with a fastback, splitter, wing, and side skirts.

IMG_7470

Sadly, the report from the practice day, which I missed, was that there was something wrong with the motor. It just wasn’t making any power. This despite paying for several hours of dyno tuning. We decided it would still be fun to race. For about 5 laps. The motor died on track. A radiator hose sprung a leak, causing the engine to overheat. After patching the hose, we got it started again, but the head was warped and mixing oil and water.

My brother came prepared with a spare head and gasket. So we set about the task of swapping heads. I say ‘we’ but it was really the other guys on the team. I would periodically fetch parts or food. Along the way, there were several unexpected adventures, like breaking the EGR pipe on the header and having to weld it back up. After some 8 hours, the car was all put back together and ready to race for day two.

The weather forecast was for rain, and it rained plenty. I went out first because I didn’t drive at all in practice. I spent most of the time watching the gauges and sniffing for trouble. Although you don’t use your nose much in racing, it’s a surprisingly useful diagnostic tool. I didn’t smell steam or oil, but there was a lot of unburned gas in the air. I didn’t have much confidence in the 15×9 225/45/15 RS4s in the rain. I tested the limits of the tire early and found they broke away very quickly. My job as first driver was to check the car out, not race for position, so I lapped at six tenths.

In my ~50 minutes on track, I saw a quite a few cars off track. Some of the slower drivers were really slow. But there were some really fast ones too. There were caution flags in some corner or other almost every lap. When I saw the course go to a full course caution, I brought it in as planned. In the pit stop we added 1 can of gas and a lot of oil. Apparently the mild oil leak was getting worse. Great, just great.

Our next driver spun the car on the first time through the inner loop. He decided the conditions were a bit too treacherous for him and brought it in a few laps later. The next driver’s stint was cut short by running out of gas. No, the fuel gauge doesn’t work, and our estimation of fuel usage was off (probably because it’s running very rich). He went back out again and was running good laps until he saw our friends’ car in the T6 graveyard.

And that’s where we decided to end the day. I suppose we could have raced more for the love of racing. But the car wasn’t competitive and had a decent chance of getting wrecked. The oil leak was also a concern. It’s much easier to drive a car up on a trailer than push it up if the motor croaked. I’m sure our competitors thanked us for the oil we weren’t leaving out there.

It was a pretty frustrating weekend for the whole team, but especially for my brother, who had nothing to show for all his time and money. After event fees, consumables, and travel expenses, it turns out I paid about $1000 for 50 minutes of yellow-flag-riddled track time. Sadly, this shit happens all the time in racing. Let’s be more specific about that last statement and take a look back at some racing history.

  • MR2: DNF, DNF, PX, PX, DNF, DNF = 2/6 finished
  • Miata 1: PX, PX, DNF, PX, PX, PX, P3, P2, PX, DNF = 8/10 finished, 2 podiums
  • Arrive-n-drive: P3, PX, DNF, PX, PX = 4/5 finished, 1 podium
  • Miata 2: PX, PX, DNF = 2/3 finished
  • Yaris: PX, PX, P3, PX = 4/4 finished, 1 podium

The actual cause of each DNF is listed below.

  • MR2 spun bearing – unknown cause
  • MR2 broken axle – inspection showed it was cracked
  • MR2 blown head gasket
  • MR2 spun bearing – clogged oil pump
  • Miata – broken suspension from off course
  • Miata – overheat from radiator hose leak
  • 240SX – blown head gasket
  • Miata – overheat from radiator hose leak

That’s only considering races. I’ve also had to go home early on HPDE days. Here’s the tally on that

  • MR2 spun bearing – unknown cause
  • 325E overheat – broken motor mount led to radiator leak
  • 325E overheat – leaky radiator hose
  • 325E blown head gasket

Certainly each platform has its specific weaknesses, but it’s surprising how often radiator hoses fail. Although we did get back out and finish, one of my arrive-n-drives involved a lot of down time from a leaky automatic transmission line.

There’s probably a bunch of mechanics out there shaking their collective heads at me. Duh, of course you need to inspect and replace hoses all the time. Sadly, I suck at car maintenance. The real problem here is that I have very little desire to become good at it. That’s written all over my racing history.

Revisited: The five stages of grief

Back when I first posted this, I was more concerned about oversteer as a problem. Well, that’s because YSAR started out as a crash analysis blog and oversteer, or rather the inability to handle oversteer, is the main cause of spins and a common culprit in car-to-car contact. I see oversteer a little differently now. Oversteer is necessary. The fastest way through a corner requires oversteer. You can’t open the throttle until the car is pointed down the track, and that requires rotating the car by swinging the back end around a little (not a lot, as drifting isn’t the fastest way around a track).

Oversteer scares passengers, understeer scares drivers

The problem with an oversteering car is that it’s more dangerous to drive. An advanced driver can handle a twitchy car and use the instability to rotate the car early in a corner. But novice and intermediate drivers will feel scared and intimidated. Countersteering isn’t second nature yet, so they will drive slower with less confidence and be more prone to accidents. In a team setting, tune the car for the slowest driver. Fast drivers can work around the handling problems of a car. It’s better for the fast drivers to be a few tenths off pace than the slow driver coming in on a wrecker.


If you hadn’t noticed, there’s some kind of numbers theme going on in this blog (two kinds of oversteer, three terrors, four temperaments). When will this end? Not this week. This time let’s riff on Dr. Kubler-Ross’ 5 stages of grief and call it the 5 stages of oversteer.

It doesn’t matter if you’re in a FWD, RWD, or 4WD vehicle, sometimes the rear of the car loses grip and the back end starts to swing out. Maybe you lifted the throttle, popped the clutch, put a tire off track, stomped on the throttle, ran over oil, or got nudged by another car. The reasons may or may not be within your control, but oversteer happens and you have to deal with it or you’ll spin.

  1. The first stage is denial. Typical thoughts as you spin off track might be “that did not just happen” or “that wasn’t my fault”. This stage generally doesn’t last long.
  2. Next is anger. You might yell “fucking track” or “stupid car” as you try to quickly assign blame. A repeat performance will leave you asking the racing gods “why does this keep happening to me?”
  3. As you continue to experience oversteer, you enter a bargaining phase where you start to put some blame on yourself and ask a lot of “if only…” questions such as “if only I had stickier tires” or “if only I had a thicker front ARB” or “if only I had a wing”.
  4. Depression is the fourth stage. It comes when you realize that all the tuning in the world won’t stop oversteer. You become dejected as you realize there’s nothing you can do about it. Your despair may make you consider quitting racing.
  5. The final stage is acceptance. You eventually realize that oversteer is a natural occurrence. A great drive is not measured by the absence of adversity, but by grace in its midst.

Coaching 86s

Normally, I coach for Hooked on Driving. I’ve also coached for a variety of other HPDE organizations, but not often. 2 weekends ago I did my first private coaching gig. I was asked by a driver who runs in the GT86 time trial series if I would be interested in doing some private coaching for some of the drivers in the series. I thought that sounded like fun and a good learning experience on both sides, so I agreed.

How much does one charge for such a thing? That’s a very good question! In my day job, I sometimes get honoraria for giving talks or reviewing grants. Those events might run $250-$1000 per day, and usually on the low side. If a company wants my services, I’d probably charge $250 per hour. But that’s my professional side. Performance driving is a hobby. So I decided that I would charge a hobby rate: $400 for the day for a group of 4.

So what does $100 get you? 2 sessions of me in the right seat, some chatting before and after each track session, and a group online telemetry analysis review a few days later (I brought my Aim Solo DL with me each session). Would I do this again? Sure, see my prices in the For Hire link at the top. You can also rent a car from me.

So let’s take a look at what one of the GT86 drivers learned. I’m particularly proud of what this driver was able to accomplish in one day.

The blue traces are a couple typical laps from his first session. The red traces are from his second session. I was in the car both times. Between those, he had one or two stints where he was working alone. The track is Thunderhill West. For some reason, Aim Solo thinks the start/finish line is on the straight between T2 and T3, so that’s where the left hand side of the graph begins.

The first thing to notice is that the red line is higher everywhere in the speed graph. He’s faster everywhere. Well except for 5900-6600 feet. We’ll get to that later. Let’s discuss the turns starting from the beginning (or rather T3).

  • T3 (500 ft) – There are 3 big changes here. The most obvious one is that his minimum corner speed is higher. Over the course of the day, he learned that he didn’t have to brake so much and therefore carry more speed through the entry. He also gained confidence in using the brake pedal to set speed rather than just scrub speed. This is why the shape of the curve is more U-shaped than V-shaped. He’s blending cornering and deceleration. Finally, the upward slope of the red line is higher, meaning he’s getting the throttle down more fully albeit slightly later.
  • T4 (1300 ft) – He brakes and accelerates at pretty much the same place, but he’s going much faster. This is mostly an improvement in confidence. An improvement in technique would also show him backing up the corner. Something to learn for next time.
  • T5 (1800 ft) –  The downward slope of the blue line shows that he used to brake for this corner. But now he just lifts off throttle a little.
  • T6 (2500-3200ft) – The red line starts to decelerate gently and then aggressively. He’s going very fast here and is a little worried about the upcoming corner. So unconsciously, he’s starting to lift before applying the brake. He could easily gain time simply by keeping his foot at 100% throttle. It’s fine with me that he doesn’t. If your self-preservation instincts kick in at the highest speed part of the track, I’m totally okay with it. The more important and impressive thing is that his minimum corner speed is 15 mph higher. Amazing!
  • T7 (4000 ft) – We didn’t really work on low speed corners, and you can see that there isn’t much difference in time from 4000-4500 ft.
  • T8 (4700 ft) – Confidence gets him accelerating over T8 rather than decelerating.
  • T9-T10 (5200-5900 ft) – Although we didn’t work on low speed corners, you can see that his minimum corner speed is higher. Not only that, but it’s happening later. He’s clearly trail-braking to make that happen. Now all he needs to do is move all that deceleration and rotation earlier in the corner. Sadly, it’s much easier to say than do, and he’ll be working on this skill for a long time.
  • T1 (6700-7400 ft) – The blue trace is actually higher than the red trace. Yes, he could go faster, but our attention was on how to drive T1 faster, not how to drag race down the main straight. Honestly, I would prefer if students never used 4th gear. Look what he does here. It’s smashing. He previously applied brakes, then accelerated, then braked again. The red trace shows him turning this complex of turns into a single turn.
  • T2 (8000 ft) – On the way into the carousel, he’s going so much faster that he bleeds a bit too much speed on the entry. Oh well, it’s an overall gain as shown by the time graph. Again, something to work on next time.

Wow, right? He gained 8-10 seconds over the course of the day. Partly it was increasing confidence, but there were changes in technique too. They go hand in hand. You can’t really trail-brake until you gain confidence. And once you gain confidence, it improves your trail-braking.

Race Report: Thunderhill West

One way to get a trophy is to be the fastest car on track. Another way is to be the slowest car. Haha no, ChampCar Endurance Series doesn’t actually give out trophies to the slowest car on track. We got the Sportsmanship trophy! I’m really proud of that because the team was full of first-time racers, and I didn’t want our inexperience to cause problems with the other racers. During the pre-race meeting I made a quick announcement along the lines of “We’ve got a bunch of first time racers here. Please don’t try to race us, we aren’t in your race. And please don’t try to scare us, we’re already scared”. We received nothing but encouragement all race long.

ChampCar uses a point system where each car has a base value and modifications with additional points. The Yaris was valued at 100 points plus 10 points per corner for lowering springs, so 140 total. Most of the field is near the target 500 points. The Yaris is really under powered for ChampCar. It was kind of depressing being the slowest car on track. But it’s a good car for a first race as it doesn’t have bad manners and is cheap as dirt to race.