Project Celica: Part 2

The stock intake box is located directly behind the radiator, but despite what seems like a very hot location, it draws cooler outside air from the small grille in the center of the bumper.

The Rod Millen Motorsports intake places a cone filter in the open space shown here. Unfortunately, this happens to be directly behind the radiator. An optional heat shield keeps the hot radiator air from entering the filter. Our tests show this to be a very effective solution.

The Celica's mass airflow sensor consists of this simple sensor element that drops into a calibrated section of the air filter box. On the Rod Millen Motorsport intake, the sensor is moved to a similar section of the RMM intake pipe just before the throttle body.

The mass airflow sensor element measures the amount of air that passes through the hole in the base of the sensor, and the ECU translates this into total airlow based on the overall size of the intake pipe. Change the size or flow characteristics of this section of the pipe and your sensor readings will be wrong!

Fully installed, the Rod Millen Motorsports intake is compact, simple and surprisingly effective.

The stock exhaust's unusual configuration is designed to reduce the car's polar moment by moving the heavy muffler closer to the center of gravity. It also makes aftermarket exhausts an exercise in creativity. A bolt-on cat-back exhaust is actually impossible, since the bottom section of the header, the catalytic converter, and the first silencer are all one piece. The first flange is at the rear suspension. The front section of the exhaust appears to be large enough anyway.

The Rod Millen exhaust uses a straight-through design and show-quality, stainless-steel construction. After a little use, of course, that shiny stainess steel is a darker bronze color, but as long as the pipe is very clean before it gets heated (wipe off your fingerprints) it still looks good.

The HKS exhaust uses a chambered muffler and a secondary muffler at the tailpipe. Combined with the catalytic converter and stock pre-silencer, it has a very subdued exhaust note. This one should never get you in trouble with the law.

Though you will need a shop with a proper spring compressor to remove the stock springs, we still feel compelled to point out that the alignment of the spring perch, rubber bellows, and spring should be marked before you disassemble everything. It doesn't look nearly as obvious when you put it back together.

Unlike many modern cars, the bump stop and dust boot are separate, allowing us to trim the bump stop without rendering the dust boot useless.

Modern bump stops are carefully calibrated parts of the suspension designed to control extreme suspension movements seamlessly. Once you lower the suspension, however, the bump stop is active far too often, and must be trimmed to allow the spring and shock to work properly.

How much to cut off the bump stop is a guess, but this is the part of the front bump stop that we used, and it works quite well.

Upon reassembly, we noticed the upper strut mounting holes were quite large. Showroom stock racers always take advantage of loose tolerances like this to get as much camber as possible. When tightening the strut top bolts, we made sure the strut was as far toward the center as possible. The difference in camber is minimal, but not as minimal as the effort.

Access to the rear upper shock mounting bolts is far from simple. The plastic tray at front of the rear hatch area must first be removed (it snaps out if you pull straight up), then the side panels must be removed to access the three bolts (two bolts are shown here and one is in front of the shock).

After unbolting the bottom of the shock, it should drop right out... but it doesn't. Unbolting the camber control link and anti-roll bar end link allows the suspension to drop farther, making removal easier. Later, the technicians at the Progress Group managed to remove a shock without unbolting all these links, but this still seems to be the easier technique.

This is why the shock is so difficult to remove: It is actually much taller than it appears.

Again, the rear bump stops are carefully calibrated, and again, we hacked them up, leaving only this much.

As of now, this is the only set of Enkei LMF-1 wheels in the country. It shouldn't be. At only 15.7 lbs for a 17 x 7-inch wheel, the LMF-1 is exceptionally light, and its simple good looks have caused heads to turn ever since we bolted them on. They should be available by the time you read this, but if not, call Enkei (its number is in the source list at the end of this story) and start bugging them. Despite the fact that the wheel wells seem to be able to accommodate wider tires, we fit 215/40ZR17 Bridgestone Potenza S-02s. While a 225 section width would probably fit, we couldn't find such a wide 17-inch tire with the correct rolling diameter.

The stock Celica GT-S is hardly weak in the brake department, but who are we to say big brakes are big enough. AEM's big rotor upgrade will allow us to stop harder more often, without fade and with better pedal feel and modulation. We don't need it, we want it.

To clear the larger rotors, the front dust shields must be bent back slightly. Alternately, you can remove the shields altogether, but since we must eventually return this car to Toyota, and removing the dust shields involved destroying them, we left them in place. In addition, some road racing readers have pointed out that the brake dust shield does isolate the ball joint from brake heat, a function we had previously overlooked.

The rear disc incorporates a small brake drum for the parking brake. This makes the rear caliper cheaper and simpler, but more importantly, it helps prevent the rotor from warping when the parking brake is set. The AEM rotor, of course, incorporates this drum.

The rear parking brake mechanism is surprisingly complex.

Because the integral drum eliminates the aluminum hat AEM usually uses to mount their rotors, the rear brakes have much more of a pie plate appearance than the fronts. They are, in fact, very similar in diameter, though the rear rotor is a much thinner, non-vented design.

Last month's introduction of Project Celica GT-S may have been a shocker for regular SCC readers, but not for Celica GT-S owners. Last month, we admitted that we actually kept our Celica stock for the first 9,000 miles. That may well be a record for an SCC project car, but the new Celica is so much fun to drive, we were able to resist the uncontrollable shaking in our wrench hands that comes whenever we're behind the wheel of something stock.

That all changed this month.

We have finally cracked the seal on our Celica, making it look better, go better, stop better and turn better. With its lower stance from a set of Progress springs and the stunning new Enkei LMF-1 wheels, our Celica has really started turning heads. Even among those wholly indifferent to cars, our Celica has induced jaw-dropped exclamations of approval. Of course, head-turning good looks wasn't really our goal, we were just lucky that Toyota's Calty Design Center penned such a stunner.

While we like to think of our project cars as examples of logical, progressive tuning, we have to admit we have started out with a rather haphazard assortment of modifications. This is a result of the car's relative newness. Very little is available for the Celica at this time, so we are basically trying whatever we can get our hands on. A few more months into the project and perhaps we will have a better idea of the most effective upgrade path. For now, we've broken down our multi-flanked attack into several different sections for your perusal.

Cold Air Induction
It all started with a bench racing session at Rod Millen Motorsports where we were ostensibly getting a preview of the Lancer EVO VI that Rhys Millen was soon to take up Pikes Peak in yet another record-breaking run. Rhys let it slip that he was working on his own Project Celica, and it was all downhill from there. An intake and exhaust had proven quite effective on his car, combining to add nearly 15 hp. That sounded good to us; next thing we knew, we were headed home in a Celica full of boxes, ready to begin our own tests.

First was the intake. When we saw our first pre-production Celica exactly one year ago, we pointed toward the intake as the most probable first step for easy horsepower gains. For an engine making 180 hp, the air filter was surprisingly small, and to make matters worse, there was an alarmingly restrictive-looking flap door before the air filter to block induction noise from the sensitive ears of Toyota's NVH engineers.

Last month we tried removing the flap door assembly and thought we felt a significant improvement in both power and drivability, but subsequently found no power on the dyno. We then tried swapping the flap door in and out without telling the driver and found that nobody could really tell the difference. Our conclusion? Taking out the flap door doesn't hurt, and may or may not improve drivability.

Enough about that. The biggest obstacle we saw in the intake was the fact the mass airflow meter was integrated into the air filter box, meaning any aftermarket intake would have to integrate the airflow meter as well. This can be more difficult than it seems. The mass airflow meter measures the amount of air entering the engine by effectively measuring the air flowing through a small orifice about a quarter inch in diameter. The total airflow is calculated based on the relationship between the air flowing through that little orifice and the air flowing around it in the main body of the airflow meter. If you change the size or airflow characteristics of the airflow meter body, that relationship will change, and the signal entering the ECU will no longer be accurate.

Normally, the airflow meter body is a separate piece that can easily be incorporated into an aftermarket system, but since the body of the sensor in the Celica is integrated into the air filter box, a carefully calibrated replacement body must be part of the intake system. The Rod Millen Motorsports intake appears to be nothing but a simple u-shaped piece of pipe, but the last section of the pipe before the throttle body is also doubling as the airflow meter body.

The simple design of the Rod Millen intake places the air filter directly behind the radiator, which would seem to be a bad idea. It isn't. Luckily, the filter is also right behind the small grille in the center of the bumper, so in addition to the hot air from the radiator, there is a source for cold air from the outside. To ensure the engine gets the cold air, a stainless steel heat shield separates the air filter from the radiator. This seemed like far too simple a solution, so we decided to put it to the test. Installing a thermocouple in the intake right before the throttle body, we measured intake air temperature at 40 mph, 70 mph, and at idle after 30 seconds and 60 seconds of heat soaking. To our surprise, not only did the heat shield do an effective job of keeping intake temperature cool, it was more effective than the stock intake! With only a 10-degree temperature rise over ambient outside air temperature, the Rod Millen Motorsports piece is actually a cold-air intake, despite its underhood location. Very impressive.

OK, so it's cold, but does it make power? We headed to our new Dynojet to find out. After a few pulls, the answer was a resounding yes... with a caveat. From 4000 rpm on, the intake made a substantial difference, adding as much as 9 hp to the Celica's already stellar high-rpm powerband. Below 4000 rpm, however, the Celica lost as much as 9 hp. This was a shocker, as we had already driven the car and didn't really notice the loss at low rpm. Are we asleep at the wheel? Not really.

To be driven fast, the Celica demands revs. To be really fast, you have to stay above 6000 rpm, but the gearing makes that impossible. Still, hard driving is all done at 5000 rpm and above. Anywhere below 5000 rpm full throttle is an exercise in futility, so we seldom bother. With the intake, the engine did seem to demand a little more throttle at low rpm, but with this particular engine, a loss of bottom end power simply isn't important.

In addition to the power, the engine note with the intake is spectacular. There has been an ongoing debate around the SCC garage as to whether Honda's B18 or Toyota's 2ZZ-GE sounds better. With the Toyota's authoritative new voice, the debate is over.

Installation of the intake was quite simple, and in fact, over the course of our testing, we swapped between the stock intake and the Rod Millen Intake several times. When you install it for the first time (and hopefully the only time), do make sure to read the instructions. There are several vacuum switches mounted on the air filter box that must be relocated to a bracket supplied in the kit, and their proper location is not necessarily obvious if you don't follow the directions. Other than that, everything fits perfectly and is well made. Installation should take no longer than an hour.

Exhausting the Exhaust
Our initial success with the intake made us expect similar success with the exhaust. While the stock intake looked restrictive, the stock exhaust did not, so perhaps similar power gains would be too much to expect. Still, Rod Millen Motorsports claims 8 hp from its intake, and we saw 9 hp, so we expected the exhaust to be as successful as they claim as well.

Like so much else on the car, the Celica's exhaust is unconventional. The exhuast manifold is basically a tri-y header, but the secondary pipes have been merged in the form of a single large pipe with a divider wall down the middle. This design reduces the surface area of the exhaust, allowing more heat to get to the catalytic converter to speed lightoff. This secondary section, the catalytic converter, the first muffler and the exhaust pipe all the way to the rear suspension is all one piece. If you want to bolt on an exhaust, you have to start at the back. This isn't as big a problem as it first appears. None of this section of the exhaust appears to be restrictive, so a simple muffler replacement could well milk all the potential benefit from the exhaust.

Did we say simple muffler replacement? You wouldn't call it simple if you had to design it. The Celica's muffler design is quite unconventional, leaving room for creativity when devising a replacement. The Rod Millen muffler is a simple, straight-through design constructed entirely of stainless steel. Installation is indeed simple--only two bolts and three rubber hangers have to be disconnected to drop the stock exhaust, and four bolts and the same hangers to re-install it. Good thing too, because we ended up changing the exhaust more than five times over the course of two weeks.

What happened? First, our dyno tests showed virtually no difference between the stock exhaust and the Rod Millen exhaust. Doubting the validity of our own test, we put the stock intake and exhaust back on and re-ran our baseline run, then re-installed the exhaust and tested it alone, without the intake. Again, no gains. Just about then, HKS called saying its new Celica exhaust was done. We loaded up all our exhausts and went to HKS' Dynojet where we again tested the Rod Millen Exhaust, the stock exhaust, and the HKS Hiper Muffler and found no meaningful difference between the three. Sitting right next to our car was the GT-S that HKS had tested and found 4 hp.

Reviewing its dyno charts, we noticed that while we were making 162 hp with the stock exhaust, HKS' test car started out making just over 150 hp. Digging a little deeper, we found that Rod Millen Motorsports' test car also started out lower than ours, making about 156 hp in stock form. Furthermore, AEM had tested two cars, a stock one that made 155 hp, and one of the cars used in the celebrity race at the Long Beach Grand prix which made 165 hp with a straight pipe replacing the muffler.

That means our car, in stock form makes as much power as a car with a straight exhaust. Our car made between 162 and 165 hp on three different dynos on four different occasions, so our numbers are no fluke. What's the story? Perhaps we just have a good engine, one in which the tolerances just stacked up in our favor, but the fact that changing the intake made exactly the change we expected and changing the exhaust did not made us suspect that the stock exhaust was the culprit. To test that theory, we installed the stock exhaust that originally came on Rod Millen Motorsport's car and returned to the dyno. Returning to our original wheels and tires to maintain the same rolling resistance and rotational inertia as our previous tests, we again found no difference between RMM's stock exhaust and our stock exhaust.

For now, we are baffled. HKS' data, RMM's data and even AEM's data on the Toyota celebrity car all suggest the exhaust is indeed worth some horsepower. Because of the unexpected and unusual results on our car, we are taking the unprecedented step of presenting the manufacturer's test data. We suspect their data may be more representative of what a typical Celica should see (see page 178).

We tested both the HKS and Rod Millen exhausts and both make essentially the same power as far as we can tell. So what else did we learn? The Rod Millen exhaust has a crisp, clear, very pleasing exhaust note that sounds incredible echoing off canyon walls or the flat slabbed body panels of Mustangs. The HKS exhaust has a deeper, quieter exhaust note that may be more pleasant for long trips, or more appropriate for sneaking out of your driveway at night. With two mufflers to the Rod Millen's one, the HKS exhaust is the choice if you are concerned about excessive noise, either for legal, or personal sanity reasons. The Rod Millen exahust is still far from loud (our Project 240SX and Project Civic Si were both louder), and the sharp, clear Toyota wail is one we can't get enough of, so the Rod Millen exhaust is probably the one that will stay on the car.

The Shocking Thing About Springs
Picking the right suspension for the Celica is a daunting task. A car that starts out with uninspired handling is easy to improve, but the Celica GT-S starts out as a dynamic gem. We could very easily make things worse with haphazard lowering, so our first instinct was to head straight for a fully developed race suspension. Before we even had the keys to our GT-S, Ground Control had already designed and tested a complete coil-over suspension for the Celica using its lightweight, double-adjustable Advance Design shocks. Though this is clearly the ultimate suspension, we thought that perhaps we should try something less expensive first.

With very few exceptions, lowering any car without adding stronger shocks is asking for trouble, but short of Ground Control's exotic dampers, there is still very little in the way of off-the-shelf performance shocks for the Celica as of this writing. Unfortunately, this is the nature of the shock absorber business. Shocks are made by big companies that move with glacial determination. Despite the fact there is a very active aftermarket waiting for performance shocks, most shock manufacturers are still using a replacement parts mentality. Why bother making a new shock when people won't be replacing their worn-out old shocks for several more years? Because our stiffer springs need stiffer shocks, that's why!

Astute followers of the Celica will point out that TRD has a complete suspension package including springs, shocks, and front and rear anti-roll bars. In fact, we tested a TRD Celica with that suspension in our April, '00 issue. The fact we have already tested it is the only reason we didn't put that suspension directly on our car. The TRD suspension performed well when we tested it (though it was hampered significantly by poor tire choice), but the whole reason we build these project cars is to test as many different new parts as possible, so we feel compelled to try something different. We will start out piece-meal, trying one part at a time to see what works and what doesn't. Don't be surprised if we end up with at least a few of the TRD suspension parts in here by the time we're through--we'll test them twice if we have to.

For now, we just started with springs. We chose Progress springs, both because of our positive experience with its suspension on our Project Civic Si, and because Progress is not secretive about spring rates. Some spring companies consider their spring rates to be proprietary, which can be quite annoying if you really want to know what you are getting. What good is knowing the spring rate? At worst, no good at all. At best, if you know what to look for, you can get an idea of whether the springs were designed for smooth ride, good handling, understeer, neutral cornering, whatever. Then you can decide which set of springs best fits your needs.

The Celica's front springs are conical, so the spring rate is not quite linear; in other words, the spring rate is different at different heights. Progress gave us the spring rate when the spring is compressed one inch more than it is at ride height. This is the position of most interest, since this is about where the outside springs will be during cornering. The Progress front springs are 180 lb/in, up from 136 for the stock springs. The Progress rear springs are 230 lb/in, up from 168 for the stock springs. What does this mean? This is 32 percent stiffer in the front and 36 percent stiffer in the rear. This is a significant enough bump in spring rate to suggest that responsiveness and handling were its goals. (Its reputation tells us the same.)

In a perfect world with stronger shocks, we would like to see even stiffer spring rates, but with the stock shocks, this is probably as stiff as we should go. The other thing we notice is that the rear has been stiffened up more than the front. This is critical on a front-wheel drive car, as it shifts more of the cornering load to the rear tires.

That's about all the speculation we can make based on numbers--time to try them on. Unless you have a spring compressor at home, you aren't likely to be swapping the springs yourself. Still, there were a few unusual points to the Celica's suspension that we thought should be pointed out, if for no other reason than to allow you to look like a smarty pants when you point them out to your installer. Peruse the installation photos here and consider yourself educated.

After settling the suspension, the car was lowered 1.4 inches in the front, and 1.3 inches in the rear. Front camber was only 0.8 degrees negative (we would prefer at least 1.5 degrees) and unlike most cars immediately after lowering, the front toe settings had not been disturbed--the car still had zero inches of toe in. The rear suspension is where an alignment is necessary. Rear camber is 2.2 degrees, and rear toe-in is 3/16 of an inch. The extreme camber and toe settings both will serve to keep the rear tires planted, promoting understeer. The trick to making a front-wheel drive car handle is to get the rear tires to slip somewhat, but to slip in a predictable manner. With these alignment settings, the rear is planted and understeer is the only cornering attitude available. This in spite of the stiffer spring rates in the rear. Camber and toe are both adjustable in the Celica's rear suspension, but we haven't corrected them yet, so we'll reserve judgement on the handling balance until we have.

As for ride quality and the ultimate "do the stock shocks work" question, we have the following answers: good and almost. The ride is on the very good end of the lowered car scale. In other words, the suspension soaks up bumps like it should, reacts more quickly than the stock suspension, but still gets out of shape over large bumps or even sudden pavement undulations. That is also the answer to the shock question. On larger bumps, or when a bump is encountered during hard cornering, the car will pogo off the bump stops. The rear is especially prone to unnecessary oscillation. Two things are needed here: more compression damping to help keep the suspension off the stops in the first place, and more rebound damping to absorb the energy of the stiffer springs and the inevitable bounce when the suspension is compressed too far. It should be noted that after 9,000 miles in completely stock form, the stock shocks softened up significantly. They are not so soft as to be called worn out, but they definitely do change as they break in. We have spoken with other Celica owners who have observed the same severe break-in. Given their current rate of wear, we expect them to get even softer now that the Progress springs are giving them such a workout.

Given our trepedation about running lower springs with stock shocks, we are pleasantly surprised with the effectiveness of the Progress springs alone. We are far from satisfied with the suspension in its current state, however. We expected worse, but we still want better. We need shocks...bad.

In the next installment, we'll have a better understanding of the car's new handling balance and hopefully some more suspension changes to make.

Wheels and Tires
Wheels and tires always seem like an easy choice to figure out, but nothing is as easy as it seems.

We'll start with wheels. We wanted 17-inch wheels for entirely cosmetic reasons. The 16-inch wheels that came on the car simply look too small, and though we have seen 18-inch and even 19-inch wheels on Celicas and have to admit they looked quite good, it is impossible to maintain anything close to the stock rolling diameter with such a big wheel. Rule #1 of Sport Compact Car projects: Nothing we do to the car can hinder performance. A taller tire means taller effective gearing, and that means a slower car. We can't have that. To even justify a 17-inch wheel, we had to find something exceptionally light, lest the extra inch add up to extra pounds. Our criteria for the perfect Celica wheel, then are the following: 17 x 7 to 17 x 8-inch size, less than 17 lbs, 5 x 100 bolt pattern, and close to the stock wheel's 39-mm offset. That narrowed down the choices considerably.

A few phone calls later, the perfect wheel appeared: Enkei's new LMF-1. Not yet available in the United States, the LMF-1 weighs only 15.7 lbs in a 17 x 7 size, and was available in a 32 mm offset, which happened to be close enough. With properly applied beg, plead and blackmail techniques, we were able to get a prototype set from Japan. Theoretically, this wheel will be available by the time you read this, but if not, we suggest you badger Enkei until it is.

Now, tires: The Celica has very accommodating wheel wells that look as if they would swallow a tire with a 225 section width. In fact, the TRD Celica we tested back in April did have 225/40ZR17 tires, but we avoided that option because a 225 tire needs to be mounted on an 8-inch wide wheel, not the 7-incher for which we had already settled. There is also very little tire selection available in that size. Instead we used a 215/40ZR17, a tire with a rolling diameter 0.3-inches smaller than the stock one. Scanning the tires available in that size, one quickly rose to the top. Bridgestone's Potenza S-02 Pole Position. We have driven many different cars with this tire (the stock S2000, King Motorsports Type R and CR-V and our own Project Impreza come to mind immediately) and have always been pleased with their grip, responsiveness and predictable breakaway. We have barely scraped the mould release compound off the tires at this point, so we'll save a more complete dissertation on the tires for after we have the suspension working properly.

Braking
If the Celica has a weakness, it certainly is not the brakes. With massive 11-inch rotors up front, and 10.5-inch rotors in the back, a stock GT-S can haul down from 60 mph in a scant 114 feet, besting even the big bindered Type R. The pedal feel when driving hard is also quite good, our only complaint being a slightly soft initial bite and a pedal that is spongier than we would like, even if it is better than the brake pedal on most stock cars. If the brakes are so good, why are we so eager to slap on AEM's massive rotors?

We could argue the cross drilled and slotted design of the rotors will help initial bite (which it does), or we could argue the larger rotors mean less pedal effort and consequently less pedal squish (also true), but in reality, we were suckered in by their good looks. In purely functional terms, the initial bite we are looking for should come from different brake pads, and the spongy pedal should be cured partially with stickier pads, and partially with less flexible braided steel brake lines. Both of those solutions are in the near future, but with the Celica still relatively new, neither was available as of this writing. Goodridge brake lines and EBC brake pads are on their way for the next installment. The true purpose of AEM's larger rotors is to increase ultimate braking power, improve fade resistance and make brake modulation easier. Or just to fill the backside of your new 17-inch wheels.

We are so vain.

When the pads and lines arrive, we will conduct a more extensive test, but for now, we tried them on just for fitment and some subjective pedal mashing. AEM's Celica rotors are very similar in concept to the AEM rotors we tested in the April 2000 issue. Basically, larger diameter rotors are used with the stock calipers by mounting the caliper farther out from axle centerline. With more leverage, the stock caliper is more effective, and the cost is far less than a complete rotor and caliper change.

The Celica brake kit is less complex to install than the Civic Si kit was. The Civic required careful re-routing and re-mounting of the brake lines, but the Celica's lines have enough slack to allow the calipers to be moved without worries. From a design perspective, however, the Celica did present some new challenges.

Traditionally, a standard size rotor is mounted to a custom-machined aluminum mounting hat. This design saves weight by using aluminum where there is no need for a hard steel friction surface, and also allows the same rotor to be used with multiple mounting hats for multiple applications. This technique could not be used the rear, however. To simplify the rear caliper design, and to help prevent warping when the parking brake is set on a car with hot rotors, Toyota uses a parking brake drum in the center of the rear brake rotor. While it is possible to incorporate this brake drum into an aluminum hat, a steel liner must be used. John Concialdi, AEM's chief parts making brain, decided a one-piece drum and rotor design would be best.

After a surprisingly simple installation, we put them to the test. As expected, initial bite was improved, as was the pedal feel. The relationship between braking force and pedal effort seems more linear with the AEM rotors than it did with the stock rotors, but again, this is something we would prefer to fix with pads. The pads, lines and rotors together should prove an unbeatable and relatively affordable combination. We'll find out soon.