How to pick an AFR?

[Slidn]

Context I'm asking about AFR in the hemi forums, but some twat told me to come in here.
I figure you guys may be more helpful.

Q: How do we pick a target AFR in general?
It's great if someone tells me "Running an AFR of 12.2-12.4 is ideal for engine X."
What this doesn't tell me is how someone reached that conclusion.
How do you find an AFR range for any engine? Start from stock and work from there?

I have also seen multiple people claim "See what the car responds best to."
What data are we logging here? Butt dyno? 0-60 E.T?
I'm looking for a measurement here.

Q: What is an acceptable range of AFR values? 0.5? 1.0?
Going from 13.0-12.0 seems like a large swing.
Going from 12.4-12.2 seems like a small swing.

Maybe I am over emphasizing the importance of AFR, but i don't understand how we figure out what AFR is best for any given engine.

[5FDP]

With a dyno is usually the best answer.

The rest can be done to what fuel you are running and honestly just a general safe zone where "we" know the engine will make good power and torque without being on the very edge of what is usually safe. It's pretty stupid if you ask me to run an engine at 12.9 AFR when it makes 5-10hp less with 12.4 AFR type of thing, you are never going to notice that 5-10hp, so why risk it at a leaner air fuel ratio.

There is no dead set answer when it comes to road tuning. Just about any LS engine I've tune has never not run good at 12.2-12.5 AFR, so that is exactly why I stay there on pump gas. I already know that a leaner air fuel ratio on E85 makes more power in most cases, so I can pick a different value for that.

Maybe for me it's easy, I just don't care enough to ask all the why questions or need a measurement for all the things. Yeah, AFR/Lambda is important but in a way it's not that important. I'm not trying to squeeze every last bit of horsepower from an engine, I just don't want it to blow up from too little fuel. So I pick an area that I know will never be a problem.

[Slidn]

With a dyno is usually the best answer.

The rest can be done to what fuel you are running and honestly just a general safe zone where "we" know the engine will make good power and torque without being on the very edge of what is usually safe. It's pretty stupid if you ask me to run an engine at 12.9 AFR when it makes 5-10hp less with 12.4 AFR type of thing, you are never going to notice that 5-10hp, so why risk it at a leaner air fuel ratio.

There is no dead set answer when it comes to road tuning. Just about any LS engine I've tune has never not run good at 12.2-12.5 AFR, so that is exactly why I stay there on pump gas. I already know that a leaner air fuel ratio on E85 makes more power in most cases, so I can pick a different value for that.

Maybe for me it's easy, I just don't care enough to ask all the why questions or need a measurement for all the things. Yeah, AFR/Lambda is important but in a way it's not that important. I'm not trying to squeeze every last bit of horsepower from an engine, I just don't want it to blow up from too little fuel. So I pick an area that I know will never be a problem.

Okay that's very helpful. It sounds like it really is a set it and forget it kind of deal.
I'm mainly surprised that AFR isn't as important as I thought considering how many different view points there are on AFR.

Thank you

[kingtal0n]

Energy comes from fuel
combustion chamber tech and engines vary in efficiency conversion of fuel energy to drivetrain energy
The Fuel also removes some energy when it evaporates and there is a limited source of oxygen molecules put restrictions on how much mass of fuel can become how much drivetrain energy

To directly answer your question is the word experience.
The full more detailed answer is a series of empirical testing and engine modifications with statistical inference to determine significant differences with respect to some goal, e.g. economy, power, noise, vibration, etc...

[gametech]

Very few manifolds (if any) have anywhere near perfect air distribution, and even if they do, the exhaust will likely not have identical flow. Add to that small flow differences in individual fuel injectors, and the situation becomes having an AVERAGE fuel air mixture that your engine can live with. To make matters more interesting, even if you tune each individual cylinder for exactly the same air/fuel ratio, the difference in power from 11/1 all the way to 13/1 will be barely noticeable. The theoretical best power, fastest burn, and all the other chemistry goes out the window when you may have more than a 2/1 difference between richest and leanest cylinders.

[SiriusC1024]

The Hemi's like a richer mixture because of chamber geometry. With LS heads the quench pads focus the mixture toward the spark plug. On the polyspherical chambers it's just an average toward the center. That's why there are two plugs per cylinder on the Hemi's. Quench has always been an issue with Hemi's, whether true hemispherical or poly.

Look at what FCA did to the quench pads when they went to Eagle heads. Tighter squish means increased turbulence and better mixture. Notice that dual plugs are retained.
original vs eagle.png

The LS head is better for combustion, but the Hemi head is better for flow (cross flow). Richer mixture is required to keep flame propagation speed up. This has to do with charge density and non-homogeneity of the mixture at the spark event. Even if timing is perfect, those effects are still there. This can be seen in the PE curve of a factory GM tune. Mid RPM's are richer than higher RPM's because the quench effect isn't as strong. Piston speed increases turbulence and air-fuel mixing as the piston reaches TDC.

So if you just want a number 11.7:1 at high RPM WOT.

LilSick is an idiot. Feel free to click the badge icon under his name and add appropriate reputation so that other people are aware. PurpleRam is the guy you want to talk to about making Hemi's go fast. He's pretty helpful, but good luck prying secrets out of him

[Alvin]

Context I'm asking about AFR in the hemi forums, but some twat told me to come in here.
I figure you guys may be more helpful.

Q: How do we pick a target AFR in general?
It's great if someone tells me "Running an AFR of 12.2-12.4 is ideal for engine X."
What this doesn't tell me is how someone reached that conclusion.
How do you find an AFR range for any engine? Start from stock and work from there?

I have also seen multiple people claim "See what the car responds best to."
What data are we logging here? Butt dyno? 0-60 E.T?
I'm looking for a measurement here.

Q: What is an acceptable range of AFR values? 0.5? 1.0?
Going from 13.0-12.0 seems like a large swing.
Going from 12.4-12.2 seems like a small swing.

Maybe I am over emphasizing the importance of AFR, but i don't understand how we figure out what AFR is best for any given engine.

I've been tuning EFI for 22 years now. So this isn't just repeated lines you can find on the internet. It's from experience.

There is typically a wide range of AFR that will produce basically the same power. For example on a NA car you really won't see much of a power difference between 12.7 and 13.2. Maybe less than 5rwhp at very best.

A car on the richer side of that range will be more tolerant to bad gas, high IATs, high ECTs, extended full throttle pulls, etc in regards to not knocking.
A car on the leaner side of that range will be easier to get to knock.

To find the best AFR I do what I call a "sweep test"

Once the car is getting finalized in the tune I'll deliberately sweep the AFR by changing the commanded AFR in PE. I'll go from 12.0 to 13.4 and back to 12.0.. or vise versa.. it doesn't matter. I just like it start from something out of bounds, go to something on the other side of out of bounds and the back. You can then overlay this dyno run with a previous one and see what type of power gains you get at what AFR. There is usually not a big swing in power even sweeping the AFR by that much. Since the pull sweep hard one way then back the other you have two chances for the best AFR to prove/show itself.

[gtstorey]

I know it's not a GEN 3 thing, but in general, what does Direct Injection do to the desired afr? There was a lot of talk about running the LNF/E69 quite a bit leaner than "normal" ratios back in the day, but haven't really been able to find guidance on what to shoot for. The factory tune seemed to run pretty lean numbers even into where it was making boost.

[SiriusC1024]

Flame kernel like I was alluding to.

"Furthermore, spray-guided direct injection (DI) combustion systems allow stratified lean combustion operation due to their ability to eliminate wall-wetting and form ignitable stratified mixtures near spark plugs."
https://www.sciencedirect.com/scienc...60544212001740

[Slidn]

Very few manifolds (if any) have anywhere near perfect air distribution, and even if they do, the exhaust will likely not have identical flow. Add to that small flow differences in individual fuel injectors, and the situation becomes having an AVERAGE fuel air mixture that your engine can live with. To make matters more interesting, even if you tune each individual cylinder for exactly the same air/fuel ratio, the difference in power from 11/1 all the way to 13/1 will be barely noticeable. The theoretical best power, fastest burn, and all the other chemistry goes out the window when you may have more than a 2/1 difference between richest and leanest cylinders.

I think this may be one of the most insightful perspectives on the topic because you are making it clear how much variance there is in the system. Here I am think everything needs to be exact as if we are in a lab. This added a much needed context. Thank you

[Slidn]

The Hemi's like a richer mixture because of chamber geometry. With LS heads the quench pads focus the mixture toward the spark plug. On the polyspherical chambers it's just an average toward the center. That's why there are two plugs per cylinder on the Hemi's. Quench has always been an issue with Hemi's, whether true hemispherical or poly.

Look at what FCA did to the quench pads when they went to Eagle heads. Tighter squish means increased turbulence and better mixture. Notice that dual plugs are retained.
original vs eagle.png

The LS head is better for combustion, but the Hemi head is better for flow (cross flow). Richer mixture is required to keep flame propagation speed up. This has to do with charge density and non-homogeneity of the mixture at the spark event. Even if timing is perfect, those effects are still there. This can be seen in the PE curve of a factory GM tune. Mid RPM's are richer than higher RPM's because the quench effect isn't as strong. Piston speed increases turbulence and air-fuel mixing as the piston reaches TDC.

So if you just want a number 11.7:1 at high RPM WOT.

LilSick is an idiot. Feel free to click the badge icon under his name and add appropriate reputation so that other people are aware. PurpleRam is the guy you want to talk to about making Hemi's go fast. He's pretty helpful, but good luck prying secrets out of him

I highly appreciate the analytical approach to tuning and the explanation of why a particular engine may be happier on the rich side of things. I didn't even consider the head chamber design. Seriously, this is the higher level discussion I was looking for and I can't thank you enough because now I have ideas as to why we tune an engine way we do.

[Slidn]

I've been tuning EFI for 22 years now. So this isn't just repeated lines you can find on the internet. It's from experience.

There is typically a wide range of AFR that will produce basically the same power. For example on a NA car you really won't see much of a power difference between 12.7 and 13.2. Maybe less than 5rwhp at very best.

A car on the richer side of that range will be more tolerant to bad gas, high IATs, high ECTs, extended full throttle pulls, etc in regards to not knocking.
A car on the leaner side of that range will be easier to get to knock.

To find the best AFR I do what I call a "sweep test"

Once the car is getting finalized in the tune I'll deliberately sweep the AFR by changing the commanded AFR in PE. I'll go from 12.0 to 13.4 and back to 12.0.. or vise versa.. it doesn't matter. I just like it start from something out of bounds, go to something on the other side of out of bounds and the back. You can then overlay this dyno run with a previous one and see what type of power gains you get at what AFR. There is usually not a big swing in power even sweeping the AFR by that much. Since the pull sweep hard one way then back the other you have two chances for the best AFR to prove/show itself.

Okay I'll implement you're sweep test as I work through this car. It seems like a good way to confidently pick an AFR on any engine.

The only question I have left is ranges for AFRs. Is there a guide line on acceptable AFR ranges or is this again something we experiment with?
What I mean by this is going from 12.2-12.8 vs 12.2-12.4 - Or is this coming back to the sweep test? I.e if the sweep test reveal we make the best power between 12.0-12.5 then that becomes our acceptable AFR range?

Again, I can not thank you guys enough for providing the context as to why we make the adjustments we do. It is invaluable to me and I greatly apricate everyone's input.

[Alvin]

Really what your after is a AFR that makes good power and is not right up next to knock. A car that runs 14:1 at wide open won't melt itself down. But it will be much easier to knock. Likely not making great power at that AFR either.

not necessarily the case for a supercharged car... I wouldn't run a supercharged car leaner than 12.2 but there are OEM supercharged engines that target mid 12.5-13:1 AFR. Many marine engines are like this. There are plenty of examples of NA engines that are low performance, low compression, usually with a big cam and free exhaust that will take a ton of timing and a 13.2 AFR.. but 13.2 is normally my upper limit. An example would be a 4.8L with a big cam, long tubes.. you can stick 28 degrees in them have a lean PE 13.2 and never see knock. Usually these you have to lean on decently hard to get good power out of them.

I consider it a bad build, as in inefficient if I can get 28 degrees of timing in a NA car.. Or say more than 21 or less than 12 degrees for a boosted car. For builds like this... I recommend changing compression or boost.

Really the take away is that your not going to make a huge difference as long as your in the ball park. Just choose something that doesn't leave the car knock sensitive.

[kingtal0n]

AFR influences ideal spark timing, injection timing, and dependent on fuel quality, energy density, air/oil/coolant temps and other factors. And ideal spark timing for economy is not the same as ideal timing for power and power spark ideal timing is not the same as ideal spark timing for safety and durability for changing conditions.

You can't just lock down a random timing number then sweep the afr from left to right and assume the best power or integral of power maximum is the ideal afr. Just like you can't lock down an injection timing and do the same thing. The timing of the injection and spark will influence the results in terms of power based on AFR and myriad other factors. When tuning a daily driver or street car or expensive engine for non-level playing field racing the goal isn't generally peak power or ideal power its more safety and longevity so you don't have to replace the engine, and this setting of safety is a very wide range of acceptable afr timing and injection spray events potentially and since there is now way to actually measure safety for all situations you would generally what we actually need to look at is the energy content and mass flow rate of the fuel to find what sort of energy related damage could occur at the peak safety settings for example the lowest possible combustion pressure at the highest air density the engine can achieve with the least amount of timing that throws the maximum percentage of fuel as heat into the exhaust system... could be negligible on a natural aspirated engine but pure meltdown on a forced induction applications while both settings achieve safety in terms of pressure and engine seals vastly apart in terms of energy flow.