Tuesday, January 31, 2012

A Little Experimentation

As I've been preparing for the start of the 2012 season, I've had a lot of questions about the changes I'm making to the car:  new differential (4.30 and urethane bushings), beefier clutch, wing, and moving back to stock sway bars.  Will the new setup work? Will it take 2 or 3 events to figure out how to drive it? Will I trust it too much and stuff the car?

There are a lot of moving parts up there--no real scientific method was used, other than trying to maximize the allowed points for my class. The wing was the most obvious aero choice, but it's not being accompanied by any corresponding front aero, so the biggest question I had was if I was seriously upsetting the handling of the car.  Most folks tend to agree that adding down-force means adding spring-rate, but that's not in the plan or the budget, so...eek?

Last night I had an idea.  One that's probably been done by countless others, and with better results, but it was new to me.  I fired up Gran Turismo 5 on the PS3, hooked up the PlaySeat & Logitech G25, and built a '91 Eunos Roadster to match my real car's configuration from last year (except for that whole 1.6L ~ 1.8L thing):

Power:  ECU, header, exhaust, intake
Suspension: -20mm ride height, -2 deg camber, 0 toe, 12.5kgf/mm front springs, 5.8kgf/mm rear springs, 7 front sway bar, 4 rear sway bar
Tires: Intermediate racing
Drivetrain: stock clutch, semi-racing flywheel

I added ballast and moved it rearward to achieve 51/49, and biased the brakes forward 7/5.

The car was nearly undriveable.  Heavy braking resulted in an extremely twitchy tail (just exactly like in my real life car), and it took all my concentration to get the car around the High Speed Ring without crashing.  That's ridiculous, as it's one of the easiest tracks ever. The best lap I could manage was a 1:26.3, and I was actually starting to get into a consistent rhythm.

So then I went and updated to this year's intended configuration:

Power: same as a bove
Suspension:  same, except sway bars:  2 front / 2 rear
Tires: same as above
Drivetrain: single-plate clutch, semi-racing flywheel
Aero:  Wing w/ downforce set to 13 (an intermediate setting)

I expected to have drag issues on the straights, but didn't. I can't decide if this is a limitation of the game or because speeds are not truly significant enough in a Miata to really affect top speed, but the difference in the corners was immediately noticeable.  No longer did the tail want to swing out under braking, and the steering was a tick heavier entering corners.  I've notoriously had a habit of tossing the car into a corner and powering out, but the end-plates on the wing prohibited that.  I could only pour the car into the corners, which is a good thing.

Ultimately, after just 3 laps driving with the new configuration, I'd taken almost 2 full seconds off my lap, and was far more confident entering turns 3 and 4. I was able to carry a full 10mph more through turn 4 and get on power far earlier in turn 5.  I only put in about 6 laps, but was turning consistent 1:24.7 laps.

That's a 1.6 second change doing nothing but adding downforce and reducing sway bars (forgot to change the old configuration to a 4.10 rear end). I'd say it was better than expected, but it's hard to say how well the video game will translate to the real world. But right now I'm feeling a little more confident in the changes I've made.  In fact, if I scale the 1.6 / 1:26.3 to ??? / 2:19.8 (my fastest at VIR), I get a 2.59 second gain, or 2:17.2.  That's huge, and could make this the best year ever.

So now the question remains:  is GT5 a good enough simulator to test real-world performance mods?

UPDATE: The gearing in GT5 was completely wrong for a stock Miata. After correcting it and updating one configuration for a 4.10 rear, I re-ran the test.  With a little practice I was able to get the old config (no wing) down to a 1:24.9, and the wing config down to a 1:24.3. It will still carry a full 10mph more through both T3 and T4, but the gearing and wing offset each other everywhere else.

No comments: