Richard Holdener
July 1, 2009
What happens when we bolt a supercharger onto the already potent GT1000 motor? Are headers still a worthwhile upgrade on a supercharged application?

Having run our header test in the last issue on the normally aspirated version of the GT1000 motor, we were anxious to retest the motor with some boost. You will remember that the 5.4L GT1000 was built to officially reach the 1,000hp mark in supercharged trim--no easy feat, but one simplified by creating a much stronger normally aspirated combination to which boost would then be applied.

Knowing that the best supercharged combinations start out as powerful normally aspirated combinations, the 5.4L was built using an all-aluminum Ford GT block, augmented with forged internals. Wanting to maximize power, the static compression was purposely increased over the traditional 8:1 numbers run on the factory GT500 and Ford GT motor. The 10:1 short-block was topped with a set of CNC-ported GT500 heads and custom Comp cams. It was the cam timing that played havoc with our header test run on the normally aspirated combination, allowing us to improve the power output by over 100 hp with the installation of a set of American Racing headers. Would the power gains be even greater on the supercharged combination?

As with the test run on the normally aspirated 5.4L combination, we compared 1-inch, 1 7/8-inch, and 2.0-inch American Racing headers against the factory exhaust manifolds.

In part one, we covered two scavenging effects offered by long-tube headers. Using both the kinetic energy of outgoing gases and reflected pressure wave, (properly designed) long-tube headers provide low pressure in the combustion chamber during overlap. This helps both evacuate the residual exhaust gases, while simultaneously drawing in the intake charge.

Of course, these negative pressure waves must be timed properly to arrive at the combustion chamber when the piston is just past TDC on the end of the exhaust stroke. Short primary tubes (like those used on most factory modular exhaust manifolds) don't provide sufficient length for proper timing of the pressure waves, thus they offer no beneficial scavenging effect. In essence, they are simple flow devices that provide an escape route for the exhaust gases. In this way, the power potential of the stock exhaust manifolds is determined by their ultimate flow rate. By comparison, long-tube headers offered additional power not through additional flow capacity (though this can be the case as well), but primarily by artificially enhancing the breathing of the motor. Think of long-tube headers as a super extractor, sort of a mini supercharger for the exhaust system.

Our test motor was once again the 5.4L GT1000. The forged rotating assembly was stuffed inside an aluminum Ford GT block, and topped with a set of CNC-ported GT heads and custom Comp cams. Run with a modified version of the Cobra R intake (and headers), the 5.4L produced 500 hp in normally aspirated form.

The scavenging effect offered by headers is similar to the charge filling offered by long-runner intake manifolds. Know that our designations of long-runner intakes and long-tube headers should more accurately be designated optimized length, as the intake runner and exhaust primary lengths are tuned for the specific combination. We use the long-runner designation here only to differentiate them from short-runner intakes (like those used with the supercharger) and stock exhaust manifolds. The difference between the charge filling offered by optimized intake runners and the scavenging offered by long-tube headers is that the pressure waves are happening in reverse. On the intake side, the opening of the intake valve and downward movement of the piston creates a strong negative pressure wave that propagates outward to the plenum and is returned as a positive pressure wave to enhance cylinder filling. On the exhaust side, the open exhaust valve sends out a positive pressure wave that is returned as a negative pressure wave to help draw out residual exhaust gases and draw in the intake charge. Not surprisingly, the two pressure waves must work together to optimize power production, meaning the intake manifold and header must be sized to operate effectively in the same rpm range. This tuning effect of both the intake manifold and exhaust is offered despite the presence of boost. Boost from a supercharger or turbocharger does not minimize the effect of the ram effect offered by the intake, nor does it alter the scavenging effect offered by long-tube headers.

For the first test, the GT1000 5.4L was equipped with a stock GT500 supercharger. Obviously the motor was never going to produce 1,000 hp with the stock Roots blower, but it did allow us to run our header test at a more realistic power level.

With our understanding of header (and intake) theory now complete, we can take a look at our header test. Having run our three sets of American Racing headers on the normally aspirated GT1000 motor, it was time to apply the same test procedure to a supercharged combination. To facilitate this test, off came the modified Cobra R intake and on went the stock blower assembly from the GT500. The M122 supercharger was installed along with the factory air-to-water intercooler.

All testing was completed with dyno water running through the intercooler core. Actually, the intercooler received water after running it through the engine--not ideal, but pretty real world in terms of water temps. (We kept the water temps to the motor low for this reason). We took the liberty of installing a set of larger Siemens injectors since we planned to install a Kenne Bell blower and crank things up to reach our goal of 1,000 hp.

Since Ford blessed the GT500 with drive-by-wire throttle activation, we had to make our own mechanical version. Apparently the boys at Westech didn't take to my vice grips on the throttle shaft, so they machined this aluminum throttle lever. Not only did it look much better, it also didn't damage the throttle shaft.

The Eaton supercharger was run with the stock blower and crank pulleys, which produced a peak boost pressure of 10 psi (with the stock exhaust manifolds). So equipped, the supercharged 5.4L produced peak numbers of 630 hp at 6,500 rpm and 547 lb-ft at 5,100 rpm. Compared to the normally aspirated numbers with the stock exhaust manifolds, we increased the power output by nearly 73 percent with an increase in boost pressure of just 68 percent.

On the surface, it appeared that the stock exhaust manifolds were less affected by the wild cam timing in supercharged form, but the only way to know for sure was to install the first set of headers. On went the 1-inch stainless headers from American Racing, and once again, up went the power. Equipped with the long-tube headers, the supercharged GT1000 motor produced 655 hp at 6,500 rpm and 578 lb-ft at 4,800 rpm. It should be noted that these gains came despite a drop in boost from 10 psi to 7.2. Whenever you get an increase in power with a drop in boost, you know you are on the right track. The peak-to-peak difference between the stock manifolds and 1-inch headers was 25 hp, not the 100-plus hp witnessed on the normally aspirated combination. True to form, the scavenging effect of the long-tube headers improved the power output from 3,500 rpm all the way to 6,500 rpm, a sure indication that the extra power came from scavenging and not simple additional flow rate. Were the gains based solely on flow rate, we'd have seen the power gains increase with engine speed, meaning less of a gain down at 3,500 rpm than at 6,500 rpm. The second test run with the Kenne Bell at a much higher power level confirmed these gains were a result of the scavenging effect. Without the wild cam timing, these are the types of gains we'd expect to see on a normally aspirated mod motor as well, possibly slightly less, but consistent gains through the entire rev range.

The first order of business was to establish a baseline with the stock exhaust manifolds. After the results of the normally aspirated combination, we were anxious to see their response to this supercharged application.

When it was time to run the larger headers, things got really interesting. Installation of the larger 1 7/8-inch headers (with larger 3.5-inch collectors and extensions) improved the power output slightly to 658 hp at 6,500 rpm and 580 lb-ft at 4,800 rpm. Note that both long-tube headers reduced the peak torque engine speed compared to the stock exhaust manifolds (4,800 rpm versus 5,100 rpm). This trend continued with the larger 2.0-inch headers (same 3.5-inch collectors and extensions), as the supercharged motor produced 663 hp at 6,500 rpm and 583 lb-ft at 4,800 rpm.

Basically, the peak power numbers improved (albeit slightly) with each subsequent increase in header diameter. What was interesting was that unlike the normally aspirated combination, there was no loss in power elsewhere along the curve. The big-boy 2.0-inch header produced as much torque down at 3,500 rpm as the smallest 1-inch headers. The only difference between the headers was at the top of the rev range.

Next on the list was a set of 1 7/8-inch headers. Equipped with these American Racing headers, the supercharged 5.4L produced 658 hp and 580 lb-ft of torque.

Now, is this insensitivity to the different header configurations a function of the boost from the blower (less likely), or the significant change in runner length (more likely) going from the Cobra R intake (in normally aspirated trim) to the supercharger intake (almost no runner length)? The only way to find out is to run the header test again on the normally aspirated 5.4L with the short-runner blower intake and no blower, or to run the same test on the 5.4L equipped with a centrifugal supercharger and the long-runner Cobra R intake. That my friends will be a test for another day, but this test does reveal that headers are a worthwhile upgrade on your supercharged mod motor.

Not content to leave you with just one test, we decided to run the headers versus the stock manifold one more time at a higher boost and power level. Off came the stock M122 and on went the Kenne Bell 2.8L H-series blower. The Kenne Bell blower was installed with a Mammoth blower intake, a dual 75mm throttle body, and 3.25-inch blower pulley. Run with the stock exhaust manifolds and 2.5-inch exhaust, the supercharged 5.4L motor produced 753 hp at 6,500 rpm and 619 lb-ft at 5,400 rpm at a peak boost reading of 14.1 psi. Available dyno time did not allow us to run all three different headers on this configuration, but installation of the 2.0-inch headers resulted in a drop in boost pressure from 14.1 psi to 11.4 psi, and a jump in peak power to 788 hp at 6,500 rpm and 645 lb-ft at 5,400 rpm. It's interesting to note that the headers offered almost exactly the same power gains at this power level as they did with the stock blower (33 hp versus 35 hp). This illustrates once again that the gains are less about absolute flow and more about scavenging. It should be noted that the power curve was climbing rapidly at our shut-off point of 6,500 rpm, but we felt that was a safe redline for the long-stroke combination.

The final test run with the stock supercharger was with a set of custom 2.0-inch headers. The largest-diameter primary tubes produced the highest peak power numbers of 663 hp and 583 lb-ft of torque. The peak boost numbers registered with all of the headers was consistent at 7.2 psi.

With our header test complete, it was time for Dynatek's GT1000 motor to earn its name or die trying. What we needed was more boost, so off came the 3.25-inch blower pulley and on went the 2.75-inch version. This was coupled with the larger 7.8-inch crank pulley (up from 7.1 inches) from Innovators West. This provided a drive ratio right near 2.84:1. This same ratio can be achieved with the 2.5-inch blower pulley and the stock crank pulley, but decreasing the blower-pulley size increases the chance of belt slippage. To ensure we had adequate fuel delivery, the dyno fuel system was augmented with a Kenne Bell Boost-a-Pump; likewise for the ignition system. We employed a Kenne Bell Boost-a-Spark to amplify the voltage to the MSD coil packs. This combined with the Autolite race plugs (gapped down at 0.020) resulted in zero misfires. With a crankcase full of Lucas 5W-30 synthetic oil and a fuel tank full of Rocket Brand race fuel, Westech's Ernie Mena manned the FAST controls at a peak boost pressure just north of 23 psi. Even running conservative air/fuel and timing values, the peak numbers easily exceeded 1,000 hp, the best being 1,036 hp with a new Street Rod induction system in place of the Mammoth and dual 75mm throttle body. With over 1,000 hp and 865 lb-ft of torque, the supercharged 5.4L was now officially the GT1000!

Since time was in short supply, we stepped right up to the 2.0-inch headers (we wanted them for our shot at 1,000 hp). Equipped with the 2.0-inch headers, the boost dropped to 11.4 psi, while the peak numbers jumped to 788 hp and 645 lb-ft of torque.

Unlike the test run on the normally aspirated combination, the 2.0-inch header performed best on the supercharged combination. In fact, the largest of the three headers not only offered the highest peak power, but there was no trade in power elsewhere along the curve. Had we elected to rev the motor to 7,000 rpm, the differences between the sizes would have been even greater. It's obvious from the graph comparing the stock exhaust manifolds to the long-tube header that the motor responded to the scavenging effect of the tuned header length. Less evident is the fact that it may have been the short-runner intake even more so than the boost supplied by the supercharger that made the motor less responsive to the different header configurations. Since the header scavenging effect must work in conjunction with the reflected waves produced by the intake manifold, the short-runners used on the blower motor provide no such ram tuning (at least, not in the rpm range of this motor). The only way to be sure is to run the motor in N/A trim with the short-runner blower intake (without boost pressure) to see if the trend continues. Another option is to run the motor equipped with a centrifugal supercharger and long-runner intake to see if the motor was more or less responsive to the changes in the headers.

Header Test: Stock GT500 Motor
After running our header test on a normally aspirated and supercharged 5.4L combination, the last thing we should want is more testing, but we decided to do just that. Since the cam timing had such a dramatic effect on the header comparison of the normally aspirated GT1000 combination, we decided to run the same test on a standard GT500 motor.

After running the header tests, we performed a pulley swap. The smallest pulley offered for the Kenne Bell measures 2.5 inches. Though this pulley has been run successfully, a better option is to increase the size of the crank pulley using this new unit from Innovators West. This will allow you to run a larger (2.75-inch) blower pulley to minimize belt slippage.

We pulled a stock GT500 motor out of a car, installed it on the engine dyno and ran the exact same series of tests. The GT500 motor was run in normally aspirated rim (using the same Cobra R set up), with the stock Eaton supercharger and then again with the Kenne Bell. In each configuration, we ran the motor with the stock exhaust manifolds; then the three different header configurations. Run in normally aspirated trim with the stock exhaust manifolds, the stock GT500 motor produced 384 hp and 413 lb-ft of torque. After installing the 1-inch headers from American Racing, the peak power numbers jumped to 410 hp and 432 lb-ft of torque. As with the normally aspirated GT1000 combination, the larger headers improved power (albeit slightly) at the top of the rev range, but lost power up to 5,000 rpm. Unlike the wild cam timing in the GT1000 combination, the stock exhaust manifolds did not kill the power on the GT500 motor.

The header test was run once again with the GT500 equipped with the stock Eaton supercharger. Run with the stock pulleys and stock exhaust manifolds, the supercharged 5.4L produced 594 hp and 559 lb-ft. Installation of the American Racing headers improved the peak power numbers to 602 hp and 566 lb-ft. As with the test run on the GT1000 motor, the headers improved the power output through the entire rev range. The larger 2-inch headers further improved the peak power output (compared to the smaller 1 headers), and the headers were worth as much as 10-11 hp and 12-13 ft-lb of torque over the stock exhaust manifolds. After running the stock supercharger, we installed the Kenne Bell 2.8L H-series Twin Screw supercharger. Equipped with the stock exhaust manifolds, the Kenne Bell-equipped 5.4L produced 719 hp and 615 ft-lb of torque.

Our fuel needs were met by a set of 60-pound Siemens injectors. It was necessary to run a spacer plate under the blower to provide room for the taller-than-stock injectors. Note also that we exceeded 1,000 hp with the stock fuel rails!

After installing the 1 7/8-inch headers, the peak power numbers jumped to 735 hp and 636 lb-ft of torque. While all of these values are a great deal to digest, testing the stock GT500 motor did verify that the stock GT500 motor was less responsive to the headers in both NA and supercharged form than the wilder GT1000 combination. Not only that, the NA configurations (both GT500 and GT1000) were more responsive to headers thanks to the presence of the long-runner Cobra R intake (versus the short-runner intake used on the supercharged applications). What all this seems to boil down to is that all applications will benefit from long-tube headers but those equipped with either after market cam timing and/or long-runner intakes will benefit even more.

Supercharged 5.4L Header Test
HP Numbers: 1¾-inch vs 1-7/8-inch vs 2.0-inch (NA 5.4L)
RPM1-3/41-7/82.0
3,{{{600}}}377376377
3,800400400402
4,000427429429
4,{{{200}}}455457458
4,400481483485
4,600506508510
4,800529530533
5,000549550554
5,200568570573
5,400588591593
5,600605608610
5,800617620622
6,000628632634
6,200640643646
6,400651654658
6,600655659663