Like most athletes, race cars must also be capable of performing at peak form in a number of different environments and conditions. And because we've done all of our racetrack evaluations at Thunderhill Park, it may be a good idea to expand our horizons a bit. With some luck--and, of course, a natural unfamiliarity of a new track--we're bound to learn something about our beloved Project RX-7.
So where do we take what is perhaps the fastest, track-ready road car we know of? Why, to the fastest road course in all of California of course! Located near Edwards Air Force base in Rosamond, Calif., Willow Springs International Raceway is an excellent complement to the more traditional Thunderhill Park. While some may argue that Thunderhill is the more technical of the two road courses, Willow Springs--with its simply frightening 120-plus mph decreasing-radius turn 9--is, by far the most intimidating. Also unique to Willow Springs is its strangely unorthodox Turn 3 through 6 sequence, a series of turns that combines off-camber, blind turns with a steep and intimidating hill that places exceptionally high demands on a car's suspension while allowing different cars to follow equally different lines.
Willow Springs was Mazda's test track of choice nearly 10 years ago during the evaluation phase of the finalized third generation RX-7 "S1" prototype. During this time, the new RX-7 was pitted against the then-current reigning "King of the Track", Honda's exotic NSX. As fate would have it, a few smartly modified NSXs were ominously present during our very own Project RX-7 testing session as well. To prepare for Willow Springs, we installed a set of Performance Friction 90 carbon metallic race pads. Although they dust terribly and squeal like amplified pigs when cold, they should be able to withstand the rigours of Willow Springs Raceway better than the set of Hawk's high-performance street pad we've been using at Thunderhill.
Track Testing At Willow Springs
Those who are also following the build-up of SCC's Project Impreza (which, by the way, shares a residential garage with the RX-7) may recall most of the Subaru's track testing is done at Willow Springs. However, never once have we blurted such phrases as "Please, for the love of God, slow this thing down" or "Oh no, I'm think I'm going to vomit" about the Subaru's driving characteristics. These pleasantries were uttered several times (unfortunately, by me) during one such testing session in the RX-7.
With an extra 200 hp and infinitely better aerodynamics, the similarly sized RX-7 blitzed through the big track like a thoroughbred race car in fast-forward mode. At least that is how it felt from the passenger seat, as I rocketed into turn 8 at 140 mph, while peering through the cracks between my trembling fingers. Thankfully, I wasn't driving the car during this primitive fit of self-preservation; it was race car driver Gary Sheehan that was given duty behind the wheel. Readers may remember that Sheehan has track tested a few SCC feature cars in the past year. During that time, I've learned Sheehan is a very critical driver--readily willing to heap mounts of trash upon a car's slightest shortcoming with little or no remorse for anyone's feelings. So what were his impressions of the car? Let's ask him.
SCC: The last time you drove Project RX-7 on the track, it was completely unmodified--driven right off the used car lot and directly to Thunderhill. What did you think of the stock car?
Gary Sheehan: I remember being very impressed with the power delivery and feel of the car. Corner exit required sensitive throttle application to prevent excessive wheel spin. I did feel the car was overpowered for its brakes [with generic brake pads] and at one point entering turn 10, the fade was so bad I contemplated what part of the embankment on the outside of the corner would inflict the least amount of damage. Luckily, no harm done.
SCC: What were your first impressions of the modified car? How was it different?
GS: From looking at it in the paddock, I did not expect it to be same car I drove at Thunderhill. It had bigger brakes, different suspension, a big rear wing, RA1 tires, yada, yada, yada. But it was only until I finally got the car out on the track that it started to hit me just how much of a change this car had gone through. The first thing I noticed during the warm up lap was the suspension. It was firm, but not harsh.
SCC: How did Project RX-7 feel when pushed harder?
GS: As the speeds started to pick up, I was very impressed with the car's stability. Immediately, I felt the car was very confidence inspiring. More importantly, once the limits of the chassis were reached, it did nothing dramatic--it was very forgiving in that respect. I also liked the steering--very light and offered good tactile feedback. That, combined with quick engine response, made the car very easy to position, rotate and manage to the exit of a corner.
The brakes were absolutely wonderful going deep into the high-speed braking zones of turns 1, 3 and 9. I wasn't even aware of them, which is how it should be when driving at the limit. All in all, the car felt much more at home on the race track than even a few race cars I have driven.
SCC: Did you notice any problems at all?
GS: Well, I did go off the track once early in the session. The car inspires so much confidence that I started driving too fast too quickly. On only the third lap I was driving the car very hard and entering the corners a bit faster than I should have been doing with such little seat time. Big increments of speed can make for some big surprises. As I learned in turn 2, the car can exhibit slight understeer during long steady state at-the-limit cornering. At the speeds we were traveling at (90-plus mph) this changed our trajectory slightly and caused a four wheel off at the exit of corner 2. It really isn't a big deal at all. Except for the cone that was laying on the ground. Which I hit. Hard. But even off track those high speeds with the car completely off the track, it did nothing that surprised me and we were able to continue the lap with hardly a lift and maybe a tick or two on a stopwatch. But I must confess, the off was completely my fault, not something the car did.
The only real problem with the car was engine temperatures that began to rise after four or five hard laps. That's too bad. I would have loved another twenty laps in the car to truly learn its capabilities and to turn in some really hot laps. But between the car running slightly hot and you whining and trembling in the passenger seat, I got the idea it was time to pit.
"Why," one may ask, "does Project RX-7 begin to run hot at Willow Springs and not at Thunderhill?" The answer most likely lies in the fact the former is 2,500 feet about sea level. Like most OEM turbo cars, the RX-7's boost control system compensates for elevation changes. That is, at higher altitudes, the turbos are asked to spin harder and faster in order to make up for the lower than normal air density.
This is one reason turbochargers are sized large enough to be able to provide the extra airflow with minimal reductions in compressor efficiency when the air gets thin. However, it looks as if we used up most, if not all, of that reserve capacity during our previous quests for more horsepower. As a result, Project RX-7's hard-working turbos are simply over-exerting themselves beyond the point of reasonable compressor efficiencies. This creates heat. A lot of it. Combined with a sultry, 105-degree Fahrenheit ambient track temperature and thin air that doesn't carry away as much heat, it appears as if we were asking a bit too much from the RX-7's upgraded cooling system.
One easy solution to this problem would be to reduce the boost level by carefully manipulating--either electronically or manually--the wastegate signal. Another safer, albeit far more expensive solution, would be upgrade to higher-flowing turbochargers. Perhaps we'll try one or both upgrades in the near future. But for now, we'll just accept the fact that, under these atypically extreme conditions, running a periodic cool-down lap is going to be a fact of life. Besides, even under these circumstances, Project RX-7 still managed to run rings around all the modified NSXs and Skylines that were present during our track testing sessions.
Pushing The Envelope
So here we are with a road car that offers incredibly high levels of track performance without extolling any significant compromises upon the driver when used as a daily commuter. Since, we've never been good at leaving well enough alone, let's continue our quest for maximum performance while keeping a vigilant eye on streetability. Many would argue we have taken the suspension tuning as far as we realistically can go without beginning to lose sight of our goals.
Never having gotten the impression that we need a stiffer suspension on the track, we would tend to agree. While we will continue to play around with weight reduction, sway bar tuning and ride height in an effort to ameliorate the ever-so-slight understeer we experienced at Willow Springs, we are confident we have found our coilover system of choice. For now, let's focus our efforts on the rest of the car.
Building a Better Chassis
Two areas that we have already addressed in practice, but not yet covered in print, are chassis reinforcement and driver's safety. Because Project RX-7 began its life as a Touring Package model, it wasn't equipped with a factory front strut tower brace (which was available only in the aggressively dampened R1 and R2 sport packages.) Since our much-higher-than-stock spring rates place a great deal of load on the car's chassis, a strut brace seemed like a good idea. As usual, M2 Performance came to our rescue by providing us with a handsome, lightweight (3.5 lb) strut tower bar manufactured by Cusco. As we expected, the effects of the additional bracing were easily noticeable--enhancing both turn-in response and steering feedback. As an added visual bonus, the bar's steel strut tower mounts nicely matched our intercooler's blue silicon connecting hoses.
At 72 lbs, a roll bar doesn't really jibe with our weight-watchers approach to performance upgrades. Nonetheless, it is absolutely necessary in our minds. Provided by Kirk Racing Products, the heavy-duty roll bar made of beefy 2-inch tubing, could potentially keep an unfortunate accident from turning into a tragic disaster. Like most production road cars, the RX-7 wasn't designed to provide significant rollover protection. Considering the speeds at which our car negotiates most road courses and the exuberance at which it is driven (as we have seen firsthand, of course), we stand by the notion that a functional roll bar is mandatory and not optional equipment.
Installation, although a bit time-consuming, is very straight-forward and best done with the seats removed from the car. It's also a good idea to wrap towels around the sharp-edged mounting brackets to keep them from scratching paint or gouging the interior panels. Once the hefty steel hoop is pushed against the rear plastic storage bins (which will still remain perfectly accessible, by the way), all one needs to do is drill four holes in the floorboard, place the reinforcing steel blacking plate under the car and in-line with the hoop's floor mounts, and tightly secure the whole assembly with the provided nuts and bolts. Similarly, the rear bracing is mounted in the hatch's tool box cabinets (which, unfortunately can no longer be used). Voila. Rollover protection. Unlike the other modifications we have subjected our project car to, we hope we will never get the chance to test this particular product.
Wanna Go Faster? Go Light
When it came to reducing weight, Mazda engineers left little to be desired. However, they were forced to make a few concessions in the face of consumer expectations. First of all, they fitted the RX-7 with a 41 lb, conventionally sized, wet-cell battery. To make matters even worse, they mounted it on the front, driver's side corner of the car--far way from the center of gravity. Fortunately, there are several ways to approach this situation, varying from a simple battery replacement to as elaborate as a complete battery relocation.
The first option, as provided by N-Tech Engineering takes no more than 10 minutes and removes 21 lbs of mass. Appropriately called the Battery Miniaturization Kit (BMK, for short), it replaces the heavy lead acid battery cell with a slim (7x3x6-inch), maintenance-free 14.7 lb dry cell. Installation involves nothing more than removing the stock battery assembly and securing the BMK in its place on the frame rail. Rated at 280 cold cranking amps, Nick Reifner of N-Tech claims the BMK has enough juice to even power cars with upgraded stereo system.
Battery relocation, as suggested by Brian Richards of M2 Performance, is the other alternative and involves a bit more effort. However, the results, in terms of overall weight distribution and battery longevity are hard to beat. Using a larger (7x7x5.5-inch) Hawker Odyssey dry cell battery, Richards completely eliminates the stock-mounted battery assembly all together. Instead, he mounts the Hawker dry cell in the small storage bin just behind the passenger seat, drawing the extended battery cables from the fender well, into the cabin, under the carpet, and into the bins (from underneath). "That sounds like a lot of work," Richards admits, "But if you have to have weight [all 25 lbs of it], you want it close to the ground and away from the front of the car." As an added benefit, relocating the battery frees up enough underhood space for another, even more massive, air-to-air intercooler. But more on that later.
Perhaps the only thing better than relocating mass where it can be put to good use is eliminating it all together. Such is the case with the RX-7's ridiculous Acoustic Wave self-amplified sub-woofer system. Shaped like a deformed anaconda, the plastic convoluted mess takes up much of the already limited rear hatch space, adding 22 lbs of dead weight in the process. A few twirls of the 10 mm sock wrench and out it goes. Did we miss the mind-numbing, headache-inducing, one-note bass boom that it provides? Not at all.
In order to quantify the changes in weight distribution the different weight manipulating modifications have made, we paid a visit to our good friends at C2 Automotive. Equipped with a state-of-the-art Intercomp Model SWJD weighing system, they tested each of the four different configurations with 160 lbs of journalist in the driver's seat. All tests were done with one-half tank of gasoline. The results, while not exactly earth-shattering, are interesting nonetheless. What the results illustrate most clearly is our driver-equipped, stock-batteried, bass-pumping RX-7 spreads its weight rather unevenly among all its tires. Not surprisingly, the right side (with a driver) is heavier than the right side. While the weight over the rear tires are nicely balanced with a differential of only 15 lbs, the front is uneven by a whopping 118 lbs! Such findings would suggest that, all other things equal, the car exhibits more understeer cornering right than it does left. Not coincidentally, the understeer race car driver Gary Sheehan complained about was only apparent on Willow Springs' turn 2 and 8-9 combination--both high-speed right handers.
As we expected, installing N-Tech's BMK helped matters noticeably. With the massive stock battery removed and replaced by a slim, lightweight dry-cell, both front-to-rear and left-to-right weight distributions improved. The weight differential over the front tires was reduced from 118 to 109 lbs. More unexpected, however, was the weight over the rear tires which evened out considerably. It appears, like a seesaw, that removing weight from the left front of the car transfers weight to the opposing right rear.
As Richards predicted, relocating a medium sized battery from the stock location to the storage bin behind the driver, does improve matters even further. Now, the weight differential over the front tires has reduced even further, from 109 to 101 lbs. But more importantly, the extra rear weight has been positioned down low, toward the center of the car. The removal of the 22 lb subwoofer improved matters as well--both the front-to-rear and left-to-right weight distributions evened out considerably. With our option-laden and roll bar-equipped Project RX-7 weighing in at a lighter-than-stock 2850 lbs (without driver)--we couldn't have asked for better results.
Intercoolers: Is Bigger Always Better?
Offering almost 50 percent more core volume than M2's Medium intercooler, the Race intercooler literally fills up the entire front end of the engine bay, requiring the installation of a polished aluminum "S" shaped elbow just upstream of the previously-installed GReddy elbow (both of which are included in the kit).
The theory behind the larger intercooler's frontal area advantage is easy to understand. According to Corky Bell, in his book Maximum Boost , "In many respects, frontal area reflects the amount of ambient air that goes through the core to cool the intake charge. The greater the mass of ambient air that can get through the core, the greater the cooling capability." To calculate the airflow rate that goes through the Race IC's core at, say 60 mph, we multiply its frontal surface area by the car's forward velocity.
In this case, with a intercooler frontal surface area of 1.49 square feet and forward velocity of 5280-feet per hour (which is only 1 mph, but any speed is valid for comparison purposes), we calculate the airflow through the core to be a whopping 7867.2 cubic feet per minute. The M2 Medium intercooler, by comparison, is cooled with only 5280 cubic feet per minute. But to put things in perspective, the laughably small stock intercooler is treated to just under 1900 cubic feet per minute. If you are saying to yourself, "It can't be that simple," you would be absolutely right. The previous calculation assumes the intercooler has direct and unlimited access to ambient air. While that would be the case in a fully exposed front mounted intercooler, it does not hold true in our case when the intercooler is mounted under the hood and fed by a duct that offers a 2x11-inch view of the outside world. This makes things rather complicated, but as we've seen by the previous intercooler testing of M2's Medium intercooler, the ducted intercooler system works. In fact, using the results from our on-road intercooler testing (see part VI), we calculate intercooler efficiencies to be a remarkable 86 percent. In other words, during that particular test, the intercooler removes 86 percent of the heat generated by compression.
In our experience, it really doesn't get much better than that. For comparison, the stock unit--which is undersized and overly restrictive for our current air flow levels--measures in at a dismal 44 percent. The million dollar question is: Will the larger Race intercooler, fed by the same limited flow of ducted air, offer better real world performance? Keep in mind that the superior intercooler wouldn't just do a better job in cooling air--it would also have to offer similar or lower levels of pressure loss. Which intercooler is better then? To find out, we compared the Medium and Race intercoolers using our standard "on the road" testing method. Simply put, we recorded peak pressure loss (using a differential pressure gauge) and intake temperatures (using a fast-reacting thermocouple/fluke meter) during a variety of conditions. The results were surprising. [See the IC Comparison Test.]
During the full-throttle run through third gear, both intercoolers exhibited identical thermal efficiencies (approximately 84 percent). However, the smaller of the two offered significantly less peak pressure drop across its core. Looking at these results, it becomes clear the smaller intercooler fares better during this particular test. One reason for its short-term transient performance advantage may lie in its number of cross-flowing air channels--it has 22 while the Race intercooler has only 18. The greater number of channels, all other things held equal, could possibly account for the lower pressure drop.
A simplified analogy would be trying to blow air through one plastic straw. Trying to blow that same amount of air through several straws would certainly take less effort, right? Now, in the case of the larger intercooler, not only are there fewer straws, but the straws are longer as well. Now, flowing the same amount of air becomes even more difficult, especially since those straws are filled with tiny fins designed to distrupt airflow--hence the 0.6-psi extra pressure loss and resultant lower boost pressures.
So, one may ask, "If the smaller intercooler offers the same cooling performance and lower levels of pressure loss, why would one ever consider the larger intercooler?" Good question.
Perhaps the answer lies within the fundamentals of our intercooler testing methods. With this particular road test, we measure intercooler outlet and turbo outlet during a single third gear run from 3000 to 7000 rpm. This takes no longer than a few seconds with the intercooler never getting warm to the touch. While this may give a good indication of real-world intercooler performance, it leaves a little to be desired when it comes to sustained, wide-open throttle hot lapping on the track.
In fact, if we were to number crunch the results of the Medium intercooler's track testing (See Part VI), it's calculated efficiency would drop down to the 65 to 70 percent range. Of course, the stock intercooler, by comparison, would register a completely miserable 20 to 25 percent. We think these testing method-dependant differences in efficiency can partly be explained by the effects of heat soak, or in our case, the amount of heat getting trapped in the intercooler's core during prolonged periods of thermal loads. That is, the smaller (i.e. less massive) intercooler will get hotter when asked to absorb the same amount of heat.
Both these characteristics, as well as its larger internal volume, may suggest the Race intercooler could have some advantages when tested on the track. But one thing for sure, an extra 0.6 psi pressure loss is not a good thing. In fact, some would agrue that in order to offset the negative effects of just one extra pound of pressure loss, the intercooler would need to provide and additional 15 percent of efficiency--the main reason being the extra pressure loss would require that the turbos to spin harder to provide the same amount of manifold pressure. And, as we have all seen by now, this causes an increase in exhaust back-pressure which, in return, causes an increase in horsepower-robbing exhaust gas reversion. The only way to compensate for this loss of power would be to cool the air even further. Enough semi-educated pontificating--we'll leave the testing of that theory for the next installment.
But before we sign off, we feel it necessary to give honorable mention to a couple of products that, quite frankly, surprised the heck out of us. For the last several months, we have gotten spoiled by driving on a set of Toyo Proxes RA1 racing rubber. For a DOT-approved tire, they provide simply staggering grip, predictable break-away characteristics and reasonably low noise.
However, being an R-compound tire, they don't exactly exhibit the wear characteristics most consumers expect from a tire. Fortunately, our well-balanced Project RX-7 has proven to be surprisingly easy on its tires. In fact, we were able to log an astounding 8,000 miles of mixed street and track driving--simply unheard of in the world of racing rubber. But, of course, all good things must come to an end. With the RA-1s currently resembling a heat-cycled, bald racing slick and the fact that northern California's rainy season is now approaching, we realize the need for a change.
While we would be more than happy to get our hands on another set RA-1s and spend the next year wearing them to their cords, it was suggested we try a more conventional, longer-lasting (read: Non R-compound) subsitution. At first, the idea of sacrificing grip and response for a... [gasp]... longer-lasting street tire didn't sit well in our stomachs. How could we possibly justify a change that would surely result in compromised road-carving performance? After doing a good deal of research, we decided to go with Yokohama's new AVS Sports--partly because its unique tread design reminded us of an honest-to-God wet weather racing slick. But the biggest question on our mind was if Yokohama's flagship street tire would live up to the racy performance its looks suggest.
To find out, we mounted 235/45-17s up front and 255/40-17 in the rear--the same staggered sizing used on Mazda's Japan-only '99 RX-7. Even before reading the product information provided by Yokohama, we knew there was something different about these tires. Like real race tires, there was a noticeable absense of tread squirm--something one doesn't really notice until it's gone.
What makes these tires different? A good part of the answer is in its strange tread design. The massive center block, constructed of one single piece of road-contacting rubber, is designed to resist a tire's natural tendency to deform or "wiggle" under loads. Surprisingly, the AVS Sports' unique character was not only felt under hard, at-the-limit cornering, but also in gentle lane changing and braking. And contributing to the AVS Sports' unsually strong resistance to "roll" onto its shoulder during maximum cornering loads are its unique, large lateral tread blocks and precisely shaped casing. But the AVS Sports' level of performance isn't without its trade-offs, perceived or real.
For one, like real race tires, the Sports are more prone to follow grooves and other irregularities in the road--requiring a bit more attention and correction from the driver. And secondly, one could possibly interpret the Sports' quick and decisive responses to be a sign of a less-forgiving tire. However, we would agrue the later is not the case with a car as predictable and intuitive as our Project RX-7. In our situation, it's certainly an advantage to run on tires that respond immediately, than on ones that feel slighty indecisive. Not only does this make the car ultimately faster around a track, it also complements the very nature of the RX-7--even when driven casually on the street. While their ultimate grip is as good as, if not slightly better than, the other UHP street tires on the market, it is the Sports' unique subjective "feel" that has won a place in our hearts.
In addition to intercooler testing at Thunderhill, we will also resume our quest for more horsepower. And as if Project RX-7 wasn't already raced-out enough, tune in next time and take a peek at what could very well be the widest and stickiest set of Hoosier racing slicks this side of a World Challenge car. Stay tuned.
1100 S. Raymond Avenue, Suite H
Fullerton, CA 92831
C2 Automotive Inc.
1009 7th Street
Oakland, CA 94607
(510) 272-9869 (Tel)
(510) 272-9882 (FAX)
Kirk Racing Products
Mark Stewart Enterprises Inc.
1433 Montgomery Hwy
Vestavia, Alabama 35216
2111 Freemont Street
Concord, CA 94520
P.O. Box 15
Cypress Gardens, FL 33884-0015
Performance Friction Corporation
83 Carbon Metallic Hwy
Clover, SC 29710-0819
(800) 521-8874 (Tel)
(803) 222-2144 (FAX)
Superior Dyno Service
1740 Enterprise Dr. # 12
Fairfield, CA 94533
(707) 425-DYNO (Office)
(707) 425-6062 (Shop)
Toyo Tires (USA)
6415 Katella Avenue
Cypress, CA 90630
(800) 678-3250 (Phone)
(714) 229-6184 (Fax)
Yokohama Tire Corp.
P.O. Box 4550
Fullerton, CA 92834