Eric English
May 16, 2012

The Mule

At home in a '65 Falcon owned by Randy Dunphy, our test engine is a 331-inch Windsor that is similar to what powers thousands of hot Fords across the land. It's fitted with GT40X aluminum heads, roller rockers, Tri-Y headers, and a nice Comp Xtreme Energy hydraulic roller with 224/230 degrees duration at 0.050, and 0.513-inch lift. Up top sat a 650-cfm carb on an Edelbrock Performer intake, a combo chosen simply because it was sitting around when the engine went together. Dunphy had been considering ideas for increasing the visual appeal and performance of his engine, but wasn't willing to commit the budget to a couple of currently popular options--EFI or supercharging. We pitched our ideas about dual quads to Dunphy, and he readily agreed to be involved in our testing. Commonly used cfm formulas call for around 600 cfm for a moderate street 331 such as this one, so the 670 put us just a bit on the aggressive side of the ballpark.

Our testing took place at Blood Enterprises' shop in Auburn, Washington, using a Mustang MD250 chassis dyno. The shop has extensive experience tuning everything from stock restorations to vintage road race and drag cars, so we knew we'd be in good hands. We made a decision to test all intakes with the same 34 degrees of timing Dunphy has found to work well with this motor. Likewise we tested all carbs sans air cleaners, not wanting restrictions there to influence our results. Virginia Classic Mustang supplied a variety of OEM style hardware such as intake water necks, carb studs, and nuts, and once we got everything tightened down, the original Cobra intake/Holley 670 combination performed well. Pulling strongly to our self imposed 6,000 rpm redline, the Holley was surprisingly spot-on out of the box, its No. 65/No. 68 jetting contributing to a near perfect 12.5:1 air fuel ratio (AFR) as read by Blood's wideband O2 sensor. After a couple of runs, dyno operator and technician James Leahy opted to quicken the vacuum secondaries by switching from a plain diaphragm spring to a purple one. Peak horsepower was 243.9 at 6,000 rpm, with 252.5 lb-ft at 3,750.

Keep in mind that a Mustang dyno typically reads in the range of 20 percent lower than the oft-quoted Dynojet, plus the reciprocating loss of the manual transmission drivetrain is along the lines of 15 percent. Factor these into the mix, and Dunphy's 331 looks to be putting out 330-340 flywheel horses--not bad for a tractable street SBF. The fact is, the most important element of dyno tuning isn't ultimate numbers, rather its repeatability and comparison of changes. We certainly got that here.

Swapping for the Edelbrock Performer RPM was quick thanks to Dunphy's mechanic Freddy Jonsson. Throughout our testing, Jonsson spun the wrenches and had carbs and intakes on and off more times that we care to count--thanks a million Freddy! With the improved airflow dynamics of the Performer RPM, we weren't surprised to see initial dyno pulls run a bit leaner than with the Cobra. The 13:1 AFR at peak power wasn't bad, but the engine was only in the realm of an extra 5 hp and 5 lb-ft, and we felt it vital to try and equalize the air/fuel ratios as much as possible. To this end, Leahy fattened up the secondary jets three sizes to No. 71s, and let it fly. The results were much more as anticipated--with A/F ratio near 12.5:1, the Performer RPM showed it's superiority over the vintage Cobra design. In the end, peak numbers were 256.6 hp at 6,000 rpm and 262.1 lb-ft of torque at 4000, an improvement of 12.7 hp and 9.6 lb-ft at the peaks. The numbers were better for the Edelbrock from 2,200 rpm and upwards, thus a very usable gain overall. Modern technology prevailed, and likely would have looked even better on an engine with 6,500rpm capability.

LOOKING AHEAD

With solid numbers for our four-barrel intakes in hand, we look forward to a dual quad smackdown in the concluding segment of this story. The tuning capability of Blood Enterprises' dyno with a wideband oxygen sensor proved valuable to this point, but will become vital as we move forward--experimenting with four combinations of two-fours. We'll be testing a currently available Blue Thunder intake, and an obsolete Ford piece that was used by 1967 Trans-Am teams and offered through catalogs for a number of years. Each will be tested with two sets of carburetors in an attempt to determine the potential of dual quads. We'll evaluate the 465-cfm Holleys, which Carl's Ford Parts typically packages in its small-block two-fours setup, but we also took the opportunity to test the even smaller Holley 390s thanks to Holley coming through for us with the necessary carbs. We'll stick our neck out a bit here and predict good things for the dual quads compared to a similar vintage four-barrel, but how will they do when stacked up against modern single-four technology? Which intake proved its worth? Which set of Holleys ended up being the better performers? Tune in next time to find out.

Photo Gallery

View Photo Gallery
9 Ford- and Shelby-lettered aluminum four-barrel intakes were sold over the counter, and are kissing cousins to the Cobra-lettered G.T. 350 intake in our test. Even Edelbrock's F4B had a strong familial look, and we suspect very similar performance. Such intakes are available on the used market today, and are popular period pieces, with this Shelby-lettered intake from a vintage Shelby parts catalog advertised to be worth 22-26 additional horsepower when teamed with a 600-cfm Holley. It isn't explained as to whether this gain is based on an original two-barrel or four-barrel configuration.
10 Our upcoming dual quad tests will use the currently available Blue Thunder intake (bottom), and an obsolete Ford/Shelby intake (top) said to be considerably better. We'll have an opportunity to evaluate 780 cfm (two 390s) and 930 cfm (two 465s) on each manifold, both of which would be commonly viewed as "too much" carburetion for a hot street 331-inch engine. Do traditional cfm formulas hold true when delivered via dual quads? We aim to find out.

Dyno Comparision Chart

CobraEdelbrock
RPMHPTQHPTQ
200088.1231.185.8225.3
220095.5229.297.0231.7
2400108.2237.1110.1241.0
2600119.9242.3120.9244.8
2800129.0241.8129.5243.3
3000138.1241.7139.8245.0
3200150.0246.1153.4251.6
3400161.3249.3165.6255.6
3600172.3251.4177.1258.3
3800182.6252.3188.2260.4
4000191.4251.3199.7262.1
4200200.5250.7208.0259.5
4400208.4248.8214.5255.6
4600215.1245.7220.7251.9
4800221.0241.8227.2248.8
5000226.2237.7234.1245.6
5200230.4232.6240.4242.6
5400233.9227.6244.9238.1
5600237.8223.2249.8234.1
5800241.7219.6254.6230.8
6000243.9215.9256.6226.4