Like Starfury said, EFI ECM's only have so many data cells to map fuel and spark, so there is a limit to the tunability with EFI, though closed loop operation helps to smooth that out. Carb's have emulsion circuits, with holes cross drilled from the primary air bleed well into the primary fuel well. At WOT as rpm increases the fuel level in the primary well drops, uncovering the bleeds from the primary air bleed well and drawing in air to the primary fuel well. The higher the rpm, the more the fuel drop, the more emulsion holes uncovered, the more air drawn in, the more the AFR is changed(this also helps to break up and emulsify the fuel, making it more readily atomizable and more easily lifted out of the primary well, as well as being closer to it's vaporization point). By controlling the number and size of those cross bleeds, along with the size of the primary fuel well, the sensitivity of the boosters, the amount of signal bled off by the primary system's air bleed/air drawn into the primary air bleed well etc, you can control the AFR at any throttle position and any RPM as well as you can with EFI, and sometimes even better depending on which 2 systems you're comparing.

A good carb that has a tunable emulsion circuit when dialed in properly(or one that's pre-drilled properly), will have just as precise an AFR as EFI will, so EFI loses it's advantage there. With a carb that has sensitive booster/metering packages you can run a larger carb and not have a high end restriction and still get low speed atomization, so EFI loses it's other advantage there. All that's left then is atomization/emulsification, and carburetion always has that advantage.

That's why a good carb will usually outperform good EFI, even if only by a small amount. You get what you pay for generally with either system. $700 may seem like a lot to setup a good carb, but to get an EFI system with comparable flexibility and performance, I'd be getting into something like FAST XFI or Accel DFI, both of which are over $4,000.

The higher end race carbs(large cfm race only) typically will run 4 or 5 circuit emulsion systems, but can set you back well over a grand just for a basic carb. Usually the comparison of EFI outperforming carbs is from comparing a medicore/crappy carb to a good EFI system. Will FAST XFI outperform a typical Holley 4bbl? Yes. Will a stock production EFI system outperform a high dollar race bred carb? No. Neither is a fair comparison though.

A really good carb is only worth the money though if the person tuning on it knows what they're doing, and there-in lies the problem. Most people really don't know how a carburetor works and how all the parts in it affect the overall system.

As a side note, Formula 1 are obviously the world's most advanced naturally aspirated engines, and though they run EFI it's either a speed density or an alpha-n(or a mix). It has no MAF, and they run throttle body type injection(called stand-off injection). They do that because delivering fuel upstream of the intake ports allows more time for fuel atomization in the intake runners, giving it at least partially some of the advantages of a carburetor. The only other naturally aspirated engines that can equal the level of performance an F1 engine produces, are carbureted.....Nextel Cup and NHRA Pro Stock.

 

Come on. First of all, a carburetor is very poor at adjusting for changes in atmospheric pressure. Take a carbureted car from sea level and drive it into the mountains at 5000 feet and it will be running slobbering rich. A carb works on Bernoulli's principal which is dependent on air density which changes with altitude, temperature, humidity, and air pressure. The carb DOES NOT adjust for all of these. A FI computer can do that. No not simple speed density, but mass air systems certainly can. On top of that, the computer interpolates the data in the fuel map, and a good computer will fit the curve rather than just linear interpolation. Then if you want to talk about multipoint FI, the fuel distribution between cylinders is very very very even because the fuel goes directly into the cylinder. On a carburetor the fuel charge has to be delivered by coming from the carb and going around all the corners in the intake. Invariably some cylinders are a little bit richer than others. No way around it. Good intake design minimizes this, but the design has to be a tradeoff for caertain design RPM's. Thats why there are dual plane, single plane, tunnel ram, etc. Lets not forget that modern ECU's can also vary ignition timing. Modern systems are closed loop and can monitor multiple parameters, and adjust fuel delivery for changing conditions as needed and only resort to the base map for sensor failures and high power application. You can say all you want about how tuneable a caburetor is, but it can never match up to fuel injection unless you are talking speed density, and even then the speed density will probably win unless you spend an inordinate amount of time tuning the carb. Face it. Most people would never spend the time to tune a carb to get anywhere near optimum and for a daily driver or weekend driver 99.9% of the time a FI will be better.

 

Nextel cup are limited to carburetor so they can't run FI anyway. If FI were allowed, I bet it would be on there. Pro stock are go fast only. Try running a pro stock on the street. For that matter, a Nextel racer doesn't spend much time at low power settings either. In any of these cases, the pit crews change carburetor adjustment for the given atmospheric condition to optimize.

 

Oh, by the way. In the case of drag racers the pit crew monitors the engine performance using an on-board computer and adjust settings between each heat compensating for atmospheric changes even between race heats. They maintain a database of every race that they have run and what worked and what didn't. Yes very tuneable, but they don't do it by themselves.

 

I actually live in the mountains, so I do drive from sea level to 5,000ft on a regular basis. My AFR doesn't change enough to notice it on the plugs, the fuel ring at the base is the same size regardless of altitude, so the AFR is changing by maybe 0.1-0.2. Carburetors don't respond to atmospheric pressure alone. They respond to the difference in pressure between the atmosphere and the barrel, and at higher altitude where atmospheric pressure is lower the engine's pumping efficiency drops due to reduced power, so airflow drops, and the pressure differential stays reasonably consistent, so AFR doesn't change all that much. Is it ideal? No. Does the AFR change? Yes, but not enough to offset the power advantages the carburetor can provide. That's what matters, not the AFR by itself, or timing by itself, but overall vehicle performance.

And race tuning isn't always about changing because of atmospheric conditions. Sometimes they're trying to tune for that last 1/2 a horsepower, but other times they're playing with tuning for reasons such as traction control, fuel consumption, detonation resistance etc. The F1 cars for instance will dial the AFR towards lean(they do this as the car is driving) and drive a tad easier in order to try and conserve fuel and make it farther on the same tank of fuel. Redialing a carb to account for atmospheric variances only matters if you're concerned about minimal power gains. It doesn't matter in a street car if you make 1% more power, but in a bracket car where competition is that tight, it does.

I agree that FI is better in 99% of cases because people don't truly tune their carburetors right. It takes a lot of work and a lot of knowledge, and most people really do NOT know what's going on inside a carburetor. But the fact remains, a good carburetor that is properly tuned will usually outperform EFI everywhere, except at extremely low rpms, but even then newer carb technology(such as annular boosters) have greatly improved low speed atomization in carbs and have closed the gap considerably on EFI's low speed advantage.

Everyone for the most part likes to accept the wive's tale about EFI being better than carburetion, when it's simply not true. There's just a lot of misinformation out there. It may have held some truth decades ago when EFI technology was more advanced than carburetion technology, but it's not decades ago any more. Does EFI have advantages? Yes. Does carburetion have advantages? Yes. In the end all that really matters is how the vehicle performs.

http://www.pro-system.com/scoop92102.html Read that. Those guys make custom carb setups for a lot of race cars, Pro Stock etc. It's rather enlightening.

 

Let me put it this way. Everything that you have mentioned says that carburetors make more power and I never said they didn't. However, all of your arguments relate to how much power the carburetor is capable of by referencing race cars whether they are Nascar, drag, or whatever. The problem with your argument is that all of these engines are designed to run optimally in a very small power range. A street car has to operate in everything from bummper to bumper traffic to however fast you want to cruise down the highway with slow acceleration to rapid acceleration. When you have a single or very narrow power band requirement like a race car, you can get more from the carburetor. You mentioned that formula 1 cars vary fuel-air ratio to achieve better economy on the fly so they don't have to pit as much. Don't you think the Nascar guys would love to do that too if they could have FI on their cars? BTW, how efficient is the drag racer engine when it is sitting on the line? Not very efficient. A carburetor whether you like it or not IS affected by atmospheric changes for which it can't compensate. Conversely, a closed loop FI WILL compensate and do so on the fly automatically over the whole rpm range. Tuneability to me for a street car is not a matter of simply which one puts out the most power, it is which one can be made to operate more efficiently across the whole rpm range and the whole atmospheric range. I stand by what I said in this regard that a closed loop FI will outperform the carburetor when considering the entire rpm range as well as the entire atmospheric spectrum. You even supported this by admitting that your carbureted car does richen up, though not much, when going from the valley to the mountains. Still very driveable but also still richer. Probably one out of a thousand guys that buys a carburetor goes to the lengths necessary to achieve the type of tuning you are talking about anyway. However a guy that goes and gets a closed loop FI system either aftermarket or off of a wreck will have a system that does the optimizing all by itself all the time every time it is driven.

 

I just love my Sleep Number bed!

Some folks like the ones made with "memory foam" (the Sweedish brand name escapes me at the moment!).

Which one is better?

Well, it depends on who we are asking and just what aspect of "better" we are considering.

There is no SINGLE answer to the above dispute.... no matter how long the answer.

 

I just gave the race cars as an example of the maximum performance that carbs can put out, that's not to say they don't operate well over larger rpm ranges and more varying throttle positions. Yeah, FI will compensate, and for the average person will perform better everywhere. But carbs do compensate as well, that's what all the design in them is about. Once you have a proper metering system developed, it should be able to adjust the AFR proportional to a VERY wide range of throttle positions and RPMs, as well as respond to atmospheric conditions(at least to some degree).

That's what developing a carburetor is all about, getting the right mix of the right systems within the metering block so that it responds the way you want it to across those varying conditions. That's also why carburetors have different circuits, to deal with those differing conditions(idle, transition, power, main etc). Do they adjust to atmospheric conditions? Yes and no, they all do to some extent just based on physics, but good ones will respond better to atmospheric changes. But yeah, it's not going to be as precise on the AFR with FI is, but it doesn't need to be. Again, it gets back to the atomization advantage for producing power.

The car doesn't perform based on what the AFR is alone, it performs based on power output at the given rpm, throttle position and load. Even though a carb may not have as ideal an AFR due to a change in conditions, the atomization still allows it to produce slightly more power whether it be part or full throttle(provided the design is capable of such). That means better acceleration and better throttle response(and sometimes better mileage). With older carbs that was mainly true at mid-higher rpm, because they had deficiencies in the design that precluded them from having their atomization advantages at low rpm, so EFI made better low end power at all throttle positions. But again, newer technology such as annular boosters has allowed a carb(a good one anyway) to close that advantage up.

My carb'd 302 makes good power at all throttle positions and all rpms, has extremely snappy throttle response at all rpm and throttle positions, and it will get around 22-25mpg on the highway, depending on how I drive. And it still needs more tuning. Obviously carburetors are more than capable of performing across a wide range of rpm and throttle position.

But in the end it gets back to 2 primary things. First is the carburetor itself, whether or not it's of good, capable design. The second is the monkey that's tuning it. If either one has a fault then performance is lost. So yes, for nearly everyone out there FI will perform better than carburetion. But when both FI and carbs are tuned properly, the carb will typically outperform EFI just about everywhere. The only thing EFI can do better is produce more very low rpm power(in nearly all cases) get better mileage in most cases, and will ALWAYS produce less emissions in normal driving situations(the primary reason that the manufacturers went to EFI).

Carburetion has been and still is widely misunderstood and underappreciated. 99.9% of what average people know about carburetion is myth, rumor and speculation. That's why I also think that nearly everyone is better off with EFI, because they'll never get a carb tuned right, and there's cirutually no one left these days that can. most dyno tuning shops these days are great at tuning EFI, but don't really know the first thing about carburetion.

And just as a side note again(not trying to prove a point here, but it was brought up and this is impressive), F1 cars powerbands are over a wider rpm range than most engines can even mechanically operate. Their best power range is from around 11,000-19,500rpm. That's an 8,500rpm operating range, on top of the other mostly unusable 11,000rpm. The few engines in the world in street cars that even operate from "off" to 8,500rpm are high end sports cars like Ferraris, and even NASCAR Cup cars barely operate much past 8,500rpm. That's extremely impressive powerplant development by any standard. 850+bhp from a 145 cubic inch naturally aspirated V8...that's what an annual $ 1/2 billion budget allows for though.

 

Discussion is good for it's intellectual value, forces people to think and exercise the brain. Iron sharpens iron and all that.

and I think you were thinking of the Tempur Pedic?

 

looks like I'm late to this thread, but just thought I'd put in that I have an edelbrock 650 cfm carb and haven't had any problems since, except for a vacuum leak at the base but the mechanic told us a trick he does is use two carb gaskets to be sure it seals, and so far so good! I've even completely removed my carb before to do some work on something and put it back on and it stayed tuned and working perfectly fine.