Logo
FULL SITE MENU

Mystery Carb Reaction

fastjeff

fastjeff

Gold Medal Contributor
This one has me guessing. Been messing with engines for half a century, and I'm totally baffled.

Marinette 32 footer with twin Mopar 360 V-8s. Port motor has a Flo Scan device that monitors fuel consumption. (I'm too cheap to buy another sensor for the starboard engine.) Going along in rough seas last week, at 2,500 rpms cruise speed, I watched the fuel usage rate CONSISTENTLY increase from 9 gph to 10 as the boat slowed down (from climbing a wave). After the wave passed, however, the rpms came back up to 2,500 and the fuel usage returned to 9 gph. And it did this consistently for hours.

1. The carb is NOT flooding--that I know for certain, both from exhaust sound and feel.

2. The boat climbing the waves was gentle--like smoothly encountering a change in the hill your car is going up.

3. The throttle is fixed in position at that rpm and wasn't moving on its own.

To me, the fuel consumption should have dropped as the rpms dropped, for less intake strokes per minute were occurring, but it went UP! And it did this consistently for over an hour.

Anyone have an explanation?

Jeff
 
do the same test but hook up a vacuum gauge to monitor the pressure in the intake manifold....

i'll bet the vacuum will drop when the wave is encountered and the fuel consumption will also start to rise...in other words, the fuel consumption will be correlated to the intake manifold's pressure....
 
I agree, big mystery. On an engine with a fixed throttle (no governor) if RPM drops it is not sucking as much air. Agree with Mark that vacuum will drop, but that means air flow though the carb will drop (for the same throttle opening). And air flow (velocity) determines how much fuel will be sucked out into the airstream, so one would expect fuel flow to drop. My first reaction is that the above is a fact (I have been led to believe), so logically the Flo Scan must be giving erroneous readings. I would look there first. Maybe the angle of the boat and the level of the fuel in the tank vis a vis the flow meters has some effect? (I doubt it but reaching for straws....) Maybe voltage related? Interesting question.
 
Good answers. Thanks for the responses.

1. There is virtually no vacuum in a boat motor working hard--like driving up hill al the time in your car. What little there is WILL decrease still further when the load on the motor increases.

2. The Flo Scan is working very well and has been for years. (It'll probably blow up next time I run the boat!) The reading is very smooth, with but a slight lag (understandable) as the wave is climbed.

Here's what the prevailing opinion is (and I'm not 100 % on board with it--yet): The carb (Edelbrock 1409) has its floats hinged in the back. As the bow rises upon climbing a wave, the angle of the carbs increases and the floats drop slightly. That causes greater fuel flow (temporarily), making the motor then run a bit richer. When the bow comes back down the rpms and fuel flow returns to normal.

Sound plausible?

Jeff
 
I don't believe you will see a 10% change in the flow rate just because of the angle of the fuel in the float bowl...

Remember, a carb works due to pressure differences...and the volume of the engine doesn't change - displacement is fixed...

Which step up springs are in those Edelbrock's?
 
I don't know about the float, but it seems possible that in carburetor design that some factors (internal plumbing, jet location, fuel level in bowel, etc.) might cause the device to flow more (or less) under acceleration. If a designer was aware of this he would choose a design that actually flowed more when the vehicle was accelerating (g force pointed on an angle backwards from straight down). That would be a good thing. In a boat if the bow rises the effect on g force direction would be exactly the same as in a car under acceleration. Maybe your Edellbrock has this feature in its design. Maybe all carburetors do. Never heard of it but that doesn't mean that the industry does not do it. Just another possibility.
 
.."It all has to do with the brief loading and unloading of the propeller. "

Agreed. but why does the fuel usage rate increase with the load (and visa versa)?

..."Which step up springs are in those Edelbrock's? "

Stock "orange" springs, that may as well not be in there. I doubt if the carb ever sees the economy needle mode except for the brief moment when the motors are throttled up to plane off.

Jeff
 
Perhaps Apples/Oranges here.

My Dodge Cummins truck has the overhead feul consumption display in the cab.
I suppose in a crude fashion, this is a low budget fuel flow meter... so to speak.

There is a hill that I routinely drive on my way out to a property that we own.
I can approach this hill and climb the hill with no change to my throttle position, yet this meter will show a substantial decrease in fuel economy while climbing the hill.
IOW, my GPH will increase, or my MPG will decrease, until I crest over the top of the hill.
 
fastjeff said:
.."It all has to do with the brief loading and unloading of the propeller. "

Agreed. but why does the fuel usage rate increase with the load (and visa versa)?

..."Which step up springs are in those Edelbrock's? "

Stock "orange" springs, that may as well not be in there. I doubt if the carb ever sees the economy needle mode except for the brief moment when the motors are throttled up to plane off.

Jeff
Click to expand...

Are you actually looking for the reason why fuel consumption increases with more load and decreases with less load without moving the throttle?
 
makomark said:
do the same test but hook up a vacuum gauge to monitor the pressure in the intake manifold....

i'll bet the vacuum will drop when the wave is encountered and the fuel consumption will also start to rise...in other words, the fuel consumption will be correlated to the intake manifold's pressure....
Click to expand...

That would be a good data point. An IC engine running variable load/constant throttle should vary fuel flow strictly by rpm. As rpm reduces with fixed throttle, the "pumping" action of the engine is less, the manifold vacuum should also be weaker. Of course, the carb knows not of rpm or loading, only pressure differences and should respond only to engine vacuum, given a throttle setting. In Jeffs case, the unexpected is happening. We would expect less vacuum with less rpm, and thereby less fuel flow. It leads one to believe the mixture is becoming richer with less rpm, and that I find unexpected. The other variable is the pitch attitude of the boat. Are the carbs reasonably level at plane? If they are tilted back at level plane, and then up a wave presents even more backward tilt, could be the mixture is being messed with. OR, the time constant of your fuel flow measurement is not in sync with the wave. That is, there is some delay in the response. OR, the dash gauge (analog) is sensitive to acceleration of the boat. OR....??

I will say my 454's, also with floscans, will reduce fuel flow when just the shifter went to forward, throttle staying at idle. I've done tons of high swell crossings, but can't remember what the flow did up/down the swells. Don't think the boat had enough time going up or down to get a stable reading.
 
..."Are the carbs reasonably level at plane? If they are tilted back at level plane, and then up a wave presents even more backward tilt, could be the mixture is being messed with. OR, the time constant of your fuel flow measurement is not in sync with the wave. That is, there is some delay in the response. OR, the dash gauge (analog) is sensitive to acceleration of the boat. OR....?"

Excellent response! Here's answers to your questions.

The carbs are indeed tilted backwards a bit on plane. (Always meant to correct that, but...) So when the bow goes up, they tilt back even more, and the floats have to drop a bit, letting more fuel in. That should make the motor run a bit richer for a second or so, but the flow reading stays high until the boat and motors speeds back up.

There has to be a lag in the reading, since its sensor is located several feet from the fuel pump.

Weird, eh!

Jeff
 
The carb is designed to meter fuel based on the mass of air flowing thru it...and the "vacuum" in the venturi isn't always the same at the vacuum in the intake manifold.

Intake manifold vacuum is a key indicator of the load on the engine...the engine will run at the maximum rpm possible for a given throttle opening so there are a lot of throttle-rpm combinations.

For a fixed throttle setting and a given load, there will be a steady state fuel flow...increase the load on the engine and the intake manifold vacuum will drop and the fuel delivered by the carb has to increase to maintain the desired air-fuel ratio... (vacuum drop shows more air mass is entering the engine)...

Look at a fuel map for a fuel injection system - as the MAP rises (intake manifold vacuum drops) fuel delivered increases for a given RPM...and comparing the fuel delivered at any two RPM settings will require the MAP values...
 
makomark said:
...increase the load on the engine and the intake manifold vacuum will drop and the fuel delivered by the carb has to increase to maintain the desired air-fuel ratio... (vacuum drop shows more air mass is entering the engine)...

Look at a fuel map for a fuel injection system - as the MAP rises (intake manifold vacuum drops) fuel delivered increases for a given RPM...and comparing the fuel delivered at any two RPM settings will require the MAP values...
Click to expand...

Let me poke at this a bit:D

I think of engines as a pump, in fact, an air pump. As one increases the load, but maintains throttle, I think we can agree that the rpm will decrease. Now, will the MAP go up or go down? I believe the vacuum will decrease (heading towards atmospheric), directly due to the lower rpm/fixed throttle orifice. I can't quite get my head around the notion of "loading". If we load the engine with higher and higher torque, using a test stand, or climbing a steep wave, keeping throttle position, your pump is slowing down, your ability to pump air thru the carb is less. The carb or MPI or whatever is keeping the mixture at 15:1 +/- , less gas, less air. Is this logical?
 
Totally. IF this was a EFI engine, one might expect the computer to do some enrichment, but it's not--just a dumb carb. Also--if this was a car with cruise control--the computer would open the throttle a bit and up would go the fuel usage.

Slick mystery, eh!

Jeff
 
makomark said:
The carb is designed to meter fuel based on the mass of air flowing thru it...and the "vacuum" in the venturi isn't always the same at the vacuum in the intake manifold.

Intake manifold vacuum is a key indicator of the load on the engine...the engine will run at the maximum rpm possible for a given throttle opening so there are a lot of throttle-rpm combinations.

For a fixed throttle setting and a given load, there will be a steady state fuel flow...increase the load on the engine and the intake manifold vacuum will drop and the fuel delivered by the carb has to increase to maintain the desired air-fuel ratio... (vacuum drop shows more air mass is entering the engine)...

Look at a fuel map for a fuel injection system - as the MAP rises (intake manifold vacuum drops) fuel delivered increases for a given RPM...and comparing the fuel delivered at any two RPM settings will require the MAP values...
Click to expand...

Exactly.
 
..."For a fixed throttle setting and a given load, there will be a steady state fuel flow...increase the load on the engine and the intake manifold vacuum will drop and the fuel delivered by the carb has to increase to maintain the desired air-fuel ratio."

With a carb there is no mechanism to automatically increase fuel flow like that!

With the increased load, manifold vacuum decreases (ie; become higher pressure). That reduces air flow through the venturis, so LESS fuel should then be drawn in. What happened was the opposite of that.

Jeff
 
Jeff, as posted earlier I am in agreement with you. When an engine with a carburetor slows down without moving the throttle plate less air flows and less fuel should flow. After thinking about it, I really do not believe that the slope of the wave would have a measurable effect on the fuel air mixture produced by the carburetor since cars go up and down steeper hills and accelerate and decelerate all the time (which would have the same g force effect as a slope). Surely they have solved this problem if it ever was a problem.

So I would still bet on the issue not being in the engine but in the Floscan. They are not positive displacement devices since they claim you will get fuel flow even if the rotor fails and does not turn. And they must have very sensitive senders to measure the low flows that the do. If they do change a little with the wave slope, the up and down motion, the fuel sloshing in the tank, etc. then over a few minutes that would all average out and you would still get accurate average results. And does your gage measure in 10ths? If not maybe you are seeing rounding errors on small changes (9.46 going to 9.52) in the readout even if the computer has an exact number for the totalizer.

It is easy for me to speculate since I have never had Floscan but like you, I have fooled around with a lot of engines. What you really need is a sea that will allow you to go up hill for an hour or so you can tell if consumption is really higher. ;);)

CaboJohn
 
diver_dave said:
Let me poke at this a bit:D

I think of engines as a pump, in fact, an air pump. As one increases the load, but maintains throttle, I think we can agree that the rpm will decrease. Now, will the MAP go up or go down? I believe the vacuum will decrease (heading towards atmospheric), directly due to the lower rpm/fixed throttle orifice. I can't quite get my head around the notion of "loading". If we load the engine with higher and higher torque, using a test stand, or climbing a steep wave, keeping throttle position, your pump is slowing down, your ability to pump air thru the carb is less. The carb or MPI or whatever is keeping the mixture at 15:1 +/- , less gas, less air. Is this logical?
Click to expand...

DD - you are looking at the "pump" strictly from a volume point...the A:F ratio is driven by the MASS of air, not the volume....and because air is easily compressed, the two need to be examined individually (where with an uncompressable fluid like water, the volume and mass can be treated the same)...

I think it is easiest to look at the A:F maps for an EFI computer as they already exist and are known to work...and they come from many sources...but they are very similar.

Fixed rpm and fixed throttle...load increases and vacuum drops (MAP increases) as you noted...this means the mass of air "in the engine" has increased even though the volume (rpm driven) has decreased...the "extra" air mass means more fuel will be delivered and thus consumed....the carb does this automatically where an EFI does this by is programming (using the fuel curve)...

Though not convenient on most marine setups, you can also measure the pressure in the exhaust pipe to assess the mass of air an engine is consuming...but that gets a lot more complex due to the combustion process...
 
Last edited:
fastjeff said:
...With a carb there is no mechanism to automatically increase fuel flow like that!
Click to expand...
Yes there is...its the change in the vacuum at the feed point in the venturi where the main fuel flow occurs...

fastjeff said:
With the increased load, manifold vacuum decreases (ie; become higher pressure). That reduces air flow through the venturis, so LESS fuel should then be drawn in. What happened was the opposite of that.

Jeff
Click to expand...
The volume may have been reduced a bit (rpm driven) but the higher pressure is due to more air molecules being "under" the throttle plate; ie more air mass has passed thru the venturi and it "sucked" more fuel along the way...
 
Gosh, this is really interesting!

Let me see if I have this straight: Under the sudden load increase, the engine slows down a tad and the manifold vacuum drops even lower. This we agree on, right? Now, since the manifold vacuum is lower, that creates a slightly higher differential pressure (atmosphere to intake manifold) which allows more air to enter the motor--and that draws in more fuel!

Sound good?

Jeff
 
The intake manifold's vacuum dropping reflects a lower the pressure differential across the carb....lower vacuum on the gauge means its connection has a pressure closer to atmospheric..

That vacuum drop is an absolute pressure rise and since the intake manifold's volume isn't changing, that means more air mass (weight) has go in and filled the intake....as it entered the carb, it grabbed some fuel for the party...
 
makomark said:
The intake manifold's vacuum dropping reflects a lower the pressure differential across the carb....lower vacuum on the gauge means its connection has a pressure closer to atmospheric..
Click to expand...
Agreed.

makomark said:
That vacuum drop is an absolute pressure rise and since the intake manifold's volume isn't changing, that means more air mass (weight) has go in and filled the intake....as it entered the carb, it grabbed some fuel for the party...
Click to expand...
I'm not there on this. At rest (engine off) is the limit case of this, where there is LOTS more mass, maximum ever, and there is NO carb action to add any gas at all. Obviously, the carb doesn't measure or respond to mass of air in the intake, but velocity of air from top to bottom side. Note that at high altitudes, carbs run rich, so I'm not onboard with the idea that carbs respond to mass flow; I believe (at least now) that the venturi responds to velocity caused pressure differentials.

If you load the constant throttle engine with more and more torque, approaching lugging and very low rpm's, what happens?

Lets look at a MPI fuel/air map. Anyone have a good 3D graph to look at? One of the benefits of FI and a mass flow meter (hot wire or moving vane) is that they ARE measuring molecules of air, not simple venturi based pressure reactions.
 
Last edited:
Ever heard of a vacuum operated power valve? I believe if throttle plate doesn't change and engine load increases you would have to increase fuel ratio to keep engine from running a lean condition and possible engine damage. Not familiar with the Edelbrock marine carb but if anything like their automotive carb it is a rebooted carter. It has to have some kind of fuel enrichment system for load changes with out a change in throttle plate position. Just out of curiosity would the Edelbrock strip kit work on marine applications?
 
Carburators operate on Bernoulli's Principle, where the relative pressure of a gas is related to the velocity of the flow. The air pressure in the fuel bowl (at atmosphere) forces fuel though the jets because they are seeing a lower pressure due to the velocity of the air entering the carb. Higher velocity, more pressure difference, resulting in more fuel flow (and vice versa). Actual mass flow is not involved and manifold vacuum is not involved (although vacuum changes will quickly result in air velocity changes).

MPI engines use a computer to calculate what fuel flow should be based upon absolute pressure in the intake manifold (less vacuum means the throttle is allowing more air flow so it uses a look up table to determine the correct injector on time. This look up number is modified by data from other sensors (temperature and throttle movement being the main ones) before being sent to the injectors. So MPI engines ignore the velocity of the air entering the manifold. Trying to figure out what is going on in Jeff's engine by considering how MPI engines respond really is not relevant.

CaboJohn
 
The Edelbrock 1409 has vacuum operated metering rods, but they aren't in action in a boat since it's "always going up hill".

Jeff
 
CaboJohn said:
...Higher velocity, more pressure difference, resulting in more fuel flow (and vice versa). Actual mass flow is not involved and manifold vacuum is not involved (although vacuum changes will quickly result in air velocity changes).
Click to expand...
I think you will find the air mass moving is what causes the pressure differences....Also, because air is compressible, you have to consider more than just one variable....

CaboJohn said:
MPI engines use a computer to calculate what fuel flow should be based upon absolute pressure in the intake manifold (less vacuum means the throttle is allowing more air flow so it uses a look up table to determine the correct injector on time. This look up number is modified by data from other sensors (temperature and throttle movement being the main ones) before being sent to the injectors. So MPI engines ignore the velocity of the air entering the manifold. Trying to figure out what is going on in Jeff's engine by considering how MPI engines respond really is not relevant.

CaboJohn
Click to expand...
I disagree....An ECU's fuel map is very relevant as a FI system has the same function as a carb...they have been around for decades and have a track record....the point is if an FI system delivers adequate engine performance then it has to be flowing the correct amount of fuel....and remember, the A:F ratio is calculated by mass, not volume or weight....as the air density changes (due to varying pressure differences - intake vacuum), the fuel metered into the air stream will change....
 
@Makomark....I perhaps did not say it very well. The carb does operate on air velocity not mass flow (you can Google it). For any given atmospheric pressure, throat velocity and geometry there would be a resultant air mass flow so they are related. And of course fuel air ratios are the ratio of fuel mass to air mass so that is important. But carbs do not measure or correct for mass flow, it is just a result of the velocity. That is why when atmospheric pressure changes (as with a car at high attitude) you need to change jets to get correct fuel air ratio. Not an issue with boats (except maybe Lake Tahoe or someplace like that). In an MPI engine one of the inputs is atmospheric pressure so the computer does correct for that to maintain correct fuel air ratio. Now whether some ECU units detect a drop in engine rpm and manifold vacuum without a change in throttle, and conclude load has gone up and therefore add more fuel....I do not know. But I don't believe that explains Jeff's question.

So to get back to Jeff's question, he wonders why fuel flow rate of his carburetor engine seems to go up (according to his Floscan) without any throttle change when he believes air flow rate is going down (because of an rpm drop) as load increases. I have yet to hear a convincing explanation of why this happens (and that includes the guesses I have thrown out). Still a mystery to me.

CaboJohn
 
John - Air compresses so velocity is NOT the only variable to pay attention to...I took your advice and found an explicit statement; see link below:

http://en.wikipedia.org/wiki/Manifold_vacuum#Manifold_vacuum_vs._venturi_vacuum


As we all know, just because it is on the internet doesn't make it factual...that's another reason for the suggestion to explore fuel maps - of course, we will have to find fuel maps for valid engine sizes and calibrations...

Maybe Jeff can take his handy vacuum gauge and verify that the manifold vacuum does indeed fall when the load is applied as well as note the drop in rpm.

Your example about external (operating) pressure changes introduce yet another variable - the atmosphere...my earlier comments were directed to a carb, factory calibrated to operate at sea level, for a given engine, operating in that environment. The jets need to be changed in that scenario because the air is less dense to begin with so changing the jets is the cost effective approach as opposed to recalibrating the carb (via the air bleeds and other factors)....An EFI system is clearly more flexible in the variable altitude regime as it is self compensating.

The last variable I'd worry about is the power piston springs...as best as I can tell, Edelbrock rates those "stock springs" at 5" (Hg inferred). Unfortunately, those springs have a "starting" and "stopping" point when the rods begin to lift and when they are fully up. I can't find anything in the library that clarifies how Edelbrock intends for their ratings to be applied...If it is fully rich at 5" Hg, it is quite possible that the rods start to move ~ 8"Hg...yet another data point Jeff could provide.
 
I think a little empirical data will fix our deadlock. If someone could take out their twin engine inboard, run at normal cruise, put a vacuum gauge on one engine. Then, simply stop the OTHER engine. Now, the remaining engine, with untouched throttle receives a higher loading, the rpm will drop, and we will know with certainty if the MAP goes up or down. My $ is on "weaker, or less vacuum". Based on the actual, we should be able to take this to the next level, and predict the air flow and then the fuel flow.

I can only wish that person could be me. But, I'm 1400 miles from the boat..:(
 
Back
Top