Richard Holdener
November 1, 2004
When it comes time to modify your '03-04 Cobra, replacing the blower pulley ranks right up on the top of the performance to-do list.

In four previous issues of MM&FF (June-August 2004), we took a hard look at the performance modifications available for the two-valve 4.6 GT motors. Heck, we covered everything from simple throttle bodies and elbows (both NA and supercharged) to nitrous and blowers. Along the way we managed to sneak in ported cylinder heads, a variety of different cam profiles and even the effect of changes in compression ratio.

All in all, we'd say the "Mods for Mods" series offered pretty extensive performance coverage for the two-valve Modular V-8. Having nearly exhausted the available bolt-ons and buildups for the two-valve GT motors, it was only logical that we turn our attention to the more performance-oriented four-valve Cobra motors. As luck would have it (it is amazing what the right amount of begging and pleading can do), we managed to secure the mother of all four-valve 4.6 test motors--an '03 Cobra crate motor assembly. Talk about the ultimate four-valve test mule!

Thanks to Ford Racing, we had the ultimate 4.6 tester at our disposal in the form of the '03 Cobra crate motor. Rated at 390 hp and 390 lb-ft of torque, the supercharged four-valve produced not only impressive power in stock trim, but also had plenty of potential once we got more serious with the modifications.

The beauty of the Cobra engine is that in addition to every nut and bolt being brand-spanking new, it is factory equipped with a number of desirable features, not the least of which is a hefty chunk of forced induction in the form of an Eaton (positive displacement) supercharger. While the presence of the supercharger is cause enough for celebration, the real key to the success of the '03 Cobra plant from Ford Racing is that it was configured from the factory to accept the rigors of supercharging. This means desirable (for forced induction) low-compression, free-flowing four-valve heads and even an efficient air-to-water intercooler to help lower the inlet charge temperature to suppress detonation. Having seen blown '03-04 Cobras already exceed 700-wheel hp, naturally we were anxious to get started with the modifications. In future "Mods for 4V Mods," look for testing on cams, ported cylinder heads and even a four-way forced induction shootout, but for part one, we will stick with some basic bolt-ons, including tuning the air/fuel and timing curves, inlet and exhaust modifications and the always popular increase in boost pressure via changes to the drive ratio.

The first order of business was to get the 4.6 installed on the engine dyno. While '03-04 Cobras have been run extensively on the chassis dyno, not many supercharged four-valve motors have found their way onto the engine dyno. The assembly from Ford Racing was shipped complete, including the serpentine drive assemblies (both blower and accessory). Since we planned on running the F.A.S.T. stand-alone engine management system, we removed a number of factory components that were unnecessary: the clutch and pressure plate to facilitate mounting the motor, plumbed water to and from the air-to-water intercooler and capped the necessary coolant and vacuum lines.

Just as with the two-valve 4.6s, the F.A.S.T. engine management system allowed us to delete the factory mass-air meter and attending inlet tubing. All that was necessary was the stock throttle body and inlet into the blower, something we would be upgrading later during testing. Though the Ford Racing Cobra mill was originally equipped with a coil-on-plug ignition system, we replaced the factory ignition with the coil-pack system used previously on the 4.6 GT motor. This in no way changed the power potential; it simply allowed us to use an existing management system on the four-valve V-8.

Using the F.A.S.T. engine management system allowed us to tune the supercharged motor to extract the maximum (safe) amount of power from each combination. The F.A.S.T. system also ensured accuracy and repeatability by producing consistent air/fuel and timing curves.

After hooking up the F.A.S.T. system, we dialed in the factory air/fuel and timing curves, which were supplied by Kenne Bell. With all its chip expertise, it knew not only what the factory timing curve (at WOT) should look like, but what the force-fed Ford would tolerate on the 91-octane California pump swill we planned on running. The idea was to run with the factory timing values (using the F.A.S.T.) then tune for additional power. Initially, we planned on running the Ford Racing crate motor with the supplied serpentine belt driving all of the accessories. With no power steering reservoir or lines to hook up a simple loop system (to lubricate the pump while in operation), we eventually decided to forgo the accessories and install a Meziere electric water pump. This allowed us to eliminate all of the accessories (with the exception of the alternator) while retaining the drive system for the blower.

The Cobra crate motor was shipped with the factory (cast iron) exhaust manifolds. We planned on replacing these with a set of headers, but to test the power gains offered by the Hooker (Flow Tech) long-tube headers, we needed to hook up a free-flowing exhaust system. The dyno exhaust consisted of a pair of 2.5-inch tubes run with no mufflers. We did not want to restrict the exhaust system with cats or mufflers.

After replacing the stock Cobra injectors with a set of 65-pounders (the rating was fine for our intended power level, but the F.A.S.T. harness was configured to accept the early style injectors) we were ready to rock. On hand for testing on the '03 Cobra motor was an Accufab throttle body and inlet, a set of Flow Tech headers, the tuning from the F.A.S.T. system and an array of blower and crank pulleys from South Florida Pulley Headquarters to increase the boost pressure supplied by the Eaton supercharger.

The stock Cobra blower pulley was pressed onto the snout shaft of the Eaton supercharger and required a dedicated puller to remove.

The first order of business was to establish a baseline. As indicated, we installed the factory timing curve using the F.A.S.T. system and to our surprise, the supercharged crate motor thumped out some pretty impressive numbers. How does 483 hp and 427 lb-ft of torque sound? We all know that the Cobra motor is rated at 390 hp and 390 lb-ft of torque. Where did the extra 90 hp come from you ask? First off, we suspect the supercharged Cobra motors are underrated by Ford. Every one we have tested on the chassis dyno has produced much more than the 390hp flywheel rating would suggest. It is also important to understand that the rating from Ford is an SAE number derived with a full exhaust including cats, full inlet system (silencer, mass air and all inlet tubing) and full accessories.

Regardless of the rating, our test motor now produced 483 hp and 427 lb-ft of torque at a peak boost reading of 8.8 psi. As is the case with these supercharged Cobra motors, the boost curve provided by the Eaton supercharger dropped off with engine speed. This drop off in boost pressure would become much more severe as we increased the drive ratio on the blower.

After establishing a baseline, we decided that the first modification would be to tune the existing combination. Using the F.A.S.T. engine management system, Tom Habryzk from Westech dialed in a new timing curve that supplied 23 degrees of total timing everywhere (from 3,000 to 6,500 rpm). In stock trim, the WOT timing was as low as 13-14 degrees according to the gang at Kenne Bell. Increasing the ignition timing had a dramatic effect on the power curve. Since the supercharged engine was already equipped with 23 degrees of ignition timing from 6,000-6,500 rpm from the factory, there was very little difference in peak power but boy howdy did the power improve down low. The extra timing improved the torque production by as much as 40 lb-ft. No changes were made to the air/fuel mixture, which was kept constant at 11.8:1. We know that the factory air/fuel mixture can be considerably richer than this, having witnessed many Cobras running in the 10s on the chassis dyno. The difference would likely have been even greater had we elected to run the factory air/fuel curves as well on our baseline. No wonder the new Cobras respond so well to custom chips.

Lucky for us, Kenne Bell offered a puller designed to remove the press-fit blower pulley.

The next performance modification on our "Mods for 4V Mods" list was a set of 1 3/4-inch (primary) Flow Tech (Hooker) headers. Keeping the ignition timing at a steady 23 degrees and the air/fuel ratio at a constant 11.8:1, the long-tube headers and 18-inch collector extensions allowed our supercharged test mule to exceed 500 hp for the first time. The headers upped the peak power to 501 hp, while the torque peak was up slightly to 461 lb-ft. Interestingly enough, the boost pressure actually dropped with the installation of the headers. The reason for this is that the exhaust scavenging helped improve the efficiency of the engine. Improving the (pumping) efficiency of the motor will result in a decrease in boost (or back) pressure. Installation of (the right) cams, ported cylinder heads or increasing the displacement of the motor will all result in an increase in power combined with a drop in boost pressure. Please don't be mislead into thinking that we should now increase the boost pressure to compensate for the drop. The real test of the effectiveness of the headers is to install them with no other change.

After improving the flow rate of the exhaust system, we turned our attention to the inlet system. Since the stand-alone F.A.S.T. system allowed us to eliminate the air silencer, mass air and air cleaner inlet assembly, we were left with the stock throttle body and inlet tube connecting it to the supercharger. In retrospect, we should have measured the inlet vacuum present between the throttle stock body and the supercharger, but we were more interested in monitoring the boost pressure. In place of the stock dual-blade throttle body and cast aluminum inlet, we installed a billet aluminum, single-blade throttle body and free-flowing inlet system from Accufab. Like most pre-blower airflow improvements, installation of the Accufab throttle body increased the boost pressure slightly (.2 psi) and upped the peak power numbers from 501 to 508 hp, while the peak torque was up 2 lb-ft to 463. As expected, the power gains improved with engine speed. A later comparison between the stock and Accufab inlet systems (at a higher boost and power level) revealed gains of 10-12 hp.

The Kenne Bell pulley removal tool made removing the stock blower pulley a snap without fear of damage.

With improved intake and exhaust systems, it was time to crank up the boost pressure. The boost pressure on the factory Cobra is supplied by a Roots-style Eaton (positive displacement) supercharger. The blower is able to supply the motor with more air than it could ingest of its own accord in normally aspirated form. The extra airflow is seen as boost. The boost pressure is controlled by the speed of the supercharger relative to the engine. This speed is determined by the drive ratio or the size of the crank and blower pulleys. The stock pulleys measured 3.6-inches (supercharger) and 7.5 inches (crank). This pulley ratio produced a blower speed 2.08 times the engine speed, meaning that our Eaton supercharger was spinning over 13,500 rpm at our self-imposed engine redline of 6,500 rpm. (6,500 rpm x 2.08 = 13,541 rpm).

Given that additional power is just a pulley change away, it is not surprising that blower (and crank) pulleys are two of the lost popular upgrades among Cobra owners. Gains of 50 hp and 50 lb-ft (or more) are not uncommon with pulley changes, making them one of the best bang for the performance buck available.

Advanced Timing
The first test involved altering the factory timing curve. From the factory, the '03 Cobra motor featured a pretty conservative ignition timing curve. Down between 3,000-3,500 rpm, Ford equipped the motors with 14-15 degrees of total timing. We have seen even less on the chassis dyno if the temperatures are high. The Ford Racing crate motor was run with the factory timing curve to establish a baseline then again after revising the timing curve to produce 23 degrees of total timing throughout the rev range. The additional timing had a dramatic effect on the power curve. There was very little gain in peak power since the factory curve already featured 23 degrees of timing from 6,000-6,500 rpm. Check out the huge gains (40 lb-ft) down near 3,000 rpm.

While pulley changes are certainly popular, Ford has seen fit to make things somewhat difficult when it comes time to up the boost pressure. The stock Cobra pulleys are pressed onto the supercharger snout shaft, making removal somewhat more difficult than a simple unbolt-and-go procedure. Luckily for us, many aftermarket sources offer removal tools for the stock Cobra pulleys. Ours came from the guys at Kenne Bell and was a first class piece that included detailed instructions, all the necessary hardware and worked flawlessly.

After removing the stock Cobra blower pulley, we installed the new Dial Ur Boost (DUB) pulley system from South Florida Pulley Headquarters. Rather than a single (press on and/or off) blower pulley, the DUB system relied on a press-on hub adapter that allowed easy installation of different blower pulleys. Once the hub adapter was installed, pulley changes required only unbolting the four retaining bolts. This DUB system is perfect for a dual-purpose Cobra where boost can be kept at a reasonable level for the street, but quickly changed at the strip for some quick e.t.'s Like the pulley removal tool from Kenne Bell, the DUB system came with detailed instructions easily understood-even for the first timer.

With the 4.6 now equipped with the Flow Tech exhaust and Accufab throttle body, we swapped out the stock blower pulley for a 3.20-inch DUB pulley. According to the information provided by South Florida, the smaller 3.20 DUB pulley could be run with the stock blower belt. As expected, the 3.20-inch blower pulley increased the peak boost pressure by a hair over 2 psi (from 8.7 to 10.9 psi) and improved the power output significantly. The peak power now checked in at 530 hp, while the torque took a monster leap from 463 lb-ft to 506. As expected of a change in boost pressure, the horsepower gains were consistent throughout the rev range. The roughly parallel horsepower curves meant that the torque gains were more pronounced at lower engine speeds. The horsepower improvements seemed to reach a point of diminishing returns as the engine speed exceeded 6,000 rpm. Perhaps we were nearing the flow limit of the Eaton supercharger, since it was designed to feed a 390hp motor and not something easily exceeding 500 hp. With over 500 lb-ft of torque, this was one impressive 281 cubic-inch street motor and proof positive that it is tough to beat forced induction for making serious power.

When it comes to Cobra owners and boost, more is always better. While this philosophy can sometimes get owners into trouble (spelled D E T O N A T I O N), there is definitely more power to be had by going beyond 10.9 psi of boost. Knowing this, we utilized the DUB system once again and swapped out the 3.20-inch blower pulley for a smaller 2.93-inch pulley. The interchangeable DUB system made installation a snap--it took just a minute or so to perform the swap.

Stock TB vs. Accufab TB (8.5 psi)--Ford Racing '03 Cobra Motor
The F.A.S.T. system allowed us to eliminate everything on the inlet system in front of the stock throttle body. Thinking that the stock throttle body and inlet into the blower was a possible restriction, we installed an Accufab throttle body and inlet upgrade. The billet aluminum Accufab pieces upped the power peak to 508 hp, while peak torque increased by 2 lb-ft.

Combining the 2.93-inch blower pulley with the stock blower belt took the stock spring-loaded belt tensioner to the end of its available adjustment. Rather than run to the store for a new blower belt, we moved the alternator on its mount to effectively decrease the belt length. This put the tensioner back in the proper range for maximum belt tension. The 2.93-inch blower pulley upped the peak boost pressure from 10.9 to 13.3 psi. As another sure sign that we were nearing the flow limit of the Eaton supercharger, the peak power leveled off at 538 hp (a gain of just 8 hp), while the peak torque took another huge jump from 506 lb-ft to 550. Notice now that for the first time, the supercharged engine produced more peak torque than horsepower. Though the peak numbers seemed to be limited, the increased boost pressure sure improved the power output throughout the remainder of the curve.

The results of the previous two pulley changes seemed to indicate that we were limited in terms of maximum power potential, but this final test confirmed it. After running the pair of DUB blower pulleys, we decided to change the crank pulley. There is a limit to the available reduction in blower pulley size, as a dramatic reduction will reduce the available surface (belt) contact, the mechanical leverage and belt wrap, all which increase the possibility of supercharger belt slippage. Lucky for us, installing a larger crank pulley has the same effect as reducing the size of the blower pulley. Knowing this, we installed a 7.75-inch crank pulley system from South Florida Pulley Headquarters.

Like the DUB blower pulley system, the 7.75-inch crank pulley system eliminated the factory crank pulley with an adjustable hub that facilitated the installation of different size crank pulleys to increase the boost pressure. The larger crank pulley increased the peak boost pressure from 13.3 to 14.5 psi. The gains were not as impressive as the previous attempts, as the peak power was up by just 1 hp, while the peak torque now stood at 569 lb-ft. The pulley changes didn't seem to be offering anything other than increased low and mid-range torque so, while we had a number of other (larger) crank pulleys at our disposal, we decided to call it a day with our Ford Racing crate motor still in one piece.