Project Nissan Silvia: Part III

So much for dialing in the suspension and reaching for 1.0g on the skidpad. Now I'm drag racing for my job.

Time to get to work. Project Silvia was making 274 hp at the wheels last time it graced these pages (June '03), and shortly after that story went to bed, we took it to the strip. With bald tires and an open diff, it managed a 13.4- second quarter mile at 109.9 mph, spinning the left rear tire all the way through first and second. By my calculations (aka wild-ass guess), we should need only three parts to make the 12.9-second quarter mile that will keep my paycheck coming. First, obviously, is a limited-slip diff. That we've come this far with an open differential is just plain stupid. With an open diff, 274 hp is completely useless. You can't accelerate hard in first or second gear. You can't initiate or control a powerslide. Hell, you can't even acccelerate and steer at the same time. It's no wonder the rear tires were bald after less than 5,000 miles.

There are countless options for a limited slip on this car. First, some 240SXs actually had one from the factory. It was a viscous limited slip, though, and they're not exactly known for longevity. If you have one already, all it takes is a few powerslides to decide if it's still good. If it is, lucky you.

The Nissan viscous limited slip works by heating up a visco-reactive thermo-hardening puddle of goo trapped between a stack of clutch plates that are close, but aren't actually touching each other. Every other plate is attached to either the right or left axle, so when the two spin at different speeds, all the viscous fluid gets tossed and tumbled and twisted and sheared between counter-rotating plates. This makes the fluid thicken, expand, and do its best to grab all the plates and hold them together.

The main advantage of a viscous limited slip is smoothness. If everything is working properly, it will tighten quickly and progressively when needed, then quietly return to its open state. A viscous limited slip is slow to react, though, requiring the wheels to first turn at different speeds before reining them back in. It also has limited torque transferring abilities. You can't have one wheel on the ground and the other in the air and expect the fluid to be strong enough to send all the power to the wheel on the ground. The biggest limitation with a car this old, however, is that with hard use, the fluid will break down, becoming less and less effective with age and abuse.

If you're junkyard shopping for a stock limited slip, or one from a Z32 300ZX or an Infiniti J30, which are similar enough that their guts can be swapped but don't necessarily have interchangeable cases, consider the fact that the viscous fluid could be broken down from hard use in its previous life, and it might not actually be much of a limited slip anymore. If the junkyard is cheap, it may be worth the risk, but if it's only a few hundred less than an aftermarket diff, step up to guaranteed performance.

Aftermarket diffs come in two basic flavors: clutch-type diffs like the KAAZ or Cusco units, or helical-gear differentials like the Quaife. Which is best depends on what you want to do.

In a clutch-type diff, the stack of clutch plates actually touch each other, and as torque is applied, they get rammed together even harder. This makes a clutch-type diff fast to react, since it only takes torque, not an actual change in wheel speed, to make the diff tighten. There's also some base resistance, depending on how much pre-load is on the clutch pack.

Another clutch diff advantage is adjustability. You can change how many plates are used and you can adjust the pre-load on the plates. Each adjustment requires a teardown, but at least it can be done.

The Quaife differential works on an entirely different principal. Each axle is attached to a gear, called the sun gear, and each sun gear has several small gears around it called planet gears. The planet gears from one side mesh with the planet gears from the other side, so when you put both drive wheels in the air and turn the left wheel clockwise, the left planet gears turn counterclockwise. They turn the right planet gears clockwise, which turn the right wheel counterclockwise. In other words, the wheels spin in opposite directions just like a normal open differential.

The magic comes from all the gears having their teeth cut helically. That means the teeth are angled so that if the gear were long like a log, the gears would spiral around the log in a helix. When you put a big load on a pair of helically cut gears, there's a reaction force pushing them sideways.

Look directly at the interface between two gear teeth and the reason for this sideways force is pretty obvious. The gear teeth meet at an angle, and when you try to push straight through an angled interface like that, you get shoved sideways. The latch on the front door of your house works the same way. Close the door until the latch touches the striker plate and you have a similarly angled interface. Push straight on the door, and the sideways reaction force pushes the latch into the door. In a helical differential like the Quaife, the reaction force shoves the sun and planet gears sideways, where they rub against the ends of their housings. This rubbing encourages both wheels to turn together.

The reaction force only happens if there's some resistance to your shove, though. If there was no resistance in our door latch example, the doorframe would just fall over. If there's no resistance from the other gear, it will just turn. This is why, technically, a helical limited slip isn't really a limited slip. Put one wheel on ice and the other on pavement, and there won't be enough resistance to prevent all the power from going to the ice.

This is why Quaife prefers the term "torque-biasing differential." Put both tires on pavement and go around a corner, however, and if the engineers who picked the gear angles did their jobs, the gears will send more torque to the grippier outside tire in just the right proportions to keep the inside tire from spinning.

If you're drag racing, where one of the diff's primary functions is to do good burnouts, a clutch-type limited slip is the better choice. As soon as torque is applied, the clutch plates will get shoved together, locking the wheels into perfect unison no matter what. If someone didn't water the burnout box evenly, a Quaife may spin one wheel more than the other. On the launch, where there's more grip, it will work properly, but everyone will still be laughing about your peg-leg burnout.

I don't really care about burnouts, though, and a Quaife will give all the grip needed at the launch of our one drag racing attempt. After that, if I still have a job, it's all about handling, and that's where the Quaife has the advantage. Driving style has a lot to do with what kind of differential you prefer, but I prefer a car with neutral, predictable handling and a smooth, progressive transition to oversteer. A clutch-type limited slip will tend to be more aggressive in locking the two rear wheels together under acceleration. With both wheels trying to go the same speed, the car will try desperately to go straight. That means understeer.

A two-way clutch type, which is active under engine braking as well as acceleration, you'll get understeer going into the corner as well as coming out. The Quaife is effectively an open diff coming into a corner, so steering feel is completely unchanged. As you power out of the corner, it prevents wheelspin, but doesn't seem to create the battle between the front and rear tires a tight clutch-type diff does. In theory, a one-way or 1.5-way clutch diff could be adjusted properly to do the same thing, but in our experience, the helical differentials are almost always smoother and more seamless as they go about their job.

With the added grip of two-wheel drive and a new pair of Falken Azenis Sport tires, we headed to Buttonwillow Raceway where Hotchkis Performance was hosting a track day. There, sandwiched between Chevelles and turbocharged M3s, I managed a 13.14-second pass at 108.95. Not bad, but we still need something else if I'm to stay out of the unemployment line.

Knocking that last 0.15 seconds off is going to mean more grip. We still had wheelspin all the way through first and a dozen feet or so into second gear. The responsiveness of the Disco Potato turbo makes it easy to modulate power delivery once the tires start spinning, but it's still nearly impossible to prevent wheelspin in the first place.

Time for part number two: some drag tires. We ordered a pair of 245/45ZR-17 Nitto NT555-R drag radials, slapped them on Project 350Z's stock rear wheels, and started thinking about part number three: the clutch.

When we first installed this engine, we replaced the smoked three-puck mystery clutch that was on it with a Jim Wolf Maxima pressure plate and an organic disc. The mild, streetable clutch was supposed to be good for 300 lb-ft of torque, according to JWT's Clark Steppler, as long as we weren't drag racing and doing burnouts all the time. Turns out he was right. A few months ago, when we tried doing a burnout shot, it went up in smoke. It was one of those horrible moments when you realize, halfway through your big glorious burnout, that the smoke isn't coming from the tires.

Amazingly, after cooling down, the clutch recovered, eventually delivering our 13.14-second run without drama. But now, the thought of adding the grip of drag radials, plus their need for a burnout to warm the rubber, has me worried. Any other day I'd try it, but my job is on the line.

Another call to Jim Wolf and a new clutch idea is hatched. The Silvia flywheel has enough surface for a larger, 250-mm clutch disc, and a non-turbo Z pressure plate will fit if we drill the right bolt pattern in the flywheel. Well, actually, if Jim Wolf Technology drills the right bolt pattern in the flywheel. Well, technically, if Jim Wolf himself drills the right bolt pattern in the flywheel. At night. On two hour's notice. That would do it.

Our 9-pound chrome-moly Toda flywheel is difficult to machine, so if you're having a similar flywheel redrilled, tell the machinist it's chrome-moly steel before they break all their drill bits on it. They'll hate you a lot less that way. Our new setup still uses an organic disc, since, after my race for employment, I'll still want a comfy, streetable clutch. The disc has grown from 240mm to 250mm in diameter, with a more aggressive, more heat-resistant organic lining, and the pressure plate's clamping force has grown from 700 to 900 kg.

After a quick installation at XS Engineering and a week of breaking in the clutch on the street, it's the night before the moment of truth. Twelve point nine or my job. The car is fast and consistent, and the NT555Rs should easily deliver the last 0.15 seconds I need. I'm not really scared. I'm cautious, though, so I get half a tank of VP 103-octane unleaded. We've heard a few pings during quarter-mile runs before, and I don't want to be unemployed and not have a car.

Driving back from the gas station, the car starts feeling slow. It's funny how after a few months, 274 hp doesn't impress as much as it used to. Maybe it's the Quaife. The car used to spin the tires any time you hit boost in first or second. Now it just squats and goes. Psychologically, that doesn't feel as fast.

Just to be sure, I jump on the freeway and do a third-gear pull. It's making 15 psi of boost, like always, but it seems a little more laggy, the turbo seems extra noisy, and what's this? It's misfiring at high rpm. Crap. I like this job.

What could this be? I've been getting crappy gas mileage this month, and the tailpipe is caked in black soot. I've been thinking the O2 sensor is bad. Maybe the plugs fouled on the way back from Buttonwillow. I need new plugs. I need fancy-pants Iridium plugs.

It's 9 p.m. on a Wednesday. Where am I supposed to find Iridium plugs for a Japanese-spec engine? Desperate, I call XS Engineering. They're actually there. Working late trying to fix another magazine's screwed-up project car.

Halfway there, the car is getting worse. The turbo is getting noisier, the misfiring is starting before it makes full boost.

Suddenly my brain turns back on. We only had two timed laps around the road course at Hotchkis' Buttonwillow extravaganza. The Silvia set top time of the day on the first lap, then blew up the compressor discharge hose on the second lap. The hose sits right next to the exhaust manifold, and we didn't think to order a Nomex-wrapped hose. The polyester fibers reinforcing the hose melted and the hose blew out. Since then I've been using a piece of stock intercooler hose cut from Vishnu Performance's EVO VIII. It's rubber, with external reinforcement. It must blow out more slowly.

That's it! A new HKS hose from the scrap box and, finally, a layer of heat wrap to protect it and everything is working again. Turns out 274 hp still feels fast.

It's 9 a.m. I'm at California Speedway. With flat-black paint, no air conditioning and a hour-long drive to get here, I've had plenty of time to realize things are going bad again. Weather is a big issue with turbocharged cars, so I had planned to hit the track early and rip off our first run before the sun had a chance to scorch the track. Didn't work. It's already 90 degrees out. We should have come at 6 a.m.

With the Nittos at 20 psi, I rip off a test run. 13.4 at 104.7. What the hell? I bogged it; let's try again. 13.6 at 103.8. Crap. The simple bleed valve boost controller we're using is very air density dependent, making more boost when the air is more dense. On the dyno it was 16 psi. On cold days, it runs as high as 18, with occasional spikes to 20. Today, it's only 14 psi. The trap speed shows we're way down on power.

I crank the knob, aiming for 18 psi. I end up with something erratic, like 18 in first, 20 in second, and 20, with a spike to 22, in third. Whatever, I have race gas. I drop the tires to 15 psi, and do a big burnout. Then have to drive up and down the track to bring the coolant temp back down.

By now the temperature is 95 degrees.

The tires hook up, the engine bogs a bit, then comes on boost and spins the tires again. 13.05 at 109.5. This weather really sucks. At 20 psi, I have the same trap speed I used to have at 16. I drop the tires to 12 psi. Too low, the tires start cupping, leaving dark bands at the edges of each burnout mark. Back to 15, I try every different launch I can think of. It's bog and spin, or spin and hop. There's no hook-up-and-go. This would probably be better on a prepped and seasoned track. A concrete pad with VHT and fresh rubber. But we're alone, on bare asphalt, like always.

I try doughnuts instead of burnouts. Small chirps instead of big roasters. The best I can get is 13.04 at 110.0. It's 99 degrees.

If it weren't for all this pressure, I'd be thrilled to see my 0-60 times running just under 4.9 seconds; instead, I'm freaking out over 0.05 seconds. Unable to get any more from the Nittos, I return to the Azenis. It's an endurance racing tire disguised as a street tire, so it shouldn't be faster than a drag tire, but at 225/45ZR-17, it's slightly shorter than the 245/45ZR-17 NT555R and the Volk CE28s are at least 10 pounds lighter than the 350Z wheels. Maybe trading a little grip off the line for better gearing will give me the 0.05-second advantage I need.

Launching the Azenis is easier. They spin cleanly without hopping and can be modulated easily. Of course, they aren't as fast. I manage a 13.09 at 110.5 mph, but no better.

Hot, sweaty and dejected, I can already hear the hum of flourescent lights at the inevitable cubicle engineering job I'm going to have to find.

Then my brain delivers my salvation. Correction factor! I don't believe in correction factors. They aren't accurate on turbo cars, they often aren't accurate on fuel-injected cars, but they always make you faster. "Motor Trend" believes in correction factors. Neil Chirico is their road test editor.

I have his phone number.

"Neil, it's Dave Coleman. What would a 13.04-second quarter mile at 110.0 mph correct to on a 99-degree day with a barometric pressure of 28.8 in. Hg and 19-percent humidity?"

"12.66 at 113.2 mph"

"Thanks, Neil. I owe you one. A big one."