It’s a contentious affair that has been ongoing now in hot rodding for roughly 30 years … carburetors versus electronic fuel injection (EFI). There are ardent supporters and vociferous cynics on both sides. There have been many converts who have taken the leap of faith and dropped their float bowls in favor of fuel injectors, but there are still those who will not put their dollars behind a black box that works its digital magic in shrouded secrecy.
In this world of polarized opinion it’s difficult to find an unbiased source of accurate information but that’s what we will try to do with this story. We decided to pitch the new FiTech self-learning EFI throttle body up against a traditional Holley carburetor. The FiTech system has been receiving quite a bit of attention lately mainly due to its affordable price. What we found will certainly not make converts of every last carburetor devotee, and that’s OK. But it’s worth a few minutes of your time to take a look at our results.
Before we dive into the testing we should look at exactly what’s happening when and where fuel is introduced into the engine. It should be common knowledge that liquid fuel does not burn very well. In order to get maximum benefit from its stored energy, fuel needs to get as close to a vapor as possible. A good way to assist this process is to introduce the fuel as far upstream in the induction system as possible.
This would appear to give a carburetor a slight advantage over a multipoint fuel injection (MPFI) system since a carburetor introduces fuel much farther upstream than the average MPFI system where the injectors are placed directly above the intake ports. One way an MPFI system can improve this situation is with high fuel pressure of 43 or 58 psi to improve atomization at the point of entry.
But in tests where an MPFI system is compared to a carburetor, the old-school fuel mixer usually makes a little bit more power. This is one reason why the throttle body style of fuel injection might have a slight advantage as the injectors introduce the fuel in roughly the same place as a carburetor.
One potential advantage favoring EFI is that carburetors are difficult to tune across a very wide rpm band. In drag racing where the engine is making power over a relatively narrow rpm band of perhaps 1,500 rpm, this is more time-consuming than difficult. On the street, carburetors are asked to meter fuel from idle to 6,000 rpm or higher—a much more challenging task. On the dyno, a typical Holley four-barrel will tend to run much richer at low engine speeds (below peak torque) versus the air/fuel (A/F) ratio at peak horsepower. While a carburetor can be expertly tuned to lean out the lower engine speeds, this is a difficult task that few bother to perfect.
Before we get into the testing, perhaps an overview of the FiTech system is in order. The FiTech is similar to many of the now familiar self-learning throttle body EFI systems on the market. FiTech offers several different models depending upon the application, so we chose the basic 600hp, four-injector model that is compatible with engines making between 250 and 600 naturally aspirated horsepower. The FiTech unit incorporates the ECU into the throttle body itself, which minimizes the number of electrical connections. Each injector is rated at 80 lb/hr and the ECU does not require a laptop computer to tune since all the inputs are placed through a handheld controller. This makes it easy to configure and tune.
This particular FiTech is the Go EFI 4 model that also offers control over the ignition curve. Because we would be testing on an LS engine, we chose to use the MSD controller for the ignition side to eliminate variables of different ignition controls. On engines with distributors, ignition control requires a locked out, two-wire distributor like an MSD.
We began our test with an iron-block 5.7L LS budget engine. This was an engine assembled by editor Rupp using parts we had laying around. It started with a 5.3L Summit Racing block, bored out to 5.7L specs using essentially a stock 5.7L LS rotating assembly and stock 5.7L cathedral port heads. The cam used is a Comp XFI hydraulic roller (PN 54-414-11) with 216/220-degree duration at 0.050, a valve lift of 0.525/0.532-inch, and a lobe separation angle (LSA) of 114 degrees.
To begin our testing, we started with a Holley Mid-Rise single-plane intake (PN 300-132) for the cathedral port heads mounted with a Holley XP 750CFM Ultra XP carburetor (PN 0-80803). After a short test run to tune the A/F ratio, the carburetor was leaned slightly but still delivered a somewhat rich A/F to make max horsepower. After tuning, Westech’s Steve Brule made three separate runs and averaged them into an overall power curve. This became the standard averaging procedure for all subsequent test results.
Next, Brule installed the FiTech and performed several runs after setting the wide-open throttle A/F at 12.5:1. The power graph revealed that the FiTech was able to improve power over the carburetor basically throughout the entire range from 3,000 to 6,000 rpm. As you can see from the graph, this represented a 5 to 7 lb-ft of torque gain nearly throughout the entire pull. We’ll take a closer look at the results in a couple of paragraphs.
Rather than leave it at just this one test, Brule replaced the single-plane intake with an Edelbrock Performer RPM dual-plane (PN 71187) that is more commonly found on mild LS street engines. For this baseline, he had to lean the carburetor by several jet sizes to achieve a good A/F for best power. But when the FiTech was installed, it went on with no changes. We half expected the same results as the single-plane test but the results quickly revealed that the Holley carburetor and the FiTech made virtually the same power throughout the entire rpm band.
These results prompted us to look a little closer at the data. Typically, carburetors are jetted for maximum rpm power. As we mentioned earlier, this most often means that at low engine speeds, carburetors tend to deliver a rich A/F ratio. We recorded the A/F numbers and fuel flow for both the single- and dual-plane tests and noticed an interesting trend. In both cases, the FiTech delivered less fuel than the carburetor, which is one of the advantages of EFI. The dual-plane test produced less of a change than the single-plane.
There are a number of different reasons for this result. Changing to leaner jets in the carburetor would have reduced fuel flow at the lower engine speeds, but this would have then pushed the A/F too lean at WOT. There are ways to balance this with high-speed air bleeds, but these are pro tuner techniques not generally available to the average guy, which is why we did not attempt them. Looking at the range of A/F ratios, the carburetor was about a full ratio richer at the lower engine speeds compared to peak rpm, which is typical.
FiTech’s Ken Farrell says that one reason the FiTech’s power numbers are so good is because of where the FiTech throttle body delivers the fuel. As you can see from our “down-the-throat” photo, the FiTech injects the fuel through a series of annular discharge holes around the circumference of the venturi, much like a carburetor. Other self-learning throttle body systems inject fuel underneath the throttle blades. Farrell says this drastically differentiates the FiTech from other systems and is responsible for a measurable power increase.
This also leads to another important advantage of the FiTech self-learning injection over a carburetor. With a simple command of 12.5:1 A/F ratio at WOT on the handheld, there is a very good chance that this is the A/F the engine will see throughout the entire rpm curve. This didn’t happen to the extent that we expected with the FiTech dual-plane test. Because we only ran the engine a very short time at WOT for three pulls, Farrell says the system needed more time at WOT than the three we used for averaging in order to complete its learning cycle.
We can also extrapolate that this same stable A/F condition will occur at part throttle. All the tuner has to do is command the part-throttle A/F ratio and the system will eventually achieve that ratio after a short learning process. It’s possible to tune a carburetor to achieve similar lean part-throttle ratios but that usually requires multiple, labor-intensive tuning changes. It’s much easier with a few keystrokes on a self-learning EFI like the FiTech.
Plus, these self-learning systems also dramatically improve cold start. While we may only have succeeded in pouring gasoline over the fire that is the carburetor versus EFI controversy, hopefully this story has shown that self-learning EFI systems can hold their own at WOT power. If nothing else, you have to admit it’s much easier to exert control over the A/F ratio with the touch of a button than draining fuel bowls and changing jets.
01. EFI has entered the 21st century and is becoming more sophisticated, including the self-learning TBI systems like the FiTech Go EFI 4 version that we’re testing here.
02. The FiTech Go EFI 4 unit (PN 30002) employs four 80 lb/hr injectors and is rated for use on naturally aspirated engines making between 250 and 600 horsepower. FiTech incorporates the ECU directly into the throttle body to reduce the number of wiring connections that must be made. Other than the main battery power and ground, there are only a few hookups necessary that require only a short time to install.
03. The FiTech throttle linkage is designed to accommodate typical linkage that will fit most automatic transmissions for either kickdown or TV cable connections.
04. The critical sensor for all self-learning EFI systems is the wide-band oxygen sensor. This provides the feedback to the ECU that helps the system learn to create the commanded A/F ratio for both part throttle and WOT.
05. The FiTech handheld color screen gives the user direct access to the tuning with a minimum of key strokes. This screen reveals a cam tuning selection that is very helpful for light acceleration tuning. FiTech offers four cams from which to choose. With this mild street cam, we chose cam number 2. Generally, big cams need more acceleration fuel compared to shorter duration camshafts. This is a tuning option some other self-learning systems don’t offer.
06. The coolant temperature sensor is the only other sensor that needs to be installed in the engine. This is a crucial input to the ECU that adds fuel when the engine is cold to improve driveability. If you know what it’s like driving a carbureted engine without a choke in cold weather then the cold engine driveability improvements might make this a major selling point. The FiTech system didn’t come with an LS-style sensor, but our engine already had one and the FiTech system and plug was compatible with it.
07. This photo shows how the fuel is introduced. The FiTech works much like an annular discharge carburetor and injects fuel through a series of small holes around the circumference of each venturi. The injectors are also staged to fire sequentially rather than in batch fire mode.
08. We used a Holley Ultra XP 750-cfm carburetor as the baseline testing for both the single- and dual-plane intake manifolds.
09-10. This graph reveals how the FiTech significantly outperformed the carburetor on the single-plane intake manifold test with a small, but consistent, torque advantage throughout the entire rpm band.
|RPM||TQ Carb||HP Carb||TQ EFI||HP EFI|
11-12. On the dual-plane intake manifold test, both the carburetor and the FiTech performed equally. We’d call this test a draw although the FiTech might have improved power at the lower engine speeds with a slightly leaner A/F ratio had there been more runs to help it learn or if we had used a laptop to tune it manually.
|RPM||TQ Carb||HP Carb||TQ EFI||HP EFI|
Air and Fuel
These two charts show the difference in air/fuel (A/F) ratio between the carburetor and the FiTech fuel injection. Generally, carburetors that are tuned for peak horsepower tend to run rich at lower engine speeds by 1 full point or more. Fuel injection tends to stabilize that ratio across the rpm band, although the FiTech dual-plane test did not deliver in this case. It’s possible it needed more test sweeps at WOT to finalize the automated tuning. We would expect the A/F at lower speeds to be consistent with the higher rpm A/F ratios.
|Single-Plane Intake Carb vs. EFI|
|RPM||Carb A/F||EFI A/F|
|Dual-Plane Carb vs. EFI|
|RPM||Carb A/F||EFI A/F|
|Large Red||Battery positive (+)|
|Large Black||Battery negative (-)|
|White||Switched 12-volts positive (+)|
|Blue||Tach lead (when ignition not controlled by FiTech)|
|Black multi-pin||Handheld controller|