carbs or fuel injection

The relative merits of carbs or fuel injection are again of interest to us snowmobilers, as tighter emission requirements may be looming on the horizon.

This discussion has probably existed as long as there have been combustion engines around, and both systems are still in widespread use. One of the first recorded instances where one system was preferred over the other goes back as far as 1903. The first Wright Brothers aircraft engine used a simple carb for its historic flight, but in 1906 the Wrights switched to a simple continuous injection into the intake manifold.

While looking for previous research on the subject, I ran into a proceeding where the top engineers in the combustion engine field presented a paper discussing the merits of carburetion and fuel injection. They divided the systems into three areas; the conventional carburetor, continuous injection into the intake system, and direct injection into the cylinder.

The authors reported that fuel injection had been found to have better starting, good idle, improved acceleration, and a more positive shut-off. No measurable improvement in fuel economy had been obtained. Direct injection into the cylinder was mentioned as a potential benefit to the fuel economy of two stroke engines. The largest objection to both continuous injection into the intake, and timed injection into the cylinder was the difficulty in controlling the quantities of fuel at all operating ranges by the present mechanical devices. So when were these recommendations made? The paper was presented by Mr. Ellor, the Chief Engineer of the Rolls Royce Aero-Engine Division, to his peers at the Royal Aeronautical Society in 1939. Fast forwarding by 60 years, it is clear that Mr. Ellor's analysis was spot on. The carbs are still with us, and his comment that "there is no need to change to a more complicated system, if your carbs are working fine," could be as valid in today's snowmobile industry as it was 60 years ago. Only one basic parameter has changed since Mr. Ellor's analysis of 1939, namely the computer.

Mechanical devices are now replaced with electronic valves controlled by computers, and this has made injection into the intake manifolds a practical reality. In the automotive world the electronic fuel injector (EFI) has totally replaced the carburetor because of constantly increasing demands on fuel economy and emissions.

In the snowmobile world we have seen the introduction of electronic fuel injection by first Polaris, and then Arctic. Polaris no longer offers a fuel injected snowmobile in its lineup, but Arctic has kept a few models. So why didn't fuel injection catch on with the snowmobile public or with the racers? "Too complicated" can best sum up the response.

Hot-rodders like to fiddle with their machines, and the EFI was too advanced for everyone but a handful of specialists. To change the calibration of an EFI you can do three things; you can change the rail pressure with an adjustable pressure regulator, you can change the size of the injectors, or you can reprogram the computer chip. The first two methods are relatively simple, but they have the drawback of lifting the whole power curve from idle to top end.

If you want to change the calibration across the power band you need to change the computer program on the chip itself.
Buying a new chip is not cheap, it can cost as much as $200.00. A main jet on a carburetor costs $2.00, so it is no wonder that most racers prefer carbs over EFI units.

While simplicity and familiarity speak for the carb, there are several potential advantages to the use of an EFI system on a racing machine. Fuel delivery is not dependent on mass flow, and while larger carbs get trickier to jet, you can use a large throttle body with an EFI.

Two stroke racing engines produce more power with larger and shorter intake systems, and fuel delivery can be controlled more consistently with an EFI than a large carb. The EFI computer also adjusts the fuel delivery to changes in temperature and barometric pressure, so constant rejetting at the race track is unnecessary.

A good example of this advantage was Darcy Ewing's win at the first Formula III race of the 1990 Sno Pro Circuit. Darcy had a set of 52 mm throttle bodies on his 700 Arctic controlled by a programmable Haltech EFI unit. By the time the Formula III finals arrived at the starting line, the temperature on this balmy December Sunday had reached 50 F. All the competitors had scrambled to rejet, but all must have guessed on the rich side. When Ted Otto dropped the flag, Darcy stormed off the line clean and crisp, while everyone else bogged. Darcy Ewing cruised to victory with a straight-away lead, without having to spend any time recalibrating his fuel system. Although the system worked well on the larger Arctic 700 twin, we later struggled for two years trying to get the EFI to work on Ewing's 10000 RPM 340cc twin FI engine. This engine proved too temperamental and we had to switch back to carbs to be successful. There was an upper rpm limit of 10000 rpm which we could not crack with the injectors at the time. With the latest development in software, hardware and injectors, race engines are now revving up to 14000 rpm, but they usually have to employ two staged injectors per cylinder to get acceptable performance both at part throttle and at top end.

To be successful in racing with EFIs, you have to spend a fair amount of money on hardware, and you must also be familiar with programming fuel curves. Once the calibration is done satisfactorily you will have advantages in both performance and reliability.

The development of carburetors has not exactly stood still in the last 20 years. Flat slide carbs are making inroads because of better acceleration and fuel economy. Flat slides create a quicker vacuum signal at the venturi, which leads to better atomization of the fuel droplets. The smaller the fuel droplets can be broken up, the better they mix with the air, improving both performance and fuel economy. The rack mounted carbs we are now seeing on snowmobiles also offer easier throttle pull and a throttle position sensor mounted in the slide shaft. This throttle position switch feeds back into a digital ignition which can now select spark advance based on both throttle opening and rpm. Several devices are also appearing to control fuel flow by changing the pressure in the float bowl. Several manufacturers elect to use these to compensate for altitude (barometric pressure), while Holtzman Engineering has one unit called "Tempaflow" which changes calibration according to temperature variations. Holtzman also has a unit which varies the pressure in the float bowl with an adjustable valve mounted on the dash. None of this is particularly new to carburetor technology. When Mr. Ellor was speaking to his learned peers in 1939, aircraft carbs already had automatic adjustment for temperature and barometric pressure. This was accomplished by mounting three main jets in parallel. One main jet worked in a conventional fashion. The second had a main jet with a tapered needle controlled by a temperature sensitive metal strip. The third jet also had a tapered needle, this one is controlled by a diaphragm sensitive to barometric pressure. Mr. Ellor and Rolls Royce were soon embroiled in the second World War, and although the carburetor performed satisfactorily on the famous Rolls Royce Merlin engine, the constant quest for more power from less weight resulted in the development of two major projects. One of the projects was a two stroke aircraft engine with direct injection. This design quickly produced substantially more power than the four stroke engine. By utilizing the exhaust gas energy to power a turbine geared to the output shaft, good fuel economy was also obtained. It seemed that the two stroke engine might take over the skies as the power plant of the future, except for the very promising results of the second project under way.

The second project was the jet engine, and we all know who won out as the future power plant of the skies. It would have been interesting to see what form the two stroke engine would have today if all the aircraft manufacturers had devoted the last 50 years to its development.

A small Australian company called Orbital Engineering started a whole new interest in the two stroke engine by introducing a unique direct-into-the-cylinder injection system. Hope for the use of two strokes as a light and compact automotive engine enticed a number of automotive manufacturers. All the major car factories had two stroke programs the last 20 years, and invested generous resources into direct injection technology. At the present time the two stroke engine is on hold, not because the injection technology didn't work, but because development is not yet at a stage where the car companies can guarantee the 200,000 mile component life consumers are accustomed to. All the new development was not in vain, most of the injection technology is now being applied to four stroke engines in the constant quest for better economy and less emissions.

The snowmobiler may be the ultimate beneficiary of all this research. Outboards and watercraft already have to meet the new emission standards. Polaris and Arctic are both working with the Ficht Injection System on license from OMC, while Ski-Doo has signed up with Orbital. There is a considerable difference between the two systems. When the fuel is injected directly into the cylinder, it has a very short time to evaporate before it is ignited. (Less than half a revolution of the crankshaft). In order to accomplish a full mixing in the short time, the fuel droplets have to be broken up into very small particles. Ficht accomplishes this by injecting the fuel at a very high pressure.

The Orbital system premixes the fuel with high pressure air from a compressor, and blows the mix into the cylinder. Both the Ficht and Orbital systems use digital computers to calibrate the injected fuel volume. Both systems have advantages and drawbacks when adapted to existing engines. As the development continues, we can expect to see engines which are specifically designed to match the new injector technology. Since the injection process can be better utilized at lower engine speeds with more time available for atomization and mixing, we may expect to see larger engines running at lower speeds.

So you wonder if those big 800 twins are just a fad? They may in fact be the next step to acceptable performance with the new direct injection fuel systems.

So is the carburetor finally dead? We think it will be around for a long time, particularly on two stroke racing engines. Want an example? Automotive engines have had EFI as standard equipment for almost 20 years, but some of the best racing around are NASCAR stock cars, and they are still using carbs.