First off a very large majority of the A/F mixture is not close to being burned before the piston will reach a point where the space starts to get very small between it & the head. I have seen discussions, not only on this forum, but many others, that discuss what is an acceptable squish before it will start to detonate on a given fuel, compression ratio, etc.

The effects of squish become more & more pronounced as RPM increases. The greater the RPM the more force that goes into the A/F mixture, or more importantly, rate of acceleration. the greater the rate of acceleration the more turbulent it becomes. The more turbulent it becomes the more ignition advance is needed to stabilize the propagation. This requirement of more advanced ignition timing is what lead to the diminishing returns of RPM rise in reflation to torque & BSFC. This is due to pumping losses. Also, the more fuel that is burned due to the need to stave off interrupted propagation means, quite simply, there is less fuel available to be burned ATDC. Increasing the net loss in the system.

The richer mixture requirements is because of homogenization not being uniform enough. This ensures that an area under squish is not "lean", thusly staving off detonation. On the opposite end of the spectrum, an overly rich mixture, not homogenized & well vaporized (only so much heat energy to go around to vaporize the fuel). Poorly vaporized fuel, is extremely proned to detonation. This is why octane, in the form of heavier hydrocarbons, is required. Heavier hydrocarbons are harder to ignite, therefore requiring even more ignition advance.

Many view the "ability" to run more ignition advance is a good thing. Not the case as this isn't an ability, but a requirement. A requirement that increases the pumping losses.

 

If it has a point of diminishing returns, then it also has a peak. Whether it's the fuel mass vaporization/homogenization efficiency, or the volumetric efficiency, or the geometric efficiency of the rotating assembly, it all has a point of diminishing returns, and it all peaks.

We are interested in the peaks, because we are trying to get all these various and independent efficiencies to work together... Efficiently.

IF peak horsepower occurs at peak fuel mass vaporization/homogenization efficiency, once again IF, and we find that you can manipulate this point with, say, 8 hole injectors instead of 5 hole, or whatever, then we have truly accomplished something.

Remember, this discussion was to explain the WHY and WHAT of where horsepower peaks. I can only assume that we want to know what and why, so we can figure out a way to control it. Knowing exactly how to manipulate the horsepower peak would be mighty useful.

 

Very well said. I think that intuitively, many have learned how to control it via cam, valve timing, head flow, compression, internal resistance, but not with a real understanding of what function is driving the power fall off. More just a working knowledge that is does under various conditions, and the know-how to work with the limitation. And I'm sure there is super expensive digital dyno's that will take basic information and spit out a pretty good estimate. And that anecdotal knowledge is valuable to say the least.

Just remember, peak power and peak efficiency do not always correlate. Simple example - 15.3 AFR is the most efficient, 13.0 makes the most power. Another simple example - hotter running engine is more efficient, and colder running engine makes more raw power.

Peak power by definition will be less efficient, because at this point, you can add as much energy as possible to the system, and it simply will not put out any more power. This is by definition an inefficient use of resources, designed to maximize energy extraction at all costs. Peak efficiency is about minimizing energy inputs.

 

I like where this was going. I do want to challenge on a couple of points, though:

1. heavier hydrocarbons IMO do not necessarily correlate to reduced detonation. Oil mist is a known source if preignition.

2. Higher octane rating is as much the geometry of the molecule as it is the size. Multiple branchings with the same number of carbon atoms is more stable and will tolerate more heat / compression without self-igniting. a best case scenario 3,3 methylethylpentane will tolerate far more compression and higher temperatures without igniting vs n-octane, though both have the chemical formula C8H18. Even iso-octane - or 2-methylseptane - is far more stable vs n-octane.

3. Due to increased resistance to combustion, the higher branched, more stable versions of octane will naturally burn slower, which does require / allow for more ignition timing, which increases pumping losses exactly as you described. However, if you do not add the timing, you leave power on the table due to the slower burn. I might be saying the same thing you said, not sure.

 

I just thought of a better way to say this:

Peak efficiency is about making the most out of your available fuel - even if it means you leave oxygen unused.

Peak power is about making the most out of your available oxygen - even if it means you leave fuel unburnt.

I honestly think this is the driving force

 

V8r, that's an excellent summation of how I am understanding the concept of efficiency versus power.

 

In regard to oil: There are a few things at work there, aside from just the Mass of the oil molecule. It should also be contemplated that it has a different stoich burn ratio as well as poor vaporization qualities. It has a binding property with various petrolenes. It also, when bonded to light & heavy hydrocarbons, has a different latent heat value as well as a much different specific Gravity. When binding occurs it doesn't just attach itself to light or heavy hydrocarbons. It will bind heavy & lights together. It will Absorb more heat energy before vaporizing, when this happens it shares heat energy with the light hydrocarbons & when turbulence occurs they can break apart due to forces exerted one the fuel. I am not sure what the correct terminology here is but it has to do with the mass & specific gravity. Once this happens those lights are on the verge of uncontrolled ignition. As soon as chamber pressure spikes to a given amount they will undoubtedly combust.

You are correct the make-up of the molecule is at play here. Where in the chains those atomic components are, & their proximity to other compounds, determine the propensity for ignition, or pre-ignition. It also determines how easily it is vaporized.

 

Now take that train of thought & reduce pumping losses. You will actually end up with a scenario that efficiency will either remain stable or increase as RPM increases.

The whole thought process that higher RPM's should require more, or even the same, inigition advance is analogous to rational thought. Chamber pressure is rising faster, the engine is getting closer to matching the energy exchange rates of the combusting gasses, the engine is flowing more A/F so dynamic compression is increasing.

This All boils down to something very wrong is going on.

 

So, if at higher rpms, the rate of compression is so much higher that the air fuel mixture will already be more prone to ignition, we should, by any logic, require LESS timing advance, at those higher rpms.

What do NASCAR engines, or any other 9k rpm motor for that matter, do with the spark timing near redline? I'm asking because I literally don't know.

What I'm saying is, we KNOW that internal combustion engines are capable of producing power in excess of 9000rpms. So we KNOW that whatever processes of ignition/combustion that take place, happen in a short enough amount of time that we are capable of extracting power from it at those rpms. We also KNOW that the forces of friction working against us, can be overcome to these rpms. And we KNOW that manifold/port injection fueling can provide whatever air fuel mixture is necessary to make power at those rpms.

So, what, exactly, changes between the 9k racecar engines and our 7k daily driver cars? If the main difference is bearings and tolerances and rotating mass reduction... Then we are literally at the practical limits of our rotating assemblies.

IF it has to do with how the 9k racecar engines deliver the air and fuel, or the timing advance of the spark, then it should be relatively easy to track down.

What do NASCAR motors or f1 cars do use for air fuel ratios at those rpms?

What do 9k racecar engines do with the timing advance at those rpms?

What steps are taken to ensure air fuel vaporization/homogenization in the 9k racecar engines?

Are their combustion chambers designed differently?

I keep coming back to the realization that we may have just reached the limits of this platform, and more importantly, the fuel we have available to us.

My thoughts lead me to believe that you could take a fully prepared NASCAR engine, with all the friction fighting tricks, and high rpm heads, and if you wanted to put it in your camaro, and run pump gas, you have to lower the compression ratio, less compression means less cam, and now you just have another ls2... Regardless of all the racecar bearings and tolerances, despite the high rpm heads... Your 100% NASCAR engine, with nothing more than a drop in compression to match the grade of fuel available at the pump, and a smaller cam to match the drop in compression now peaks and falls just like a non-NASCAR normal person bearings and normal person tolerances.

I think ~93 octane fuel, and ~10.5:1 compression, yields ~7500rpms in a overhead valve internal combustion engine. This being a limitation of design and physics, not the components used.

I, personally, believe that it has to be related to the fuel we are using more than friction losses. Even racecars are limited by the octane rating. I guarantee you that we human critters can invent bearings that would last a whole 500 mile race spinning an engine at 20,000rpms. We have the technology. I guarantee you that we can figure out stronger and lighter materials for the rotating assembly. And we have done exactly this over the last couple decades... But we actually have not raised the redline much, especially in the racecar engines. Still capped at 10k... Why?

We either don't have access to a fuel with a high enough octane rating (resistance to spontaneous compression ignition, whatever) to allow for any more static compression to allow for any more rpms. Or we don't have the ability to extract power from the combustion event in the amount of time it occurs at higher rpms. Or we don't have the ability to deliver the fuel in a way that homogenizes properly for combustion in the amount of time the process occurs at those rpms.

Whatever it is, it's not the friction or materials, because we all know that there's amazing **** out there doing far more outrageous things than anything experienced in a simple car motor. To think we are at the limits of our ability to combat friction is a gross underestimation of mankind.

I think it has much more to do with the quality of dinosaurs we are burning, and our ability to get the fuel and air to mix properly in the amount of time we have available at the higher rpms.

 

So, if I'm understanding you here, the temperature and mass of the oil vapor is providing an ignition point to prematurely start combustion, and not necessarily reaching its flash point and starting its own burn. In other words an oil vapor "drop" is behaving similarly to a hot spot in the combustion chamber? if so, I would agree that makes sense.

In point of fact, I'm seeing pumping losses decrease as RPM increases on the model PER REVOLUTION, but the number of revolutions also comes into play. Also, the increased accelerative forces come into play and reduce net output.

Or else, I completely misunderstood what you were trying to say here

 

To continue my little diatribe...

I think we run out of time for the fuel and air to mix properly by the time it is ignited. That's the easiest way I can explain my thought process on this.

Maybe a short runner intake with both port and plenum injection? Start mixing the fuel and air in the plenum, with port injectors to guarantee every cylinder gets at least enough fuel to not be lean, relying on the pre-mixed fuel and air in the plenum to provide the rest of the fuel. I don't know.

 

I think this is an issue here, more than people realize. Our materials and lubricants have advanced tremendously, but our fuels have actually degraded, due to misguided environmental intentions. We're not burning dynosaurs, but dynosaurs mixed with corn.

but the corn has less chemical potential energy vs the dynosaurs, so we end up burning more pounds of fuel to get the same amount of work. And we end up forcing lower compression ratios to try to not burn nitrates, so we reduce the AKI of the fuel to prevent high compression ratios.

but, at the race track, we can burn 85 pounds of fuel to go 1/4 mile, and there is no environmental penalty. I guarantee you if us performance enthusiasts could easily get the gasoline with 100 AKI at the pump, then fuel economy would actually increase, and total tons of emissions would actually decrease.

Instead, we choke down the motor to make it less efficient, and we handicap the fuel, resulting in needing more fuel to do the same work.

OK. Done preaching now. But to your point, David, consumer grade fuel quality has not kept pace with the materials, lubrication, and understanding of how to generate power.

 

Sounds a lot like throttle body injection to me

 

But yes, you are correct. The oil droplets create hotspots, essentially. Moreso, they grab on to the easily combustible hydrocarbons, charge them up with a bunch of heat energy, and release them at the exact wrong time. The oil doesn't burn, it just absorbs and holds a bunch of heat, then gives this heat energy to all the fuel it touches.

 

I don't want to quote all of your guys'a posts so allow me to surmise:

Fist of all NASCAR & other types of injected engines are like apples & oranges. NASCAR still use traditional carbs with tweaks to the emulsion system. While certainly better, it is far from ideal. LS engines were based off valve train geometry of those engines, so in essence you are correct, Dave, if you take those motors, drop compression & valve timing you end up with an LS, essentially.

You still come back to the vaporization, meaning a very small uniform droplet size & delivered in a manner that is indicative to a homogenous mixture. In other words, at the point of introduction of fuel the droplets they are exposed to as much air as possible. Once this is the case the heat energy can take that mixture & vaporize the fuel the rest of the way.

Once you are there, & the chamber is closer to correct, meaning the plug lights off the light hydrocarbons first, in a big way, then the heavies follow, while directing the flame propagation in the right direction, a uniform propagation occurs. This translates into a more even burn & force exertion over the piston.

The big difference here can be seen in plug heat range. The ideal plug would generate a HUGE initial burn & flame kernel. So a very hot plug. Try that with a modern engine with inadequate everything I've been discussing & you will break stuff because the mixture needs X & X time to mix & continue vaporizing as well as getting the A/F mixture in place to all be burned. You simply can't control the flame kernel unless that stuff is right.

V8R: the rotating assembly will ofcourse possess more inertia at higher RPM's. That undoubtedly means more power. Pumping losses need to pace in an inverse to this inertia to be able to continue to sustain RPM increases.

Wanting to spin an engine, even when stuff isn't perfect, really only takes a certain dynamic compression. The cheat is to crank the duration & balance residual heat energy in the cylinder. The fuel just needs to "need" that residual energy otherwise things will go bad very quickly.

As an aside we can look at the method of fuel delivery itself: The ideal is to have the inertia transfer into the fuel happen by the air itself. Basically the air velocity must draw the fuel from a static atmospheric pressure state. Pressurizing fuel is counterproductive. It packs the molecules together, increasing the work required to separate them. Higher the pressure, the tighter packed together they are. I know this is opposite of what we typically see in spray patterns from injectors, but that is because of the nozzles designs. The orifices are big compared to what is really needed. So you get a dispersion effect from the added pressure.

To give you an idea an orifice size in the .0010-.0018" is closer to where proper vaporization can occur.

 

Fuel has states like all matter. liquid or gaseous in this case. If it evaporates, it becomes a gas. If it slides down the side of the port wall, associating closely with it's molecular partners, it is a liquid. Vaporization by definition means it has become gas.

Gas state molecules seek the ends of their containers, because of entropy. Electric fields can exert forces on charged particles (like with a mass spec) so I am considering gasoline in a combustion engine in this specific example. Although the instantaneous relationship if we were to follow a molecule as it escapes it's liquid brethren and becomes a gaseous molecule starts off near it's original group, any universe where entropy exists will have it moving away and seeking to spread itself apart from other gaseous state molecules, lest it become associated once more and condense back to a liquid state.



Im trying to show moles of air -> moles of fuel while referencing actual instantaneous injector flow volume and final BSFC which should show a relationship we are interested in.


Well, none of this is used in practice anyways. As I said, a tap on the keyboard and your engine is tuned and on to the next one. The only thing I really want to gain for myself is to answer this question (or similar): "How many more combustion events (in revolutions) can an engine turn around 8,000rpm with a 1200cc injector after fuel cut, and what is the air fuel ratio for each event"


Imagine an injector spraying 3uL of fuel into the airway of a runner. It begins to spread, being gaseous, working its way towards the plenum and valve simultaneously. And suddenly, the valve opens and draws all of that air, including all diffusing fuel molecules, and even more air from the plenum, cleansing the runner of fuel molecules and then it closes. The mass of fuel and air is reacted, pressure moves the piston, the remaining exhaust gas expansion slows as the next valve opens, the hot exhaust gas needs no encouragement to start moving out, but the piston moves up against it anyways to help, and as the piston nears the top, slowing down, changing direction as the intake valve is opening and the exhaust valve is closing, and the two systems interact for a brief instant- exhaust and intake tracts- fresh incoming air helps purge the remaining exhaust gasses as the exhaust valve shuts finally, and the piston is moving down fast now, intake valve fully open. THe piston reaches the bottom and slows, air carrying momentum is allowed a moment longer to get inside before the intake valve shuts as the piston is moving back up.


So I thought of something interesting. If you plotted a graph of bearing size increasing (X axis) vs friction (y-axis) what do you think it would look like? I know the true relationship is probably NOT a linear one; as with many such ideas in physics, the classical definition is fine for practical application in our macroscopic world. That said, I would be willing to bet such a classical definition graph of that (Bearing clearance increase vs friction loss) would demonstrate a linear relationship (the graph would look like a simple straight line) since intuition tells me that while there may be some off-handed exponential relationship at the far ends of such a graph, that somewhere in the "middle" ground for our purposes (.0018"-.0028") it would be linear in that area.

Good, and here is why: it does not vaporize well, and counts as a "liquid" which does not compress, raising compression ratio.

So this is interesting. Fuel injector flow rates are in weight/time, can we say mass/time? Within reason since masses vary but within a very narrow specific range for 93 octane gasoline pump fuel.,The n-heptane (zero octane, 0 octane pump gas nobody can use) is a flat chain molecule, capable of packing very well. branching leaves spaces between molecules. Every biologist should know fats that pack well (non kinky fats in cell membranes) are more likely to be found as a solid at the same temperature. the packing is also in association with nearby liquid phase hydrocarbons- excluding O2 when it leaves the injector in little packs, and I belive oxygen diffuses through gasoline depending on temperature.

 

The problem with throttle body injection (and carburetors also) mentioned above is that fuel puddles at the bottom of the plenum for various reasons (inertia, geometry, pressure, temperature, other), so we're back to modelling puddling/evaporation rates.

 

First off, a large portion is not gaseous out of the injector, or even before it hits the valve.

Second, you touch on something very important: O2 diffusion through petrolenes. This is a very important aspect to what I have been talking about. In the scenario to small & uniform atomization, in its relation to orifice size, the process has to be such that that diffusion has a specific impact on the fuel molecule if used in a specific manner.

You know those big paper cutters that have the big blade on the handle & the paper is sheared against a flat support? Well, that is what you want to do. The air is the blade & the orifice the flat table. You are working with a specific range of molecule sizes so the shear components need to be a specific size to atomize the molecules that are bonded together.

You must shear the fuel, in the right manner conducive to achieving a uniform droplet size, not pump it though an orifice & hope for the best. If you do this you create a fog, at ambient temperature even, that make the fuel very readily vaporized.

On a typical carburetor & Injection system, yes. This is because the petrolene compounds are not broken down, or light enough, if you will. The easiest way to explain it is comparing rain to fog. Rain drops to the ground & forms puddles. Fog stays in a state of suspended animation above the ground. Now imagine a huge wind blowing rain drops vs a Fog, like vertical rain that still drops to the ground va clouds blowing over mountains. The fog will be much more easily accelerated & stay in a relative suspension. Now when the sun suddenly peaks out what happens to the fog hovering above the ground vs the puddled water on the ground?

 

Good analogy with the fog/rain comparison. We want our spark plug "sun" to instantly evaporate our "fog".

As for the plenum injection system, with port injectors. I am not talking about throttle body injection, unless you were to use it in conjunction with port injectors, as well.

Either have long runners and two injectors per runner, or short runners with one injector per runner and a couple injectors in the plenum. The idea is to have port injectors, just like the LSx engines have from the factory, operating just like they already do. And the supplemental injectors are pre-mixing fuel into the air either in the plenum, or in the runners above the "regular" port injectors.

If anything, what I'm talking about is closer to meth injection, but with fuel. Just add the fuel fog to the intake tract upstream of the normal injectors. Probably too dangerous to actually employ, given that an intake backfire turns the manifold into a firebomb.

But you could do it with a long runner intake with two injectors per runner. Just have the injectors furthest away from the intake valve spray a little bit before the main/normal/cloest to the valve injectors. You just need a ecu that can control 16 injectors.

 

In my analogy the "sun" is simply a reference to IAT & residual heat energy in the cylinder.

What you describe, in the air mixed injector, is what is used in turbine engines. It works but the aerosol is still confined to a spray pattern & mixing becomes a bit harder to achieve with the incoming air.

Not sure how long you've been at this but many years ago running dual injectors was the "thing". PMS systems (remember those?) were used to drive additional injectors on top of the factory ECU. It worked in conjuction but the vaporization still wasn't there compared to ideal. The big advantage was being able to control a smaller injector down low & the other injector would come in at a higher air mass flow.

Engines are too sensitive to runner length, in relation to RPM, limiting the speed by pulse length of the induction cycle. So just running a real long runner become counterproductive to getting it up to the speed where the energy exchange rates need to be.

It is variable for a given rod/stroke ratio, but generally speaking about 8,500RPM is where the fun begins.

I am beginning to think starting another thread would have been better as where I am going with what I am discussing is not very conducive to the topic at hand. The time is not right for me to do that. Don't get me wrong, I love discussing this stuff, it just isn't overly beneficial at the moment.

So with that I will refrain from the discussion, if V8R wishes, & take this up at a later date. I am working on an extrapolation of an already existing mechanical carb, that has patents, mine being patentable as well, using the same conceptual delivery method, but In a completely different configuration. I will be able to get into more detail when the time is right. That may be a year or more depending on what other resources become available to me. But it will happen.