Compression Ratio Affects on MFI Fuel Requirements

Jeff Higgins · 01-05-2025, 02:08 PM

As many of you know, I just rebuilt my 3.0 liter and lowered its compression ratio from about 11:1 to 9.5:1 (documented here):

The Mighty 3.0 MFI Motor Has Finally Had It

I have found that the mixture was altered across the range of throttle positions and RPM. With no changes to the pump or throttle bodies, what had worked splendidly at 11:1 compression was now running quite lean under certain throttle openings at certain RPM's. The worst was in the 3,000-4,200 RPM range at part throttle, where the mix would go as lean as 15:1 or worse. High RPM full throttle was the next most affected.

I have to say, I was surprised to see this. I'm struggling with how a change in compression could have such an affect. I can see displacement changes, induction changes (throttle body and stacks), exhaust changes, cam changes, and port changes having this affect on fuel requirements, but compression? I would have never guessed. Any thoughts on why?

And yes, I did wind up having to pull the pump, remove the space cam, and regrind it. Going through my notes, it's been 14 years since the last time I had to do this. Thank God for those notes, though, as they made this go-around so much easier, knowing where I was starting, the affect of previous changes, how much material to remove to get about the change I was after, and all of that. Less than five hours from parking the car after the last test run to firing it up for the first with the new profile.

Best news, though, is that I think I got it right on the first try this time. I was very conservative regarding how much material I removed, to the point that I was concerned it may not be enough. But, well, I can always remove more, I just can't put it back. For now, though, it looks pretty good.

stownsen914 · 01-05-2025, 04:50 PM

Interesting, thank you for sharing. I might have thought that lowering the CR would lower the fuel requirement, not increase it.

Jeff Higgins · 01-05-2025, 05:58 PM

Yes, this was very unexpected. I wouldn't think there could be any affect.

MFI is a wonderfully instructive, revealing system. It's what I call a "double blind" system, in that the fuel delivery does not "see" the airflow, and vis versa. So what this situation has told me is that because it needs more fuel in those throttle positions and rpm's, it is actually sucking in more air. Volumetric efficiency has improved. It's ingesting and scavenging better, which means more power at those throttle openings and revs. Don't get me wrong, I'll take it, but I'm left wondering "why?".

The motor does feel "snappier" than before. Some suggested that may be because the old build was getting tired, but I'm convinced that is not the case. Other than one broken piston and one valve stem that was about to fail, it looked pretty darn good inside. So, here is my theory...

With pump gas as the "control", which will make more power? A high compression motor that must run compromised timing and A/F ratios so as not to detonate, or a lower compression motor that can run optimized timing and A/F ratios? Granted, the high compression motor, when fed higher octane race gas which would allow the timing and mixture to be optimized would win hands down, but that is not what I was able to do with a car that is primarily street driven. On some long trips, too, like my 4,500 mile round trip from Seattle to Santa Fe last spring. I have to be able to fill it with whatever I can get.

I think this whole adventure has answered that question. The lower compression motor, with optimized timing and mixture, makes more power on pump gas. This latest discovery, this increase in volumetric efficiency, is just icing on the cake. But I really want to understand what happened here...

rowingone · rowingone's Garage

Shouldn't be unexpected. Lowering an engine's compression ratio will make it run leaner, as less air is being compressed in the cylinder, meaning less fuel is needed to maintain the ideal air-fuel ratio, thus resulting in a leaner mixture.

Alan L · Alan L's Garage

A lower compression motor will spin easier - isn't working against the compression so much. So should breath easier.
"Lowering an engine's compression ratio will make it run leaner, as less air is being compressed in the cylinder, meaning less fuel is needed to maintain the ideal air-fuel ratio, thus resulting in a leaner mixture."
Not sure I buy that bit tho. You should actually have a bigger cc displacement. So more air in the chamber. Just squeezed less.
Alan

r lane · 01-06-2025, 07:46 AM

Any time I have changed compression in an engine, assuming other things stayed the same, timing and octane, the more compression the more power and snap. As has been stated, when you raise the compression, you shrink the chamber, so a smaller fuel air charge. But the air is compressed more meaning denser then needing more fuel. I am not a scientist, but this is my assumption. In Jeffs motor, he was not dealing with just apples. Having to run a motor on the wrong fuel and reduced ignition will not ever give that motors best. All about the package. As a coach once said,'' sometimes on given play I will not have my best 1 or 2 players on the field as they don't fit well in that particular package''. Bob

r lane · 01-06-2025, 07:50 AM

Quote:

Originally Posted by Jeff Higgins

As many of you know, I just rebuilt my 3.0 liter and lowered its compression ratio from about 11:1 to 9.5:1 (documented here):

The Mighty 3.0 MFI Motor Has Finally Had It

I have found that the mixture was altered across the range of throttle positions and RPM. With no changes to the pump or throttle bodies, what had worked splendidly at 11:1 compression was now running quite lean under certain throttle openings at certain RPM's. The worst was in the 3,000-4,200 RPM range at part throttle, where the mix would go as lean as 15:1 or worse. High RPM full throttle was the next most affected.

I have to say, I was surprised to see this. I'm struggling with how a change in compression could have such an affect. I can see displacement changes, induction changes (throttle body and stacks), exhaust changes, cam changes, and port changes having this affect on fuel requirements, but compression? I would have never guessed. Any thoughts on why?

And yes, I did wind up having to pull the pump, remove the space cam, and regrind it. Going through my notes, it's been 14 years since the last time I had to do this. Thank God for those notes, though, as they made this go-around so much easier, knowing where I was starting, the affect of previous changes, how much material to remove to get about the change I was after, and all of that. Less than five hours from parking the car after the last test run to firing it up for the first with the new profile.

Best news, though, is that I think I got it right on the first try this time. I was very conservative regarding how much material I removed, to the point that I was concerned it may not be enough. But, well, I can always remove more, I just can't put it back. For now, though, it looks pretty good.

I am getting ready to massage a space cam. Obviously depends on what you start with, but I am thinking the specified spots .002-.004'' to start. If you don't mind sharing, how much did you remove. Bob

Jeff Higgins · 01-06-2025, 10:56 AM

Quote:

Originally Posted by r lane

I am getting ready to massage a space cam. Obviously depends on what you start with, but I am thinking the specified spots .002-.004'' to start. If you don't mind sharing, how much did you remove. Bob

This is all kind of guesswork to some degree, Bob, but here is what I think the affects are when we remove material. It appears that removing .5mm (.020") results in about a .70 to 1.0 point shift in A/F. Removing 1.0mm (.039") appears to result in about a 1.5 to 2.0 point shift.

It is, of course, almost impossible to determine exactly where the stylus is under any given throttle and rpm condition. We're making educated guesses. And, when removing material, we want to wind up with smooth blends, so this will kind of muddy the waters a bit. Sometimes I will take what I think is the right amount off in what i think is the right area and it will have either more or less affect than it did when I took that much off in another area. There is a lot of intuition and guesswork here.

My advice is to go slow, and understand that you will have to remove and replace the pump and regrind the space cam several times, at least. The key is to take really, really good notes, and to mark the space cam in a permanent way so that every time you set it up to measure it, you are measuring in the same places. I took a little needle file and put witness marks on the forward facing edge, on the thin lip outside of where the big spring nests in it. Then every time I remove it, I can use a little machinist square to draw reference lines down the face of the cam with a felt pen.

I stand the cam up on its own axle, the flyweight assembly. I set my dial indicator to read the top of the cam (front of it in the car), then I use adjustable parallel bars to raise the cam by placing them between the cam and the face of the flyweight assy. I raise it and measure it at full throttle and two heights in between. Like this:

I then draw a diagram of the cam and the measurements I've taken. Here is a really old picture, with only one step between idle and full throttle. I've since learned that isn't enough fidelity, and have gone to two or three steps in between. You can see that I didn't change much, but this was the first try. I was going very slowly - we can always take more off, but we cannot put it back.

Good luck. Go slow. You will get pretty good at removing and installing that pump...

Old H2S · 01-06-2025, 01:42 PM

Jeff, I have a laser welder. Manule type, for doing very small work by hand, beam shot is .004" wide at a time on the smallest setting. I have 30 years experience doing build up work on small molds and dies.
If you want an area built up I can do it in thin layers to your needs. The end result is a very fine pebble finish that would need to be cut down with a fine stone to get a smooth surface, this is how sharp edges on expensive molds are repaired from wear.

Old H2S · 01-06-2025, 02:01 PM

r lane · 01-06-2025, 02:31 PM

Jeff, Got mit himmel, what a lot of work, but how else to get there. I assume your minus and plus numbers are starting from zero and as the pintle drops into a divot the number is a minus and as it rises past zero, then a positive reading. I would think the amount to be removed at different spots would be non linier as would be subject to a percentage of the differing amount of fuel required at each particular spot. I have not tried to figure out the difference in leverage which the cam applies to the rack at different points. That has to play into the material removal differences. Do you have some sort of data log system tapped into your AFR. Or a companion with a note pad. I am going to attempt this with CCs and a chart as I have fashioned a bench of sorts to run the pump. As you have determined, locating a spot by throttle position is not so difficult, but by RPM much more so. I appreciate your providing this info. Informative to see how others successfully handle a problem. Bob

Jeff Higgins · 03:18 PM

Very cool laser welder. I hope you won't be offended if I say I hope I never need it. I did take too much material off of one space cam when I started this adventure, so I might have been able to use it then. But, well, I just grabbed another one off the shelf and started over.

Bob, you are correct regarding my plus and minus numbers. My "zero" from which those are derived is the idle position on the space cam. Minus numbers mean that portion of the space cam is closer to its axis and therefor richer than the idle "zero", plus means further from its axis and therefor leaner. And yes, this is a fair bit of work. After some practice, though, I can remover or install the pump in less than 45 minutes. You get pretty good at that, out of necessity.

Your intuition that removing identical amounts of material from different places on the space cam seems logical, it's what we might think is happening, but it is actually incorrect. Removing like amounts of material anywhere on the space cam has the exact same affect, regardless of throttle position or rpm. There is a reason for that.

We need to think in terms of per stroke delivery. Intuitively, we think that because a motor is at redline and full throttle, it must be using the maximum amount of fuel. And it is. But the per stroke delivery is down. The only reason its using more fuel is because it is spinning faster, and there are more strokes.

If you look at my crude map of my space cam, you will notice that the greatest negative number is not in the upper left corner, which represents full throttle and maximum rpm. The greatest negative numbers are along the bottom and in the middle, which represent the low to mid rpm range. That is where the greatest per stroke delivery is needed.

The reason for that is volumetric efficiency. It's filling the cylinders better. There is more air going in, so we need to send more fuel with it. It is in this rpm range where we see any given engine producing its peak torque numbers. That's because each individual combustion event is creating the maximum power of which the engine is capable. It's in this range that it needs its maximum per stroke fuel delivery.

At higher rpm's, the number of combustion events, even though they are weaker, overwhelms the lesser number at lower rpm's and produces more horsepower. And the engine is using more fuel, but each individual combustion event is actually using less fuel.

In light of this, we need to view the space cam as to how it provides for these changes in fuel requirements. We need not even consider the total amount of fuel required to sustain a given rpm and throttle position. What is important to understand is that we are bound by the requirements of each individual combustion event, each individual stroke. Then it makes sense that removing material is linear and consistent regardless of where on the space cam we remove it. Same affect everywhere, which is actually a huge relief. It removes one very difficult variable. That's a good thing. This stuff is difficult enough...

Oh, and yes, I do use a data logger and A/F meter that is hooked up to a wideband O2 sensor. Mine is the Innovate Motorsports LM-1. Now almost 20 years old, they no longer support it, but I did download their program files while they still did. They now sell and support their LM-2. My position is that we simply cannot do this without such an instrument.