This month, we're testing carburetor engine theory against actual dyno results. You know the deal about bigger carburetion being better and more fuel equaling more power. Well, too often people can't seem to get beyond the air/fuel metering myths. How big is too big, and what happens if it's too small? Many enthusiasts can tune their carbs with their eyes closed, but when it comes to choosing the proper initial carburetor size, there seems to be a lot of confusion.
The idea is to choose a main body with a large enough bore and venturi area to allow maximum airflow through the carburetor and into the engine. At the same time, however, these areas must also be properly sized to promote a high-speed airflow signal into carburetor, a measurement of air speed that the carburetor reads in order to determine how much fuel to deliver to the incoming air stream. In theory, if the bore and venturi are too large, the signal will be weak at low engine speeds and the fuel will not be pulled from the carburetor's discharge ports accurately. The opposite of this effect is when the carburetor's bore and venturi are too small for the engine's airflow demands. While throttle response will be outstanding, the engine's maximum airflow (power) potential will be diminished. The ideal carburetor sizing combination will be able to move the maximum amount of airflow through the carburetor while maintaining the strongest possible signal.
For those readers who like to crunch numbers, there is a commonly used formula for determining carburetor size, but it provides only a rough starting point as all the engine's variables are not taken into consideration. This equation assumes 100 percent volumetric efficiency (VE) and can be worked out using the following equation: VE = (displacement x rpm)/3,456. For example, if we were to run a 350ci engine at 6,000 rpm through this equation, the formula would tell us that a 350ci engine requires a 608-cfm carburetor. However, most 350ci engines do not produce 100 percent volumetric efficiency, even at peak torque. They more likely produce around 85 percent as a strong street/strip engine. The same 350ci engine with 85 percent VE would require a much smaller carburetor (516 cfm) using this equation. As you can see, compression ratio, bore and stroke ratio, combustion chamber design, and much, much more are not even considered. In this case, a good street/strip engine would probably work best with at least a 600-cfm carburetor, and the same engine dedicated to the dragstrip would use every bit of a 650-cfm unit and sometimes more.
Depending on the engine's total displacement, operating range, and intended horsepower output, the most efficient carburetor size varies. While theories make great bench racing conversation, we wanted to see what would happen if one engine were tested using carburetors of various size.
After pondering how to perform this test, it became obvious that a high-horsepower engine would show the broadest results between carburetors, but a 700-plus-horsepower small-block is hard on parts and isn't realistic. For every one of these engines roaming the race tracks, there are a thousand or more making 450 hp or less on the street. We chose a Smeding Performance 383ci engine rated at 440 lb-ft of torque and 440 hp at less than 6,000 rpm. Durability is a key issue when performing a test of so many variables, and Smeding has a good reputation for keeping its engines together.
When our motor arrived, it was complete from intake manifold to oil pan but without the accessories (not needed for the dyno flog). All vacuum, oil, and water holes were plugged prior to shipping to ensure that the motor was clean and ready for action. Since the engine was already assembled, we added the needed parts (see Parts List sidebar) and got ready for the dyno. Since the testing procedure called for every carburetor imaginable, we contacted one of the industry leaders in carburetor performance. Holley couldn't wait to test its different-sized carburetors under the editorial eye. After speaking with a knowledgeable representative, we decided on a standard 4150 HP mechanical secondary four-barrel carburetor in 390-, 600-, 650-, 750-, 830-, 950-, and 1,000-cfm airflow ratings. Because a test of this caliber can be overwhelming, Holley also offered us some West Coast technical support from The Carburetor Shop in Ontario, California. With the dyno mule ready to rumble and the carburetors in hand, we headed for The Carburetor Shop, where we would perform our shootout on the next door Vrbancic Brothers' engine dynamometer.
George and Bob V. also run The Carburetor Shop, so they are intimate with carburetor tuning; they recommended their 13/4-inch Hooker test headers outfitted with exhaust gas temperature (EGT) sensors and oxygen (O2) sensors. These sensors allowed the dyno operator to monitor the amount of unburned fuel passing through the headers, which helped us decide whether to jet the carburetors up, down, or not at all. As long as the EGTs stayed between 1,200 and 1,300 degrees F and the air/fuel ratio hovered around 12.5:1, everything would be considered optimum. We plopped the 390-cfm carburetor on the Edelbrock RPM Air Gap intake manifold, set total timing at 36 degrees, and fired the engine up. Follow along. The results may surprise you.
390 cfm: Big Power, Little Carb
Jet Size: Primary-65; Secondary-65
So long as the appropriate supporting items are used, the Smeding-built 383ci is rated at 440 lb-ft of torque and 440 hp. By following even the roughest formula, it's safe to say that 390 cfm wouldn't cut it with this application, but we wanted to see how much of a difference a small carburetor would make. The smallest four-barrel carburetor Holley offers in a 4150 HP-style mechanical secondary configuration is this one. We bolted the NASCAR 390 cfm on the Smeding piece and made a few pulls.
By reviewing the dyno results, it's easy to see that the small bore and venturi of the carburetor definitely restricted airflow. Compare this test to the later 750-cfm and 830-cfm tests, and you'll see that power is down considerably across the entire curve. By using the smaller carburetor, we restricted airflow and hurt the potential power the engine could have made. However, the torque did reach its advertised 440-rating even with the teeny-weenie carb. Surprisingly, the EGT and O2 sensors both showed a near perfect oxygen/fuel content in the exhaust, and carburetor tuning was not required.
600 cfm: Bigger Gets Better
Jet Size: Primary-70; Secondary-70
The next step up in the line is a 600-cfm carburetor. Had we used something smaller, it would have been a vacuum secondary unit and we didn't want to alter any factors in the testing procedure other than the cfm rating. We bolted on the 600-cfm carburetor but made no other adjustments.
As you can see, the engine produced considerably more power than with the small-bore 390-cfm unit. Average torque numbers jumped by almost 20 lb-ft, and the horsepower hopped up 17 more than before. As you follow the 500-rpm increments, it's easy to see horsepower gains from nearly 10 hp down low to upward of 25 hp or more higher in the rpm band. Again, carburetor jetting was near perfect out of the box, and we didn't alter anything.
650 cfm: Changing of the Guard
Jet Size: Primary-70; Secondary-70
Though power was on the rise, we still hadn't reached the level we sought. This time, we swapped on a 650-cfm pot. Surprisingly, more cfm only netted an average gain of 2 lb-ft of torque and 3 hp. However, the larger carburetor developed 444 hp and was 21 lb-ft of torque stronger than the advertised number. Reviewing the 600-cfm and 650-cfm horsepower curves at 500-rpm increments will again show an increase of approximately 2-3 hp down low and 5-8 hp higher in the rpm curve. Again, the air/fuel ratio was right on out of the box.
750 cfm: Still Gaining
Jet Size: Primary-73; Secondary-73
The important thing to mention here is that of all the carburetors tested, only the 750-cfm HP Holley came with a vacuum port below the base of the carburetor. We would have liked to include engine vacuum in the entire testing sequence, but due to a lack of accessible portable vacuum sources, we were unable to do so. Although the Air Gap manifold does have a tapped source for vacuum, it is inaccessible unless there is a spacer below the carburetor.
To our amazement, the average numbers were almost negligible: 1 lb-ft of torque and 1 hp. Peak torque was still up by 21 lb-ft and horsepower remained the same at 4 hp above the advertised peak. This test also showed optimum air/fuel ratio results, which revealed how well the HP Holley carburetors are calibrated from the factory. If you are wondering how the 390-cfm carburetor was able to supply as good an air/fuel ratio as the 750-cfm unit using smaller jets, it's because the smaller carburetor created a stronger airflow signal, causing the fuel discharge ports to deliver the maximum amount of additional fuel. The 750-cfm carburetor may have had an equally good air/fuel ratio with larger jets, but the air speed signal was slower; therefore, it drew less fuel from the jets and delivered nearly the same air/fuel readings.
830 cfm: Peaky Power
Jet Size: Primary-86; Secondary-86
We were sure the engine would stumble and rattle at idle with the 830-cfm carburetor, but we were wrong. Surprisingly, the larger carburetor ran considerably well down low. There was a definite throttle response difference between the 750-cfm and 830-cfm off-idle feeling, but both idle quality and power worked out well.
The 830-cfm carburetor posted the best peak numbers and average numbers of the entire test. While the average power gain only increased 2 lb-ft and 1 hp, it was nice to see some new peak numbers. This time around, the 383 made 22 lb-ft of torque more than its rating and made an additional 9 hp. The 500-rpm increments remained about the same as power increased with engine speed. Also, the air/fuel ratio was right on.
950 cfm: Super-Sized
Jet Size: Primary-78; Secondary-78
We were hoping to see even more power with the 950. It definitely produced a low-speed stumble, and we thought about tweaking the idle-air bleeds but refrained since we had come this far without touching any of the other carburetors. As soon as the engine passed the 1,200-rpm mark, the air speed signal through the carburetor improved and the engine ran smoothly. Typically, a large carburetor like this one is only run on an extremely high-horsepower engine, where the camshaft depletes most of the low-speed vacuum. With the idle speed set a little higher, we began the next pull in hopes of more power.
Unfortunately, the powertrain had maxed out, and we were now merely maintaining the best power curve we could. The peak numbers still posted 19 lb-ft of peak torque better than the advertised engine, and the horsepower was still up by 3 hp. In comparison to the previous test with the 830-cfm carb, the numbers reveal a loss in power at every point. The air speed signal through the carburetor slowed even at high engine speeds and caused the air/fuel mixture to become slightly less than ideal. While we could have altered the carburetor to promote a better power curve and perhaps equal the previous tested power, we would have been bandaging the situation rather than doing ourselves a favor.
1,000 cfm: Big Daddy
Jet Size: Primary-84; Secondary-84
After experiencing the 950-cfm carburetor, we weren't too sure of what to expect from an even larger piece. Usually, once you reach the limit of beyond, it's common sense to head in the opposite direction. However, since we're talking about carburetor sizing, we thought it would be interesting to see what happens way beyond big. Now, into Dominator cfm, we figured the engine would hardly run at all below 1,500 rpm.
Again, the Holley HP series amazed us with its ability to meter incoming air speed and adjust for the proper amount of fuel needed to run the engine. Of course, the idle was extremely choppy below 1,000 rpm, so we raised it to 1,200, where it seemed happy. The touch of the dyno handle had changed as well. Throttle response went out the window from anything below 2,000 rpm, and at that point, we just wanted to see what the extra cfm would do to our previous power curves. Amazingly, we saw one more peak lb-ft of torque and still managed to stay at a 5hp peak above the advertised horsepower rating. Reviewing the 500-rpm incremental test points, the 1,000-cfm carburetor did better than the 950-cfm unit but was still making considerably less power than the 830-cfm piece. Perhaps the air speed signal through 1,000-cfm carburetor boosters was stronger than that of the 950-cfm, which is why it did a little better. Either way, the 950-cfm and 1,000-cfm carburetors were too big for our engine.
The biggest difference in power was the change from a restricted 390-cfm carburetor to a larger 600-cfm unit. This part of the test showed just how much power was lost with too small a carburetor. The most interesting part of the test is how little difference in power there was out of the box between a 600-cfm carburetor through to a 1,000-cfm unit. As mentioned before, this had a lot to do with the air speed signal being picked up from the boosters. All of the tested carburetors featured down-leg boosters, which means the air speed is calibrated farther down in the venturi to help improve the signal. An annular booster design sits farther above the venturi so incoming air is not blocked, but at low engine speeds, the signal becomes extremely weak if the carburetor is too large. Judging from the test results, we'd say that the 750-cfm carburetor seemed to provide the best power curve and throttle response. In the event that the engine made 50 hp or more and was dedicated for racing, we'd be willing to say that the 830-cfm piece with down-leg boosters might add a few extra horsepower and still be livable on the street. As you can see, carburetor sizing isn't a science. As long as you are not all-out heads-up racing, simply balance throttle response, low-speed driveability, and peak power as close as you can, and the carburetor will do the rest.
Part; Mfr.; PN
Valve covers; Holley; 241-80
Valve cover gaskets; Earl's; 29E03B
Oil filter adapter; GM; 3952301
Oil filter; Pep Boys; 0-76333-75134
Oil (10W30); Pep Boys; --
4 ½-inch water plugs; Pep Boys; 0-37495-40081
3 3/8-inch vacuum plugs; Pep Boys; 0-89601-60180
Distributor; Pertronix; D100701
Coil; Pertronix; 45111
Plug wires (universal fit); Pertronix; 808490
¾-inch reach, non-tapered spark plugs; Pep Boys/Autolite; AR51
390 cfm; 0-80507-1
600 cfm; 0-80540-1
650 cfm; 0-80541-1
750 cfm; 0-80528-1
830 cfm; 0-80509-1
950 cfm; 0-80496-1
1,000 cfm; 0-80513-1
Headers; Hooker; 2117HKR
Header gaskets; Fel Pro; 1405
Header bolts; Used; --
Intake carburetor studs; Used; --
RPM Air Gap; Edelbrock; 7501
RPM Performer heads; Edelbrock; 6090
Smeding Performance 383
Block: GM four-bolt main
Compression ratio: 9.8:1
Crankshaft: Nodular iron
Rods: 4340 forged steel
Pistons: Hypereutectic with graphite coating
Camshaft: Roller hydraulic, 0.488/0.510-inch lift, 234/244 degrees duration at 0.050 inch
Rockers: Roller, 1.5:1 ratio
Heads: Edelbrock RPM Performer, 2.02/1.60 intake/exhaust valves
Intake: Edelbrock RPM Air Gap
HP: 440 at 5,500 rpm
Torque: 440 lb-ft at 4,300 rpm
As it turned out, our 383ci engine matched with HP series Holley carburetors didn't require much attention to tuning. We were lucky to have EGT and 02 sensors to tell us exactly what the air/fuel ratio was doing. However, we are aware that most enthusiasts purchase a carburetor, head for the track, and tune according to performance. Most of this performance usually comes from changing jet sizes and altering the air/fuel mixture screws. While the external air bleed screws are easy to access, the main metering jets can be quite a pain, which is why Holley offers several parts and tools to help with this task. We did not have a need to use our tools (pictured above) in the story at hand, but we did feel they are worth mentioning. This is what we bring whenever we plan on doing carburetor work.
Metering block gaskets; 108-89-2
Bowl gaskets; 108-83-2
Jet change tool; 26-68
Quick-change jet bowl; 34-25
Quick-change jet bowl; 34-26
Air bleed assortment kit; 36-240
Jet assortment kit (sizes 64-99); 36-181
Close limit jet assortment kit; 36-213