Fuel atomization: Carb vs. EFI

There are discussions about which is faster, which is easier, which is cheaper, which is.............etc, etc.

My question is: Which system will atomize the fuel better (I suspect EFI) ?

How can this be measured ? Two otherwise identical engines, one carb, one EFI, So which will produce the most power ?
Or alternately, if both engines can be adjusted to the same HP output at the same RPM, which engine will use the least fuel ?

Not sure if it's feasible (for us hotrodders) to measure unburned hydrocarbons, but that would indicate lack of fuel atomization.

Now, I read that carbs can be tuned run extremely efficiently in a certain RPM band, but likely not idle - to - peak RPM.

Of course, if the engine in question produces highest HP with a rich mixture, meaning there will be unburned hydrocarbons,
both carbs and EFI can do that, but which system can achieve that power level with a lower BSFC ?

I doubt anyone on here is going to know that answer to that. There might be a few OEM engineers who know although my guess is that there aren't very many engineers at the OEMs who have any experience with carbs. The last new car I owned that had a carb was a 1984 Honda. I think Honda switched to EFI in 1985 which is 35 years ago.

The best place to find an answer to that question these days would probably be NASCAR. They just switched to EFI a few years back so there are lots of engine builders who have experience with both carbs and EFI. Also, those guys have very sophisticated data logging systems so they probably know the BSFC of their engines to 5 or 6 decimal places.

My personal opinion is that it doesn't matter. Nobody on here is trying to build high mileage emission compliant cars. EFI and carbs make roughly the same power on the types of engines that we build.

Almost all the race motors we build get fuel injected and go across the dyno, they get broke-in with a carb and then get switched over to fuel injection, engines will almost always make more horsepower with a carb but usually overall driveablility and fuel control is better with fuel injection.

You know the answer already, it depends of course on the engine, EFI system and the tuner on the carbs
You can make a carb really good if you have the time, tools, knowledge and know how on what and where to tune on the carb, same thing on EFI
I learned that a good O2 wide ban system is a great aide on tuning all the circuits on carbs, not just the WOT results
I like to see around 14.8 AFR or leaner on hot idle, 12.8 to 13.5 AFR on light accelerating part throttle cruise and as lean as possible(15.3+) on steady light throttle cruising RPM
ls far as atomizing the fuel if you mounted the EFI injectors up high in the tunnel ram intake runners instead of down low at the manifold flange you might be pleasantly surprised on those results
I hear comments on fuel droplets falling out of the air in the manifold and I wonder what size hole that fuel is coming out of

From my understanding when Bruce Robertson aka Shrinker was still alive and posting on the Motorsport Village Forum he stated that Carbs were better because there was a constant supply of atomized fuel in the plenum as soon as the valve opened this column would fill the chamber with a homogenized mixture. With injection the injector cycles on and off to limit the amount of fuel but in so doing there will be parts of the air column which wont be completely saturated with atomized fuel leading to pockets in the chamber that are leaner that other areas.

A 5 gas analyser would tell how well the mixture is being utilized.

Its all about the quality of the mixture.

Hope this helps.

And that is why they have this, takes the cam timing into account on WHEN the injector sprays fuel. this is my Hilborn sequential EFI on my Charger.

Also with EFI I can run it leaner than when it had a carb at cruise rpms, without lean misfires

You would think carb given the time for fuel to move through a heated intake manifold,

but,

injection is now direct on some engines...................

The Factory must know something if they are moving the fuel closer to the piston.

A Fellow I worked with once told me: " put a few ounces of fuel in a 45 gallon drum, shake it up, leave it in the sun for a week and then throw a match in it.

Heat and vapor seem to be the ticket to massive combustion.

Purely atomization, carb.

I see no one understands the question.
Rather than cut-n-paste, read DV on "displaced charge factor" in any of his books.

My thoughts, and I am no engine scientist lol. I think a carb vs THROTTLE BODY EFI the carb is better, fuel flows through it 100% of the time there is no pulsing of injectors to interrupt the stream of fuel.

I think IF port injection doesn't atomize better it at least allows better control of where the mixture goes, and with some EFI systems you can adjust that per cylinder.

Most of the EFI vs. Carb testing that I've done on the dyno shows that a carb has a slight power advantage at peak power, but generally gives up some power in the mid-range. This isn't always true but seems to be true about 80% of the time. I don't worry about it or pay much attention to it since the power difference is irrelevant for the customer. If we were building EMC type engines then we would explore it.

The couple of experts I've talked to say it has more to do with the intake design than anything else. The intake for dry air should be designed differently than a wet manifold. Most aftermarket EFI systems use wet manifolds for port injection just because that is what they already have tooled up. Eventually we might see the aftermarket start to design dry air intakes. Once that happens the power difference will swing the other way. The G3 Hemi guys are seeing this.

Absolutely no doubt about this. Direct injection may change this but that's OEM only right now.

I find this fascinating. My guess is that liquid under high pressure suddenly escaping to a much lower pressure
would vaporize much quicker. The liquid shearing across the injector orifice would be much more violent
than the shearing across a venturi booster.

I would think that the Carb atomizes fuel better just by it's design of mixing air into the fuel stream. The Injector is shooting raw fuel into the air stream, although different injectors try to break up the fuel into a fine mist.
To be fair, in a test you would have to test the carb against a TBI EFI system (non-port injected) so both would have the same intake manifold distribution and timing.
With port EFI, each port can be adjusted for fuel and injector timing, so manifolds can also be designed differently. I think usually some type of tuned port length (or adjustable port length) single plane design as the is not the concern of port signal to the carb or fuel fall-out. I think the direct port injection allows higher cylinder pressures, but I think the fuel is not atomized very well. Good question to ask on the SAE blog?

Not sure I agree or if you can support your premise here, and is maybe a separate topic, seems there is a lot more at play in the process then strictly atomization, or that it can be determined by unburned carbons measurement alone.

Let's not forget the cooling effect of a "wet" manifold compared to hot incoming air prior to the injectors.............

Direct injection gets you the benefit of in cylinder charge cooling, and should make more power than a carb, with the obvious tradeoff of complexity. Direct injection is out for racing applications... just not common at the 3000+ hp level you are playing with. Pretty much all F1rally/IMSA/enduro/whatever are using Bosch Motorsport DI fuel systems on them. You can also claw back atomization deficiencies with injector placement. Early F1 fuel injection had the injectors pretty much dead center, aimed straight into the individual throttle bodies.

A few OEM's have both port and DI fuel systems on them now, and that is done primarily because of soot production issues at high load. Newer emissions standards depending on the country have a standard for tailpipe soot, which led to the second fuel system.

In the aviation world, a carburetor in an engine running at constant rpm and power level (above the transition circuit power level) will make as much or more power than FI. The FI will have an advantage in varying levels of power and lower power levels because it is more adaptable than the idle circuit and intermediate circuits of a carburetor. Airplane carbs have variable mixture via an adjustable port for the main jet in the carb. 1935 technology at its finest.

There are a lot of different (good) questions and answers being discussed......but to simply answer the OP's short Q -

The simple answer is...EFI. It's down to fuel pressure, and any injection system has higher fuel pressure than a carb. Modern Direct injection systems run at crazy pressure. The reason - better atomisation.

What you do with it after it gets atomised is part of the discussion here, but better atomisation - EFI every time.

Full atomization?

The "technology" to produce any size droplet is now over 100 years old.
As I said - is that what you want?

Direct Fuel Injection allows you to control the timing of the fuel, and since the injector is downstream of the intake valve (directly into the cylinder) you have more flexibility with cam profile and compression ratio. No worries about pre-ignition when there's no fuel present.

I did a study on fuel atomization at Chrysler and the main factor was pressure

Which technology was that?

Polyspheric, I find your posts excellent and informative, but I do not understand your question here.
Why would it be desirable to have poor fuel atomization ?
The more vaporized the fuel, the more complete the combustion, unless of course the mixture is stoichiometrically rich.

The fully atomized fuel displaces its equivalent volume of air in the charge mass; gasoline as a gas is many, many times larger than as a liquid. In engines with poor breathing (DV was specifically referring to the BMC Mini) air is more important.
A hemi would have the opposite argument: large intake port means fuel drop-out may occur if they're too large.

OK I see now.

Can anyone please explain how a carb atomizes fuel better than fuel injection.

For the sake of my question both the carb and fuel injection are tuned properly.

The fuel in MPFI is injected onto the back of the valve while it is closed so the fuel is the first thing in as the valve opens. Think of it like putting cream in your coffee, if you add the cream to your cup first(small amount) then add the coffee(larger amount) then just the act of pouring the coffee in will mix it all up and you don't need to stir it at all, on the other hand you pour the coffee first then squirt in the creamer you are gonna have to stir it because it won't mix on it's own.

Carbs on the other hand spew the mix in the chamber at the same time, it also has more time in the runner to evaporate a little more or for the fuel to seperate out if the manifold or port is poorly designed.

EFI also comes out at a much higher pressure and this also helps to atomize so 6 of 1 or 1/2 dozen the other way.

Direct injection is more about MPG and emmisions than power or even atomization, it allows the whole combustion chamber to filled with a homogenious mix under WOT or more importantly, part of the chamber can be filled with a real good mix near the spark plug while the rest is just air, this minimizes heat rejection into the metal because the air acts like insulation, you can intentionally have lean and rich areas at the same time in the chamber to improve MPG and emmisions, this is not really utilized under WOT because a carb or port injection both can do a fine job of filling the entire chamber with a good mix.

In my personal opinion if MPG were the primary concern you would want all the fuel vapoorized and evenly mixed so you would want the high pressure fuel atomization of MPFI and a longer time in the manifold that you get from injectiing it at the beggining of the runner instead of at the valve, my TBI motors I have owned always got the best MPG, chevy, dodge both got better MPG with throttle body injection than MPFI magnum and vortech style engines at the cost of power.

I don't think anyone on this thread knows the answer to that question. That is a very complicated technical question and nobody would be able to prove it one way or the other without a lab full of equipment. A carb works opposite of injection so the behavior is totally different. In a carb the low pressure area in the venturi sucks the fuel out of the bowl. The pressure differential in a carb is only 5 or 6 psi while an injector operates around 50 psi. The metering block in a carb adds air bubbles to the fuel which helps. There is no air bubble system in an injector.

If someone really wants to know the answer to this question they would need to talk to an injector engineer at Bosch. My guess is that they know how to measure it and and could explain it.

Thats the issue here... if you look at the question you can have 2 answers.. a injector by itself will atomize
better but in a full system as in TBI you have a intake(mixer area) a MPI has higher pressure than a TBI
and the MPI will atomize better
EDIT
I had 10 years of flowing injectors and building the flow stand

I was surprised with the level of uncertainty in this thread.

Liquid fuel does not burn. It has to be vapor to burn. New cars are all about low emissions. Our favorite classic mopars had high emissions because carburetors while good at making power are not so good at low emissions. If a carburetor could atomize fuel as well as fuel injection - new cars would sport carbs.

The smaller the fuel droplet the more surface area for evaporation. Carbs drip the fuel, fuel injection sprays in a fine mist. The mist has more surface area resulting in better mixture with the air. Fuel injection is also constantly "jetting" the injectors. Making small adjustments in pulse width for any throttle position or engine load. A graph of the oxygen sensor signal on a properly running engine will cycle just above and just below stoichiometry. If the oxygen sensor signal spends too much time rich or lean it flags a code. If the fuel trims get too far rich or lean they flag a code. Carburetors are good but they are never as good as injection.

Carbs always run rich - especially on choke. That extra fuel is what wore out old engines. It washed the oil off the cylinders causing the ring ridge. It then ended up in the oil pan diluting the oil wearing out the bearings. Modern engines are barely broke in at 100,000 miles. It's because of the fuel injection mixing the air/fuel well enough that after combustion minimal fuel is left over.

Plus the blocks are much harder now days

Aren't most of the modern retrofit throttle body EFI systems now running basically 8 MPI injectors bunched together and running high pressure (50PSI or so) and squirting it in the throttle body?

Another benefit for modern cars is that a dry manifold can be made more effective at ram tuning and physically fitting into any given engine compartment than one designed to carry a mix of air with fuel suspended evenly into it. The beer barrel intake on a magnum or modern hemi is a perfect example, they have even length and shape from one runner to the next as well as a long length that would let fuel drop out at low velocipy. They would both have terrible distribution with a carb (either a down draft on a magnum or side draft on hemi).

The last time I work with either system the TBI was 14# and multi point was 55#.. I started out flowing
single injectors at multiple voltages and seeing the differences in the area around each zone of
the flow pattern and the different duty cylcles/pulse width

Holley Sniper runs at 58 psi while Holley MPFI runs at 43 psi. But the Sniper injectors fire into an annular ring and then the fuel dribbles out of there. So it isn't really injecting directly into the air. I have no idea how a Sniper compares to a carb in terms of atomization since the dynamics are very different. The average guy has no way to measure this stuff.

I have no way to measure that stuff now days without that flow bench I built up

Or whole factors of normal sea level atmospheric pressures.

There's really no debate to be had on which technology atomises better. You've had at least two people who have worked in the industry say the same thing (Mr P Body and myself). Higher pressure gives better atomisation. EFI is ALWAYS higher pressure therefore, leaving the carb or injector, better atomisation, every time, period.

Per my original reply - what you do with it after that is getting a lot of good discussion here, but there's no uncertainty about which device atomises better.

I am still in doubt about which system, EFI or a dual quad tunnel ram set up, is better for a all out drag race motor Hence NHRA Pro Stock not going faster now with EFI than they did with dual carbs on a tunnel ram intakes
I remember several different mechanical fuel injector companies testing the location of the injectors heights in there all out race set ups for dirt and asphalt track roundy round all out race cars

My understanding is that the higher up you can place the injector the more power you can make, downside is they are a little harder to dial in for street driving with the injector way up the intake.

It would have been interesting to see what happened with pro stock if the EFI conversion didn't require the loss of the scoops and tunnel rams. Holley makes a 4500 bolt pattern throttle body that is 2500CFM, a pair of those on a tunnel ram would be fun.

If EFI is so much better(which I don't believe it is in every case)then why did Pro Stock slow down across the board. Plus there's some well known heads up racers I know that sold there's and went BACK to carbs and picked up LOTS of power......My opinion on the 2 places it shines, 1, on our daily drivers get to work stuff and 2, super high end high power(2000+hp)stuff but in between that carbs shine and I've been part of it personally...........

Pro stock didn't just swap to EFI, they also got a 10,500 RPM limit, lost the scoop, and run a spec throttle body and intake tube.

Not to say that the carb still wouldn't make more power but I feel it would be MUCH closer or equal had they got to keep the tunnel rams.

But does that mean that it produces a better homogenized mixture?

Exactly. When I look down the throttle bores of a Sniper it drools big drops of fuel into the intake manifold just like a carb does. The injectors in a Sniper do not spray fuel into the intake which is what most people seem to be assuming.

My understanding is that a 5 gas analyzer would tell you how the fractions are being converted. What happens to the flame front when the mixture isn't ideally homogenized? Just because you atomize fuel doesn't mean the mixture is ideal at the time of ignition. I'm no expert but maybe the question of atomization is just one part of the puzzle.

Maybe a thorough reading of Ricardo's work might shed some light on the matter

Would be interesting to make a clear manifold and actually see the differences in fuel delivery between a Holley a Carter TQ/Rochestor Qjet and a Sniper.

Had this discussion with a friend about the difference in performance of efi and carbs and he told me about his mate who went from a well tuned Rochester Qjet to an expensive injection system and only gained 3 foot pounds of torque in the mid range and nothing at the top all measured on the dyno. He did say the only real benefit was when starting it cold it idled better. From memory it cost him 3K to convert over so not exactly bang for buck.

I still have some tuning to do, because self tuning EFI is more of a selling point & not a true reality, but the EFI HILBORN 8 Stack I have on my Street Strip car is AWESOME, Individual runners, so the low vacuum, drivabilty issues from a large duration cam are gone, cam is a hyd roller 255, 258 @ .050 & it idles like a stock 440 & drives just as smooth. a carb on a common plenum intake CANNOT do what this EFI set up does, also I can change runner length to tune for power @ different RPM's, as I said earlier I can run much leaner without lean misfires, that tells me better fuel atomization & Obviously better fuel distribution than my ported Indy SP intake. Will keep Y'all updated on my progress at the track.

The RPM limit didn't hurt the GM architecture.

The hood scoop was on of the MOST cried about pieces of a Pro Stock car ever. WJ sniveled about them for years and claimed a huge gain in MPH if the big "mail box" was gone from the car.

EFI retarded Pro Stock in more than one way.

AndyF,
I had no idea the Sniper was a "drooler" not a "fogger" !!
To me, EFI should not only meter fuel, it should also atomize it.

A buddy of mine in the R&D biz did that years ago. First thing he learned was that fuel staying in suspension in a wet manifold is basically a myth. There is a cloud of it directly under the carburetor and then 90% of it goes directly to the floor of the intake manifold.

Kevin

Atomization does not necessarily = more power.

Direct injection (nozzle exits directly into intake runner bowl) does not make the most power of FI configurations, for a traditional V8 engine.

Google "Darin Morgan wet flow test" and there is a link to a gearhead podcast where he's interviewed for about an hour. One of the topics is FI nozzle placement, in which he talks about why Sprint Cars ended up running both direct injection for improved throttle response AND additional injectors farther upstream in the runners, because the upstream location is more flexible in its tuning parameters and makes more power at high(er) RPM.

I never heard of direct injection above the valve. Direct injection has always been in-cylinder nozzles at extremely high pressures. Port injection, as used on sprint car engines generally have the nozzles entering the port from the exhaust side of the head, straight into the bowl. Other than the clarification, I think you/ he is spot-on. The higher the rpm range, the farther away from the intake valve the injector should be. F1 engines live life way above 10,000 rpm. They have had injectors at great distances for years.

Yes, that's what I mean; I'm not familiar with all the correct terminology.

https://www.youtube.com/watch?v=y2iBbwocYZw

this is how the most expensive highest tech engines are done

I would imagine the "Mist" component could change direction very quickly to go to the cylinder with the opening valve. It would be harder for the fuel on the walls of the manifold to change direction.

Anyway a heated manifold would help vaporize some/most of the fuel on the walls and floor of the intake helping it find its way into the cylinder in a better state than liquid.

Well known road race car here in Australia had the injector nozzles mounted before the entrance of the IR trumpet.

I remember they changed a few things but most interviews I saw it seemed the drivers/tuners preferred carbs and a forward scoop plus I'm sure singing the HEMI'S way up is missed as well...........I saw the spec sheet on Erica Enders carbs and it was quite eye opening..........