AFR Tech Question 2

[Mike Laws]

Guys:

I posted this nearly a year ago and it started getting good but went off-track a little and died. Rather than resurrect that one, let's try again with a slight change.

A hypothetical/tech question for comparing intake systems.

Given the same engine, same atmospheric conditions, same fuel, same inlet size (area), same dyno/race car, same wide band O2's, etc. (In other words – an ideal scenario for performance comparison of intake systems)...

If this engine produces the same air fuel ratio value & trend from beginning rpm to ending rpm – does it matter what the induction system is as far as power is concerned? [Carburetor(s), carb location, EFI, MFI, inlet location, injector location, # of injectors…]

This article rekindled my question: https://www.enginelabs.com/engine-tech/ ... nd-lambda/

[Roundybout]

The induction system that atomized the fuel better would be better. Can the AFR be the same with different degrees of atomization? Better atomized fuel = easier vaporization leading to more complete combustion, no?

[BradH]

The induction system that atomized the fuel better would be better. Can the AFR be the same with different degrees of atomization? Better atomized fuel = easier vaporization leading to more complete combustion, no?

I'm paraphrasing something I read by DV, but the gist is the atomization can be "too good", resulting in high efficiency from a fuel utilization standpoint, but less power.

Put in the context of Mike's question, I interpret that to mean that different induction systems having the same AFR may have noticeably different power outputs.

[Roundybout]

The induction system that atomized the fuel better would be better. Can the AFR be the same with different degrees of atomization? Better atomized fuel = easier vaporization leading to more complete combustion, no?

I'm paraphrasing something I read by DV, but the gist is the atomization can be "too good", resulting in high efficiency from a fuel utilization standpoint, but less power.

Put in the context of Mike's question, I interpret that to mean that different induction systems having the same AFR may have noticeably different power outputs.

Wouldn't anything that increases efficiency increase power fuel wise? Or put another way, less fuel needed to make the same power. If two engines are running the same AFR I couldn't see how the one that better atomizes the fuel, increasing the surface area for vaporization could be at a handicap. I think what matters is what's happening at the point of combustion. Two engines can have the same AFR but one uses more fuel. That excess fuel can be in liquid form coating the intake runners and floor yet what's making it into the chamber is the same AFR as a highly atomized fuel system with nothing wasted or yet to be burned hanging around in the intake manifold. So yes I think two engines running the same AFR can have different HP levels and drivability. The fuel doesn't care if a carb, injector or what intake system is being used. The final product in the combustion chamber at ignition is what it cares about power wise.

[In-Tech]

Hello, before I get too deep into this conversation...has anyone seen a wide band sensor accurate at below .65 Lambda? Years ago I was doing some methanol testing and it started getting unrepeatable around .7 lambda. This was with a Bosch LSU4.2. I did not try my ntk horiba nor was the LSU4.9 out at that time. I am looking for feedback on this, David Redzsus?

To hopefully compliment the OP's question. I know there is always an argument about BSFC, I would like to spend some time with a proper response of my opinion. Keep in mind, the sensor reads combusted fuel. I'm sure most have seen the dyno SAYS 10:1(laminar element and fuel flow meter) and the wide band says much leaner and all that leads back to is why tuning to BSFC can be very misleading due to inefficient combustion.

Sorry to cut off for today, I went to the shop early this morning and took off at 2 to attend a mass fathers day celebration and I'm beat tired at 9:30pm

edit, the o2 sensor doesn't just read combustion, it gets skewed by oxygen during high overlap too. AND BSFC gets skewed(and calculated AFR) with over scavenging.

Enough for tonight but I wasn't going to sleep with an absolute error in my previous statement.

[RevTheory]

Wouldn't the one that provides the best charge cooling paired with fuel droplets that are sized right for said charge temp and pre-fire cylinder pressure show the best power?

[nitro2]

Just to add another dimension into the mix, let's say that at max power the ideal AFR happened to be 12 for some application. Well is the AFR 12 on EVERY combustion cycle, or is it for example 11 on some cycles 12 on some cycles 13 on other cycles all averaging to a nice 12.

When AFR is looked at for every combustion cycle it can be quite steady or it can be fluctuating, obviously fluctuating (but generating the right average) is not as good as steady and generating the right average.

AFR sensors are not quite as fast as we would like to see but we can still see fluctuation, and if the AFR sensors were faster then we'd see it even better.

If AFR is fluctuating cycle to cycle then what ends up happening is that most cycles are not optimized, and worse yet for applications where detonation is a possibility and limiting factor, the detonation limit is reached sooner with a fluctuating cycle to cycle AFR than with a steady cycle to cycle AFR.

[Newold1]

First the thought to use AFR as a power determiner is somewhat flawed by two major factors. One, Wide band O2 sensors are not that accurate in the first place. You can take 5 new same brand same type Wide band sensors out of the box and run them on the same setup and conditions and you will see marked differences. Two the actual placement of the O2-lambda sensor has a huge affect on the readings the sensor produces.
Wide band O2 sensors are a tool, not an answer and as such the limitations and capability of their usage has to be factored into any scientific analysis of an engines efficiency. The measurement of power and torque is the real determiner of how well an intake system is working for the production of power and torque on a particular engine not the measurement of what the AFR is. One could very easily have an engine with a particular intake system and with tuning for maximum power and torque and an average AFR could be measured and recorded. Then the same engine equipped with let's say a better performing intake system and the engine could be tuned for maximum power and torque output and the AFR could be measured again at the same AFR.
The AFR did not determine the power differences if both engines are correctly tuned and the AFR's could be equal the obvious power differences would be measured either by a dyno or in a vehicle's performance. The AFR only would be a poor predictor of performance. My feeling is that an O2 sensor is a tuning tool not to be used to select intake systems for the production of maximum power and torque. JMHO

[Mike Laws]

Good input!

Will more efficient droplet sizes, charge cooling & vaporization produce a different AFR? If so, why/how?

nitro2: I 'think' I've seen the fluctuations you describe on unfiltered dyno and racepak data (8 individual O2's) and remember thinking to myself that there is something going on there. Ignorantly, I added filtering in order to knock the high & low points down. It's interesting that you brought this up and it makes complete sense.

[BradH]

The induction system that atomized the fuel better would be better. Can the AFR be the same with different degrees of atomization? Better atomized fuel = easier vaporization leading to more complete combustion, no?

I'm paraphrasing something I read by DV, but the gist is the atomization can be "too good", resulting in high efficiency from a fuel utilization standpoint, but less power.

Put in the context of Mike's question, I interpret that to mean that different induction systems having the same AFR may have noticeably different power outputs.

Wouldn't anything that increases efficiency increase power fuel wise? Or put another way, less fuel needed to make the same power. If two engines are running the same AFR I couldn't see how the one that better atomizes the fuel, increasing the surface area for vaporization could be at a handicap. I think what matters is what's happening at the point of combustion. Two engines can have the same AFR but one uses more fuel. That excess fuel can be in liquid form coating the intake runners and floor yet what's making it into the chamber is the same AFR as a highly atomized fuel system with nothing wasted or yet to be burned hanging around in the intake manifold. So yes I think two engines running the same AFR can have different HP levels and drivability. The fuel doesn't care if a carb, injector or what intake system is being used. The final product in the combustion chamber at ignition is what it cares about power wise.

FWIW, this is what I was referring to (https://www.hotrod.com/articles/0511phr ... ters-tech/):

Atomization Requirements

Looking at the results so far, it would seem that the more gain a booster has, the better our carb will work. That assumption is indeed correct if we are looking at a wide operating band, but if it’s a race engine where optimizing output over, say, 2000 rpm, is the goal then things are a little different. Why? Because it is possible to have a booster that brings about a too finely atomized charge. The smaller the fuel droplets are, the more readily they evaporate into a vapor. When a droplet becomes vaporized to a gas, it takes up much more room in the intake and cuts the volumetric efficiency of the induction system. Although a charge of vaporized fuel and air will display about the best combustion characteristics, it won’t produce the best power because the amount of air in the cylinder with the vaporized fuel is less than with liquid fuel. Although a charge with fully vaporized fuel delivers the best in terms of drivability and fuel efficiency, to make maximum power a certain optimum droplet size is required.

[David Redszus]

The induction system that atomized the fuel better would be better. Can the AFR be the same with different degrees of atomization? Better atomized fuel = easier vaporization leading to more complete combustion, no?

Let's take this one step at a time.
The AFR refers to the mass of air divided by the mass of fuel delivered to the engine. The fuel can be in liquid form or in gaseous form (propane, etc). Since the engine cannot use fuel in liquid form, it must first be divided into very small droplets and then evaporated into a gas. The evaporation of fuel lowers the charge temperature, increasing charge density, and makes more power.

However, it the fuel does not evaporate completely, the charge cooling effect is reduced. If fuel droplets impact the inlet duct wall, the rivlets of fuel will slowly crawl along the wall surface but the mixture will be lean.

If the fuel does evaporate completely, it can cool the aluminum wall surface, reducing charge cooling.

If the fuel does evaporate completely, the fuel vapors will displace air molecules, causing the mixture to become slightly richer. If the fuel evaporates completely, cooling the charge, it may have an effect on the ignition delay period, shifting the combustion peak pressure angle to a later angle. This may increase or decrease performance, depending on the pressure location starting point.

The Lambda sensor does not read AF mixture; it reads the oxygen balance in the exhaust. Excess oxygen partial pressure is viewed as a lean condition and vice versa. The oxygen sensor responds very quickly, but most data systems do not sample quickly enough to capture the fluctuations in readings. But they are there.

Changes in lambda sensor tip temperature will alter the readings. While many sensors are heated to compensate for lower temperatures, some are incorrectly located providing incorrect readings. Most lambda broadband sensors make use of controllers that regulate tip temperatures and adjust for temperature variances. But the tip heater cannot cool down an overheated sensor. Location is important.

Air leaks and ignition misfires will show a false lean reading. The sensor tells us what it sees, not what we wish or want.

Changes in fuel composition will alter AFR settings, even without a jet size or map change. Variations in fuel do exist and some (pump gas) can be quite large.

Lambda sensors that are located in collectors will provide average readings; some cylinders may run lean while others run rich. You won't know. Even if a lambda sensor is located in a single header tube, there will be cycle to cycle variations. A gauge display will not reveal this variance or range of variance. A high speed data collector system is required to see the mixture variations at speed.

[Roundybout]

What happens with direct injection vs a carb? DI uses a high pressure to inject directly a highly atomized liquid fuel into the chamber. (I'm assuming the intake and exhaust are both closed at the point of injection). Does the fuel instantly turn into a vapor? All of it? How does that affect the density of the charge?

I see a combo of injection close to the valve and and also way up stream as a secondary/primary means of fueling. Is that a case of the best of both worlds? Maybe instead of AFR should we be measuring BSFC to compare two intake systems? I realize an O2 sensor has limitations, especially when it's seeing an average if using them in each bank of cylinders.

[Kevin Johnson]

http://ltces.dem.ist.utl.pt/lxlaser/lxl ... 5paper.pdf


https://www.azom.com/article.aspx?ArticleID=2810


https://ilot.edu.pl/KONES/2002/01/str323.pdf

Big topic.

[David Redszus]

What happens with direct injection vs a carb? DI uses a high pressure to inject directly a highly atomized liquid fuel into the chamber. (I'm assuming the intake and exhaust are both closed at the point of injection). Does the fuel instantly turn into a vapor? All of it? How does that affect the density of the charge?

With a carb or throttle body injection, the fuel vapors will displace and reduce inlet air mass. Direct injection eliminates this problem and increases air mass flow. Whether the fuel is properly atomized is dependent on injector dynamics (pressure, spray pattern, etc.).

I see a combo of injection close to the valve and and also way up stream as a secondary/primary means of fueling. Is that a case of the best of both worlds?

The system is commonly used on engines that must operate across a wide rpm range, i.e. road racing.

Maybe instead of AFR should we be measuring BSFC to compare two intake systems?

Bad idea. Computing the amount of fuel used is not easy. Fuel blown out the exhaust compounds the problem. BSFC, either high or low, does not indicate best power. It is used for economy or efficiency evaluations.

[Mike Laws]

First the thought to use AFR as a power determiner is somewhat flawed by two major factors. One, Wide band O2 sensors are not that accurate in the first place. You can take 5 new same brand same type Wide band sensors out of the box and run them on the same setup and conditions and you will see marked differences. Two the actual placement of the O2-lambda sensor has a huge affect on the readings the sensor produces.
Wide band O2 sensors are a tool, not an answer and as such the limitations and capability of their usage has to be factored into any scientific analysis of an engines efficiency. The measurement of power and torque is the real determiner of how well an intake system is working for the production of power and torque on a particular engine not the measurement of what the AFR is. One could very easily have an engine with a particular intake system and with tuning for maximum power and torque and an average AFR could be measured and recorded. Then the same engine equipped with let's say a better performing intake system and the engine could be tuned for maximum power and torque output and the AFR could be measured again at the same AFR.
The AFR did not determine the power differences if both engines are correctly tuned and the AFR's could be equal the obvious power differences would be measured either by a dyno or in a vehicle's performance. The AFR only would be a poor predictor of performance. My feeling is that an O2 sensor is a tuning tool not to be used to select intake systems for the production of maximum power and torque. JMHO

I am NOT EFI-savvy, however it's my understanding that AFR is heavily used while in closed loop. If so; how would the the inaccuracy of wbo2 sensors effect performance?