DRIVEN TO SLOW DOWN - TOYOTA GT86
Embedded Files
Before we begin tweaking the car, we first have to understand how it came into being. It began from a Toyota board meeting in 2007 where concerned directors discussed the slowing growth of the automotive industry worldwide. They decided an affordable sports car was needed to introduce driving pleasure to a new generation, and assigned Tetsuya Tada to lead as chief engineer.
After initial research on grassroots enthusiasts, he realised the the industry was trapped in an arms race of quantitative variables on performance such as grip levels and zero to 60 times. However, the improved performance on paper did not neccesarily translate into a vehicle that was fun to drive.
“For the 86 we didn’t set up any numerical targets like lap times or acceleration, We had one test driver, & after each set of tests, the only thing we’d ask was, ‘did you enjoy it’? "
Yoshinori Sasaki
Assistant Chief Engineer
The OEM+ Approach:
Starting from where Tetsuya and Yoshinori ended, we have to consider that the GT86 / Scion Frs / Subaru BRZ is a car that has been made for a wide variety of drivers and road conditions. Correctly deducing areas of compromise they had to make will be one of the keys to unlocking exactly how to make the car feel better.
Please note that if you are only looking to make the car faster, this may not be a good guide for you. Achieving maximum fun is what we are going for! Let's dive in.
Contents:
Weight Reduction
Weight Reduction
The law of diminishing returns means that at a certain point, the dollar spent becomes exponentially more compared to the weight lost. Take the amount you spend ($) and divide it by the weight (kg or lbs). This will give you a good idea of how best to address weight reduction. Here is the definitive forum thread on the subject.
“Simplify and add lightness. "
Colin Chapman
Founder, Lotus
This guide focuses on parts that bolt right in and out and is ranked in increasing order of cost/kg lost. FT86 forum user Anthony has brought this approach to near conclusion by stripping an incredible 181kg out of his car. While this might work for him, the car that we will be developing is designed for maximum driving pleasure on normal roads.
This means that the car still needs to be usable for a daily commute, with some basic comforts thrown in. The list below is by no means exhaustive (two links to the bottom that are more comprehensive), and decisions were made looking at the money spent relative to the weight lost.
The system I use to calculate the price to benefit of weight loss in a car is based on the following criteria:
1) Free - Usually free (eg. spare tyre removal).
2) Cheap - Simple bolt in parts (eg. lightweight battery).
3) Expensive - Complex part replacement or removal (eg. carbon roof)
For category 2 and 3, you can make a better decision if you list the parts and the relative performance and or weight loss. If you are trying to shave seconds off a lap-time, the cost goes up exponentially, so consider value (price to performance) if it is a road car.
The list below is by no means exhaustive (two links to the bottom that are more comprehensive), and decisions were made looking at the money spent relative to the weight lost.
Starting Weight: 1250kg
Note: all weight figures represent net weight loss : (original weight minus new weight)
Free Weight Loss:
19.0 kg - Front Passenger Seat
14.5 kg - Spare Tire & Tools
3.0 kg - Manuals and Carpet
2.0 kg - Window Washer + Fluid
1.9 kg - Head Unit
1.1 kg - 1xRear Seatbelt Delete
0.9 kg - Sunshades
0.8 kg - Sound Generator Delete
Cheap Weight Loss:
11.8 kg - Shorai Lithium Battery
8.8 kg AP Essex Front Brake Kit
6.0 kg - Recaro SR6 Driver Seat
3.8 kg - TRD Flywheel
3.1 kg - JDL UEL Headers
2.7 kg - HKS Spec L Carbon Tip
2.2 kg - Beatrush Front Bumper Brace
1.6 kg - Momo Mod 78 330mm Steering Wheel
Total Weight Loss: 83.2kg
The expensive weight loss will lie in the carbon roof, lexan rear window, carbon hood and trunk, and carbon doors which will net another 57.5kg of weight reduction. However, given the high cost of these items and safety compromises, I have elected to stop the weight reduction at 7% (1167kg) from the original weight (1250kg).
: : : : : : : : : :
References:
General weight reduction:
http://www.ft86club.com/forums/showthread.php?t=7881
Race weight reduction:
Final Drive Gear Ratio
Final Drive Gear Ratio
One way that manufacturers achieve fuel efficency is by gearing the cars longer to provide low rpm at highway speeds for better economy. A plausible explanation for the stock 4.1 ratio. Theoretical top speed at 7400 rpm is 270km/h or 168mph, which is unachievable with the stock engine and gearing.
Reducing the final drive ratio increases the torque multiplication to the road for every gear. The cost to this is top speed in each gear. Given that this car was never about the numbers, sacrificing 0-100 km/h (0-60mph times) fits in our objective on improving driving feeling.
The two main options for manual cars are to move to the 4.56 or 4.8 ratios. In keeping with trying not to deviate too far from the original choice by Toyota, the 5.1 ratio will not be covered.
One way that manufacturers achieve fuel efficency is by gearing the cars longer to provide low rpm at highway speeds for better economy. A plausible explanation for the stock 4.1 ratio. Theoretical top speed at 7400 rpm is 270km/h or 168mph, which is unachievable with the stock engine and gearing.
Reducing the final drive ratio increases the torque multiplication to the road for every gear. The cost to this is top speed in each gear. Given that this car was never about the numbers, sacrificing 0-100 km/h (0-60mph times) fits in our objective on improving driving feeling.
The two main options for manual cars are to move to the 4.56 or 4.8 ratios. In keeping with trying not to deviate too far from the original choice by Toyota, the 5.1 ratio will not be covered.
Stock 4.1 Final Drive:
2nd Gear Top Speed: 61 MPH
Crusing @ 60 mph: 2600 RPM
Top Speed: : 174 MPH
Cusco 4.56 Final Drive:
2nd Gear Top Speed: 55 MPH
Crusing @ 60 mph: 2900 RPM
Top Speed: : 157 MPH
MFactory 4.8 Final Drive:
2nd Gear Top Speed: 52 MPH
Crusing @ 60 mph: 3100 RPM
Top Speed: : 149 MPH
Two videos can illustrate the difference in shift points on Tskuba track in Japan. This is a stock car with the 4.1 ratio, and here is the TRD Griffon with a 4.8 ratio. This is not a great comparison since the latter is much lighter, however the drivetrains in both cars are exactly the same.
O.S. Giken 1.5 Way LSD
The stock Torsen LSD is flawless except under a very specific set of conditions. Under high cornering load, the inside wheel may lift high enough that all power gets transferred to it, which effectively makes the differential open.
During aggressive cornering over chicanes, corner exit or drifting this causes an interruption of power which will resolve as contact returns. Based on my style of driving and objective to make this a fun car to drive, I chose a 1.5 way lockup. Here are more details on the O.S. Giken Superlock LSD .
The following graphic from Cusco explain the 1, 1.5 and 2 way options available. Given my tendancy to hit curbs hard, this will help lap times.
Update!
Since installing the 4.8 final drive, my fingers got a little itchy and I went with a 5.1 final drive to push the limits of the gearing possibilities. Since then, there has been a 5.26 version that appeared as well. Like everything, there is a law of diminishing returns and the 4.8 might have been the best compromise with the stock transmission ratios. Again, this is heavily dependent on individual application to find the right ratio for you!
Grip Levels / Tire Choice
Grip Levels / Tire Choice
The Toyota GT86 / Scion FRS / Subaru BRZ came equipped with Michelin Primacy HP 87W GreenX summer touring tires sized at 215/40/17.
Given the shorter final drive and corresponding increase in torque, more grip is required, but too much will make the car harder to modulate and lose its fun to drive character. This whole build diary is concerned with fun, but I have decided to include a track setup if you are interested.
Analysis from Hot Version 116
Japan car program Hot Version 116 (source below) mounted the widest tire size they could on the car to experiment with increased grip. Keiichi Tsuchiya seemed to like the grip on the time attack, and the fast lap time reflected this. However, when asked to drift, the grip of the car made it very tricky. As mentioned the car was not developed to clock the most number of G's on the skidpad, instead engineered for fun.
Rim Width by Tread Width
I came across an article on Moto IQ which presented Billy's rule of thumb:
“For ideal handling: when sizing a tire for a given wheel, I usually target a tire’s tread width to be as wide as the wheel width, or 0.5” narrower than the wheel width.”
From my personal experience working on and driving countless track days, time attacks, street cars and race cars, this typically optimizes the tire’s carcass for response and outright grip. It also gives the sidewall a very slight ‘stretch’ or preload, which will improve the tire’s response and break-away characteristics.
OEM’s have to size tires to meet a ton of different targets like kerb to wheel damage protection, cost, snow chain clearance (almost all cars have to take this into consideration) all which affects the decision on the size of the tire and is usually a compromise at the expense of peak performance.
Naturally, I was curious to see what OEM manufacturers were up to and did a quick bit of research. Measuring the rim width of each of the cars, I subtracted it from the tread width (taken from tirerack.com) here were the results:
Mazda ND MX-5 0.5
Porsche GT4 -0.2
Scion FRS -0.5
Lotus Elise -0.7
Honda FD2 Type R -1.2
The cars above were selected due to their reputation as 'drivers cars' as opposed to outright performance. We can see manufactuters leaning towards tires with a slightly smaller treadwidth comparted to the rim width. You might be on to something Billy!
Road Wheel / Tire Setup
Wheels - TSW Nurburgring 17x8 ET35 (8kg)
1kg less per wheel for an inch more of width and a taller offset than the stock wheels. Ideally would have liked to go forged for stiffness, but I do like the CSL style design of these spokes.
Tire - Michelin Pilot Sport 4 (215/45/17)
The balance of grip and slip is about perfect, and 215 width on an 8 inch tire aligns with above.
Race Wheel /Tire Setup
My race tires go the opposite direction, maximising tire width and coupling with rim stiffness. Initially, I experimented with the RE71 tire as a street tire but quickly realised that its lack of progression, compromised wet performance, noise and raw grip make it more suited as a track tire.
Wheel - Yokohama Advan RC2 @ 17x8.5
Reason: Was sold to me at a price I could not refuse. This is a tarmac and gravel rally wheel which is perfect for some over the curb action on track days. Weight is: 8.4 kg.
Ideal Wheel - Rays Volk Racing CE-28N @ 17x9
Reason: Weighing 7,2kg, the CE28 is a forged monoblock wheel for stiffness and weight.
Tire - Bridgestone Potenza RE-71 @ 245/40/17
Reason: Winner of tirerack track oriented tire shootout for July 2016. The tread width of the 245/40 RE-71 is 8.6 inch (the 245/45 has a treadwidth of 8.2") which is a perfect match for the 8.5" rim.
Resources:
Road & Track's Tire Analysis - Does incremental upgrades of tires and reports findings.
Hot Version 116 - Explores basic modifications to improve lap times including tires.
Auto Express TRD Toyota GT86 - Review of the TRD version of the GT86 in the UK.
Size tyres for Performance - MotoIQ's article on proper tire selection
Rim Stiffness - A great article on the overemphasis of lightweight rims.
Track Tire Shootout July 2016 - Tirerack
Brakes
Brakes
Brake pad testing requires time and money, neither of which I did, so this setup represents more of a best guess of ideas at this point of time.
Pads, Pads, Pads
My first foray into brake pads entailed installing Project Mu Club Racer (RC09) pads and a new front rotor. Along with the stainless lines and better tires, this was a large step forward from the stock setup on the track.
Project Mu is a preferred brand for my simply because of its willingness to share some technical information about its pads, which makes them comparable. I discovered that Toyota created a one make race series called the 86/BRZ Race in Japan, and Project Mu supplies a front (RSF01A) and rear pad (RSR02).
Since I have a pair of Project Mu Club racers still on, I decided to keep the front pad and change out the rear pad. The following table describes the potential changes. The idea was to (slightly) move brake bias rearward to allow for more rotation under braking.
The operational temperatures of both pads (RSF01A & RSR02) are similar at around 200 deg to 800 deg (900deg for the front). The key difference lies in the friction coefficients. We can see that initial bite is indicated as 10% higher in the rear at 200 degrees, and 18% higher at the 800-degree mark. In their promotional video, the calipers and discs seem stock.
This is intriguing as it seems like the team at project Mu was trying to push the brake-bias rearward without changing the mechanical parts which are common to the car. The diagrams supplied for both the front (left) and rear (pad) seem to confirm this idea.
I do not think this is an outright replacement for a well engineered big kit, particularly for cars that see multiple laps on longer tracks. For more information on that, please check out the excellent article by Point Me By below.
The AP Sprint Brake Kit
A chance came up to buy a brand new AP Sprint Brake kit which a friend decided to sell, and in a moment of insanity, I pulled the trigger. 4.5kg of unsprung weight reduction drew me in, but the real benefit has been in brake feel.
It's subjective, but there is a huge pleasure in braking down hard from high speed. Brake pads became more of a challenge to get right, and I bought 3 sets below.
Carbotech 1521
Temperature Range: 0-426 ºC
Peak Friction: 0.47
My current street setup with just the right amount of feel, and NVH levels.
Fedoro DS2500
Temperature Range: 150-650 ºC
Peak Friction: 0.42
This was meant to be my original street/track pad. The reality is that it did neither well, with excessive noise on the street, and average braking on the track.
Carbotech XP12
Temperature Range: 121 - 1093 ºC
Peak Friction: 0.65
Balls to the wall track pad, which sadly I haven't gotten much use of this year.
Brake Master Mount
This is a harder one to comment on, but there is a brace that seems to solve a problem that I did not experience myself. That said, in the pursuit of perfect feel, I decided to add it.
Suspension Setup
Suspension Setup
Reviewing some interviews on the development of the car, an engineer described how their target to lower the center of gravity allowed the use of softer springs while keeping body roll minimal. In an episode of Jay Leno 's garage featuring his Lotus Elan race car build, he explained that Colin Chapman directed his engineers to give a generous amount of spring rate for the car, to maximise the contact patch on British b-roads.
An important element to the car's fun to drive character stems from the suspension needing stroke in order to transfer the weight to rotate the car, while maintaining the right amount of grip. Here is an excellent video that describes the Toyota's approach. I have two different setups that I have developed, the Street Alignment and Race Alignment.
Factory Suspension Setup:
1. Spring Rates
The main objective here is to keep maximum stroke and compliance by choosing a spring that is not too stiff. Many people look at the spring rates (2.3kg/mm F, 3.3kg/mm R) and automatically conclude that Toyota wanted to create a tail happy car with the stiffer rear. The truth is a little more complicated.
Spring Rates selection need to consider the multi-link rear suspension setup. This affects the motion ratio, which means that the rated spring rate at the rear of the car is not truly reflective of the spring rate that you experience.
To get the "wheel" rate of the rear spring, you multiply the spring rate by the square of the motion ratio. A simple way to do this is to multiply the rear spring rate by 0.75. Therefore, the true spring rate for the car is (2.3kg/mm F, 2.47 kg/mm R) which means the car is only a touch stiffer in the rear. From that starting point, we can study the popular coilover kits:
* RR actual is the rear spring rate (RR) x 0.75
I decided to go with the stock dampers matched to the RCE Yellow Springs which drops the car around 0.5". It is a little more of a drop than I would have liked, but the price was a deciding factor. If money was no object the KW Clubsports would be on the car.
2. Camber
Front: 4°
Rear: 1.5°
Camber is non-adjustable on the stock GT86 / FRS / BRZ unless you install camber bolts, plates or adjustable bushings. The Whiteline Camber Bolt Kit can add up to 1.5° of negative camber and I combine it with a camber plate with the objective of improving turn in and reducing understeer.
3. Caster
Caster Is not adjustable on the stock GT86 / FRS / BRZ. Given this car's street oriented usage, it is not something that you would usually adjust unless you were looking for more braking stability under high loads.
4. Toe
Toe is adjustable via factory rear eccentric bolts and front tie rods, and a touch of 'toe out' will help increase steering feel and turn in while slightly compromising steering stability at higher speed.
Personally while I can attest to the advantage of running the car with slight toe out on the front, it is a great contributor to tire wear, so I have made the change to go to a 0° setup.
5. Anti-Roll / Sway Bars
Given that the car now has stiffer springs, I'm still testing if adjustments need to be made to the anti-roll bar setup. For now, it will remain stock.
6. Tire Pressures:
I run 20% lower pressures in the rear in order to maximise grip on corner exit. Therefore, trail braking technique becomes very important to lighten the rear of the car during entry. For an in-depth guide on how to adjust tire pressures for competition, I would highly suggest reading Kenny Brown's excellent article.
Headers & Tuning
Headers & Tuning
The FA20 engine is highly optimised at the factory, and there are no major gains to be made aside from forced induction. Keeping the 'fun to drive' in check, throttle response is paramount, hence this guide will be looking at purely normally aspirated solutions.
Intake Boot
On the 2017 models, Toyota made the decision to straighten the intake boot that goes from the air filter into the throttle body. Perrin makes an aftermarket replacement that performs the same function, in addition to deleting the sound tube.
Headers
There are two types, equal length headers (which came stock) generally gives more peak power, and unequal length more torque lower down. One other subjective advantage of the unequal length headers is more rumble in the exhaust note.
Many reviewers have commented on the somewhat uninspiring sound of the FA20, and given that this car was never about the numbers, it is arguable that this adds to the overall driving experience. Another minor advantage is a weight saving from the OEM.
Tuning
With a fresh air filter and a HKS Hi-Power exhaust installed, the car was tuned. Breathing through the uncorked headers, it gained around 20whp that peaked at 185whp @ expected 6,900 rpm. Torque also went up around 10-12%. One of the advantages of the Ecutek tune was changing the throttle map. After 8-9 tries, a map with a good balance of aggressiveness and drivability was achieved.
Crawford Power Block Spacers
Cobb power blocks were installed after the tune. Installed purely on the basis of this article from Moto IQ which demonstrated how the increased intake manifold volume translated into more area under the curve. While there may have been some power gained under the curve, I felt the original powerband was a little more fun, so ended up removing them.
Other Areas
Other Areas
This section is random, but covers the stuff I could not fit anywhere else.
Engine Oil
After endless debates both offline and online, Gspeed from the FT86 forums posted this excellent thread that shares data he has captured on engine oil degradation in relation to temperature.
As my car is setup for sprints rather than longer races, I have chosen not to install an oil cooler. Monitoring oil and water temperature becomes crucial and is done with my Pivot DPW gauge mounted in the driver's side air conditioning area.
Once my oil temp hits 265f (130c) or my water temp hits 225f (108c), I will back off till it reduces to 210f or 100c before I start my next run.
Best Motoring ran an endurance race a number of years ago which demonstrated how professional drivers manage oil, water and cabin temperature in different ways. It is a good resource to understand the stresses a car and driver goes through lap after lap.
One compromise I have considering is to add a water to oil cooler kit. This basically revolves around creating a pathway for the oil to be cooled by the radiator, and is an OEM feature of FA20DIT Subaru Forresters. Tested results of such as setup will be around 10% lower oil temp, 5% higher coolant temperature, and faster warmup. These results will be negated the harder and longer a car is driven, so it is not a replacement for a dedicated oil cooler for hard use track cars.
Flywheel
The stock FRS/GT86/BRZ flywheel weighs 9.45kg. While much has been written about the improved engine response and free revving nature of the lighter flywheels, less has been written about the compromises that they bring.
Without getting technical, the flywheel does not add power to the car, it just makes the time taken to get from one rpm point to another shorter. Ideal for a race track, tiring day to day.
Stock flywheels are designed to reduce noise, vibrations, and harshness (NVH), and the lighter you go, the noisier they get. Check out this video of one of the lightest flywheels available (prolite), to get an idea of what this means.
After much consideration, I have decided to go with the Toyota TRD flywheel. As this car is being built for response, it should be ideal. Even their official website warns against the harsher nature of this part, but I think it should be the best current compromise.
Alternator, Crank and Air Conditioner Pulleys
Toda is a well respected Japanese racing company that produced a lightweight replacements for the alternator, crank and a/c pulleys that shaves 2kg. Most of that weight savings comes from the crank pulley and I was skeptical of their claim of a 3-10hp gain throughout the power curve. After installing them and tuning the car, I can verify that they were correct, and am extremely happy with the results.
Page updated
Google Sites
Report abuse