Richard Holdener
July 1, 2006

Back in our "Mods for Four-Valve Mods" series, we experimented with a Kenne Bell supercharger to replace the factory Eaton M112 blower. In Part 2 of the "Mods" series, we managed to exceed 700 hp with the Kenne Bell supercharger on the otherwise stock '03 Cobra crate engine (supplied by Ford Racing).

In our subsequent "Boost Bash" series, we once again exceeded 700 hp with the '03 Cobra 4.6, but at a reduced boost level thanks to the introduction of a set of Comp XE262AH cam profiles.

If you have ever read any of our testing on forced induction, the power output of any supercharged motor is a function of the power output of the normally aspirated engine multiplied by the boost pressure. Thus, improving the power output of the normally aspirated combination can, and often does, yield larger power dividends once you add boost to the equation. Of course, it is imperative that the normally aspirated combination be able to withstand the extra power offered by the supercharger.

After reviewing the power gains offered by the Kenne Bell twin-screw supercharger, we began to wonder just how much power was available from a more serious supercharged Four-Valve combination. Having run the 4.6 up to 20 psi with a pair of turbos from HP Performance (to the tune of 990 hp), we decided it was high time to see just how much power was available from the Kenne Bell supercharger at a similar boost level. Having already exceeded 700 flywheel horsepower during the testing for "Mods for Four-Valve Mods," we knew we would easily surpass that power number. We have personally witnessed a modified '03 Cobra exceed 700 wheel horsepower on the Dynojet, so we know that translated to more than 700 flywheel horsepower. The question now was, how much more?

As luck would have it, the very same powerplant that exceeded 700 wheel horsepower (at 28 psi of boost) was used for this high-horsepower Four-Valve buildup. Of course, the engine was removed from the Cobra, and freshened up (and modified) in order to increase the power potential, but it was nice to know that we were working with a familiar face.

The proper route to more power is a bit more complicated than simply cranking up the boost. When this 4.6 exceeded 700 wheel horsepower, it was equipped with a 9.5-inch crank pulley and a 3.39-inch blower pulley, giving us a drive ratio of 2.80:1. Running 6,500 rpm, this pushed the blower speed over 18,000 rpm. According to Kenne Bell, there was a little more blower speed available, but both the blower speed and boost levels (28 psi) were nearing the danger zone. The ideal situation would be to increase the power output while simultaneously reducing the boost pressure.

Wanting not only more power but more reliable power, the '03 4.6 was taken to Accufab's John Mihovetz. No stranger to huge modular power, he took on the buildup and allowed us to follow along during the assembly and subsequent dyno thrash. Naturally, the dyno testing involved more than just the set-it-and-forget-it tuning session, as Mihovetz tried a number of different trick components (many custom machined) in his quest for serious Kenne Bell power. Before he could get to the induction, intercooling and exhaust tricks, he had to build a suitable long-block. He took the stock steel Cobra crank and added a set of forged connecting rods and forged pistons, and stuffed them inside a race-prepped aluminum Cobra block. The forged pistons deserve special mention as they not only featured crown and skirt coatings, but were flat-top slugs (with valve reliefs), which upped the static compression to 10.0:1. The pistons were equipped with stainless steel top rings to withstand the extreme conditions generated by the blower motor.

0607mmfp_01z 900hp_kenne_bell_supercharger John_mihovetz
Any time John Mihovetz gets involved in a modular project, you can expect big power numbers.
0607mmfp_02z 900hp_kenne_bell_supercharger Pistons
The buildup started out with a fresh set of forged pistons. The flat-top pistons (coated on both the skirt and crown) upped the static compression ratio from 8.5:1 to an even 10.0:1. These are the same pistons used by Mihovetz on his 1,500-plus-horsepower, twin-turbo mod motor.
0607mmfp_03z 900hp_kenne_bell_supercharger I_beam
The factory H-beam rods were ditched in favor of a set of stronger I-beam rods. Naturally, the factory never intended the 4.6 to produce anywhere near 900 hp.
0607mmfp_04z 900hp_kenne_bell_supercharger Crank
The new forged rods and pistons required the factory steel crank to be rebalanced (note the weld marks on the counterweights).
0607mmfp_05z 900hp_kenne_bell_supercharger Block
The forged reciprocating assembly was stuffed into a fully prepped aluminum Cobra block. The block was also treated to a main stud girdle to further improve bottom end strength. Mihovetz even went to the considerable trouble to source custom ARP cross bolts for the buildup.

Increased compression was the first step toward improving the power output of the normally aspirated combination. Improving the NA power would actually reduce the boost pressure supplied to the engine (at any given drive ratio). The elevated compression ratio was combined with slightly wilder cam timing in the form of a set of Comp XE262AH cams. The Comp cams offered 0.425 lift and a 226/22 duration split that favored the intake side. Having tested these cams before in our "Mods for Four-Valve Mods" series (and this author's new book on modular performance), they work well on the Terminator engines despite not having more exhaust duration as is typical of blower cams. To make the most effective use of the newfound lift and duration, Mihovetz worked some of his magic on the Four-Valve cylinder heads as well.

Naturally, there was some minor porting involved, but according to Mihovetz, the real key is to have the valve guides properly sized, the valve job spot on, and the all-important valve height/lash adjuster relationship optimized. These lessons (along with a few we were sworn to secrecy over) were learned over a decade of campaigning modular motors. Even without the porting, the tricks performed to the Four-Valve heads can yield 30-50 hp on his wild turbo combination. Though the gains may be less on this somewhat milder supercharged motor, every bit helps when you are looking to maximize the power output of any given combination.

While testing for our "Mods for Four-Valve Mods series," we ran the 2.2L blower from Kenne Bell. With the larger 2.4L blower available, we naturally stepped up to this configuration to maximize the airflow potential of the blower. The 2.4L twin-screw supercharger offered a 9-percent increase in displacement, and therefore increased the flow potential over the smaller 2.2L blower. It was fed by an Accufab oval throttle body and the Kenne Bell blower inlet.

This last bit of info is important as we tested a modified version of the blower inlet with excellent results. In the car, the engine was equipped with a set of Kooks 151/48-inch long-tube headers featuring 2.5-inch collectors. Mihovetz suspected these might be a tad on the small side for this application, so we had a set of Hooker headers (same 151/48-inch primaries but with 3-inch collectors) and a set of (171/48- to 2-inch) race headers from his early supercharged race motor ready to go.

The supercharged 4.6 was also equipped with a Dial-Ur-Boost (DUB) crank pulley assembly from South Florida Pulley Headquarters. Naturally, the Kenne Bell blower pulleys were also inter-changeable, allowing us to dial in the blower speed and attending boost pressure at will. The blower assembly also included the factory '03 Cobra air-to-water intercooler assembly. For dyno testing, we relied on the dyno water for the intercooler, but stepped up to ice water for the last runs on this combination.

The Four-Valve 4.6 was installed on the dyno in normally aspirated trim to allow for proper break-in. The motor was filled with Lucas conventional (non-synthetic) 5W-30 oil and equipped with 36-pound injectors. The FAST engine management system was used to dial in the air/fuel mixture, and we were eventually rewarded with over 400 hp and nearly as much torque.

These power numbers were actually down quite a bit from what was expected and there was evidence of blow-by. It should be noted that the engine was subjected to a minor mishap during the initial break-in procedure. An air pocket kept water from circulating in the block, allowing the Four-Valve motor to get hot enough to boil the water in the block. All the tempo gauges (both on the dyno and in the FAST management system) showed the water temp to be, if anything, on the low side. The problem was that they were reading air and not water. Did the overheating torch the rings, or were the stainless rings simply reluctant to break in? Concerned about the condition of the freshly built motor, we pressed on with our testing after a compression test revealed that all eight holes were still present and accounted for.

0607mmfp_06z 900hp_kenne_bell_supercharger Chart_1
Na vs. Kenne Bell Supercharged 4.6L four-valve Motor (9.2 psi and 19.9 psi) - We began this adventure by running the modified '03 Cobra 4.6 in normally aspirated trim. This was done both to establish a baseline and allow some break-in time for the rings and bearings. The NA version produced 416 hp and 393 lb-ft of torque. While this was less than expected, we installed the Kenne Bell blower and ran it at 9.2 psi where the engine produced 570 hp and 488 lb-ft of torque. Stepping things up to 19.9 psi with a 3.75-inch blower pulley brought the peak power to 767 hp and 698 lb-ft of torque.
0607mmfp_07z 900hp_kenne_bell_supercharger Chart_2
Stock (KB) vs. Modified Blower Inlet - Suspecting that the inlet supplied with the Kenne Bell supercharger kit may become a flow restriction at these elevated power levels, Mihovetz modified one of the aluminum intakes to increase the flow rate. After porting and welding, the revised intake (from the throttle body to the inlet of the blower) improved the power significantly, upping the peak numbers to 800 hp and 704 lb-ft of torque. Note that most of the power gains offered by the revised inlet occurred at the top of the rev range.
0607mmfp_08z 900hp_kenne_bell_supercharger Heads
The '03 Cobra heads received minor porting to improve the flow potential.
0607mmfp_09z 900hp_kenne_bell_supercharger Serdi
A SerDI machine was used to produce a concentric valve job (critical for proper valve sealing). The heads also received new guides to eliminate run-out that will also hurt valve seal.

Step By Step

View Photo Gallery
0607mmfp_10z 900hp_kenne_bell_supercharger Chart_3
Headers: 1 5/8-inch vs. 1 7/8- to 2 1/8-inch Stepped Race Headers
- With the intake side improved, we turned our attention to the exhaust. Mihovetz had a set of 1 7/8- to 2 1/8-inch step headers for his old supercharged race motor, which were installed in order to improve the flow rate of the exhaust system. The headers offered a significant power jump, upping the peak power from 800 hp at 6,500 rpm to 852 hp at 6,800 rpm. The peak torque was up to 740 lb-ft at 5,300 rpm. Note that we were now running to 6,900 rpm.
0607mmfp_11z 900hp_kenne_bell_supercharger Chart_4
Intercooler Upgrades, Pulley Change, and Ice Water
- Since the factory Cobra intercooler was never designed to run these elevated power and flow levels, Mihovetz made some modifications to the entry into the intercooler core. He even went to the trouble of modifying the system to accept a larger air-to-water core from the 5.4-liter Ford GT. When combined with a change in blower pulley size from 3.75 inches to 3.39 inches, the combined effect was a jump in power to 902 hp and 813 lb-ft of torque. The final mod was to replace the ambient (dyno) water running through the intercooler core with ice water. This brought the power up to 920 hp and 841 lb-ft of torque, making it the most powerful Kenne Bell-blown Cobra engine in existence.
0607mmfp_12z 900hp_kenne_bell_supercharger Intake
We selected this '01 Cobra intake and throttle body for our NA dyno test before subjecting the powerplant to boost from the Kenne Bell blower.
0607mmfp_13z 900hp_kenne_bell_supercharger Engine_block
The 4.6 was run first in normally aspirated trim to provide time for a proper break-in procedure. It exceeded 400 hp and nearly as much torque in NA trim. Though down from the expected power output, we decided to press on regardless.
0607mmfp_14z 900hp_kenne_bell_supercharger Autorotor
The boost for our buildup came courtesy of this 2.4L twin-screw Autorotor blower from Kenne Bell. It offered plenty of flow, and bolted right onto the factory '03 air-to-water intercooler assembly.

The first step was to install the Kenne Bell intercooled blower. We installed it along with the 61-pound Siemens Deka injectors (used in the car). The blower was set up with a 9.5-inch DUB crank pulley and a 4-inch blower pulley. Initial runs were made with the compressor bypass valve open to reduce the total boost pressure. Running a peak boost pressure of 9.2 psi, it produced 570 hp and 488 lb-ft of torque.

After dialing in the air/fuel and timing curves at the lower boost level, we closed the bypass valve and allowed every bit of the boost supplied by the blower to enter the motor. Running a peak boost pressure of 17.6 psi, the 4.6 produced 746 hp and 664 lb-ft of torque. Stepping down in blower pulley size from the 4.0-inch pulley to a smaller 3.75-inch pulley (using the same 9.5-inch crank pulley) resulted in a jump in boost pressure to 19.9 psi. The result of the pulley change was an increase in power from 746 hp to 767 hp. The peak torque was up from 664 lb-ft to 698.

Obviously, increasing the blower speed improved the power, but we knew there was a limit to the gains we could expect from more boost. You simply can't keep cranking up the boost. Looking elsewhere for additional power, we turned our attention to the intake and exhaust of the supercharged combination. First up was the intake. Since we were running without any MAF or attending air intake assembly, we looked at the throttle body and intake manifold between the throttle body and blower. Since the blower manifold was already equipped with an Accufab single-blade oval throttle body, Mihovetz decided to try a modified version of the blower manifold.

The revised intake featured cutting, welding, and porting, but obviously the power gains indicated the effort was worthwhile. The new intake upped the power output of the super-charged combination from 767 hp to an even 800 hp. As expected, the majority of the power gains were near the top of the rev range. This is indicated by a jump in peak torque of only 6 lb-ft (from 698 lb- ft to 704). Out near 6,500 rpm, the power gains offered by the revised intake exceeded 40 hp.

With the intake improved, we turned our attention to the exhaust. In the "Mods for Four-Valve Mods" series, we tested a set of 151/48-inch headers against a set of 131/44-inch headers on a Kenne Bell supercharged engine producing near 650 hp. At that power level, the smaller 151/48-inch headers offered the best overall power curve. With this combo exceeding that power level by some 150 hp, Mihovetz was sure the larger headers would be worth some additional power. As luck would have it, he had a set of headers from his old supercharged race car. The headers featured 171/48- to 211/48-inch step primaries and large 4-inch collectors. These race headers represented a significant step up from the 151/48-inch headers (with 2.5-inch collectors) we had been running previously.

The headers offered a significant step up in power, bringing the peak power numbers to 852 hp and 740 lb-ft of torque. Obviously, the 151/48-inch street headers represented a restriction on this high-horsepower supercharged combination. Before all you Cobra owners rush out to buy huge headers for your street cars, know that these headers will probably not fit in the chassis, and that most street applications (at least below 650 hp) will run better with the smaller primary tubes.

With the beginning and end of the airflow taken care of, we took a look at the middle portion, namely the intercooler. Suspecting the factory intercooler core was a restriction, Mihovetz modified the intermediate plate to change the angle of the air entering the core. He also modified the system to allow the use of the larger air-to-water core from the supercharged 5.4 Ford GT.

The drop in pulley size from the 3.75-inch used in all the previous testing to a 3.39-inch pulley combined with the intercooler upgrades resulted in an extra 50 hp, from 852 hp to 902 hp at 6,800 rpm. Torque production now exceeded 800 lb-ft, with a peak of 813 lb-ft at 5,000 rpm.

The final step was to replace the ambient dyno water run through the intercooler core with ice water (something that should be employed at the dragstrip at this elevated boost level). With discharge temps coming out of the blower exceeding 335 degrees (before the intercooler), the ice water offered a much-needed drop in inlet temps. The ice water proved to be a worthwhile modification, as the peak power jumped to 920 hp and 841 lb-ft of torque. Interestingly enough, use of the ice water dropped the boost pressure nearly 2 psi (from 24 psi to 22 psi). Given that this engine originally produced 720 wheel horsepower at 28 psi and was now producing 920 flywheel horsepower at just 22 psi, we'd say the exercise was a success.

Photo Gallery

View Photo Gallery