Z7What
Powered By Eaton!
PART 1
Quite a bit of emphasis is put on AFR or “Air to Fuel Ratio” on the Internet. With the emergence of ethanol blends such as E85 and the increasingly common E10 around town, any discussion AFR is not as straight forward as it would seem. AFR can only be calculated if one knows the stoich of the fuel. Most individuals do not fully understand tuning with alternative fuels and how wideband readings are affected. Some popular wideband tests could be improved from a more thorough understanding of the concepts we’re about to discuss in this two-part series.
One thing we should clear up first is the difference between your vehicles stock oxygen sensors (narrowband) and an aftermarket wideband sensor and controller. Your stock narrow band sensor is essentially a switch - it can tell you if you are higher or lower than a set point (a lambda of 1.00 - read further to understand what this is). A wideband sensor - and more importantly the controller driving it - is capable of interpreting a wide range of lambda, not just which side of a single fixed point you are on.
A discussion of widebands and air to fuel ratio must begin with an understanding of lambda (λ). So on to lambda… what is it?
Lambda is quite simply a multiplier used to determine how far off from the fuel’s stoich value you are. Lambda is the language spoken by all narrow and wideband oxygen sensors. When talking about premium unleaded gasoline, a value of around 14.64 parts of air to one part fuel is accepted as the (stoichiometric) mixture at which all fuel molecules are burned/oxidized with all oxygen molecules to form CO2 and H2O. It is generally assumed that a value of 14.64 is stoich for generic unleaded fuel which is not always the case.
Here are a few common fuel types and their approximate stoich value. Each of these stoich values are also represented as a lambda of 1.00:
* Unleaded 93 octane: 14.64
* E5 93 octane: 14.35
* E10 93 octane: 14.08
* E15 93 octane: 13.79
* VP C16: 14.77
* e85: 9.76
* e100: 9.00
Now say you want a richer mixture for a supercharged car under boost – typically I will command about 0.80 lambda. To get this into an air to fuel ratio you will take the stoichimetric value of your fuel (14.64 in this example) and multiply it by the lambda – 0.80 * 14.64 = 11.712:1. In order to use your O2 sensor and meter you must know the stoich point (14.64 used in the previous equation) if converting the O2 output to A/F and one can only use a wideband sensor for output not a narrow band sensor.
Let’s take it a bit further and check out the stock fuel table for a Shelby GT500. Understand this table is only used under certain throttle conditions but this is a good example to get you thinking…
Looking at this table, you’ll see RPM across the top and TP (throttle position) down the left. At 6000RPM and 800TP (floored) the PCM commands 0.819 lambda. What is 0.819 lambda? It is 14.64 & 0.819 = 11.99:1. The Ford PCM uses 14.64 as the stoichiometric value.
Now a caveat… just because you change this table does not mean the car will magically hit the lambda you command, in fact, there are quite a few other items that will add/subtract fueling commanded in this table. One of the many things I do while tuning your car is to make the desired fueling under all conditions match what is actually seen on the wideband, keeping in mind the various lambdas that need to be commanded during wide open throttle, cruising, idling, cold start, hot start, the list goes on.
Here’s the kicker, with the ethanol blends floating around (E5, E10, E20) the stoich of the fueling will change, changing what you see on the wideband. Scary? Yes possibly but that is why I wanted to show you a bit of what goes into properly calibrating a vehicle and what is taken into consideration, this is only the tip of the iceberg.
In my next article, the rubber meets the road. I will explain how you can datalog your air to fuel ratio/lambda, things to keep in mind while using widebands and lastly the effect of the fuel change we’ve seen here in Baton Rouge going from unleaded gasoline to an ethanol blend. For the curious I’ll also touch on my experiences with converting to e85 in non flex fuel vehicles, especially as they relate to the higher octane provided.
PART 2
In the last article I left you with some questions - with the changing stoich values for fuel, how do you know what your wideband should read? I also alluded to how fuel with different stoich values can lean out your vehicle. In this article we are going to nail down how to know that you can trust your wideband (or can you) and how to protect your vehicle when mixing in race fuels at the dragstrip. Just because it is a race fuel doesn’t mean it will give you additional safety!
So let’s setup a real world example. Say your vehicle is tuned for unleaded gasoline. It has a stoich point of 14.64:1. When you are cruising (termed “Closed Loop”) your vehicle is taking readings from the O2 sensors. The PCM will run a calculation to shoot fuel based upon its built in programming. The PCM then polls the O2 sensors after a short period (some longtubes need tuning for this delay because of how the O2 is moved) and measures the results of this calculation to see how its calculation worked out. These errors are collected and learned over time and a correction is applied to the calculation to keep things in check for fuel variations, changing conditions, aging sensors etc. If you are far enough out on your tune or a sensor the car will throw a check engine light along with a rich or lean code. Typically a vehicle can learn a 15% correction either way before throwing a light. When I tune a vehicle I shoot for no more than 3-5% correction in cruising conditions.
So say that your vehicle is perfectly tuned and the trims are within a few percent of each other. Now you get a tank of E10 which requires more fueling to make that same lambda of 1.00. Essentially when you fill up with E10 you have now leaned your car out. The closed loop correction will take care of you and enrichen the mixture back to a lambda of 1. The trims will be adding roughly 4% more fuel with the E10.
So with this learning things should always be kosher. Not quite… when you nail the skinny pedal you leave the happy world of closed loop. The vehicle now requires a mixture richer than stoich and the stock narrowbands cannot accurately read outside of a small range around a lambda of 1. Because of this you get no correction to the fueling, what is programmed in is what gets squirted out of the injectors. You can see why this might be bad if your car is run with a different fuel than it was programmed for. Some new cars can learn this and apply a correction. Most cannot.
This brings us to a discussion of wideband air to fuel meters. They are capable of reading outside a lambda of 1. Not all widebands are created equal. Not even close. Most consumer widebands use the Bosch LSU-4 sensor. It is a very cost effective sensor but over time as the sensor ages the readings can drift. When I dyno tune I use a sensor/controller combo that is proven accurate and does not exhibit this drifting phenomenon. While it works fine in a daily driver type scenario to get a sense of if things are going horribly wrong, on the dyno I have to trust what I am seeing - the Bosch sensor in my experience does not give me that warm fuzzy feeling.
Ok so we have a wideband controller and display, it says we are running 12.8:1 at wide open throttle. Is this correct? Is 12.8:1 really 12.8:1? What is stoich? To understand this you must first realize that the wideband reads in lambda and then multiplies the data by a hardcoded (or programmable on some) stoich value. This is important to know when you are using a wideband. Some widebands use a stoich of 14.7, some use 14.54, some 14.64. This will skew your numbers so keep it in mind. The error is not a huge deal but it is something you need to realize when tuning or watching your gauges. On your higher end widebands you are able to program the stoich point or forgo reading in AFR alltogether and instead just see the lambda values.
The short answer is to know the stoich of your fuel, educate the tune appropriately and then interpret the readings of your wideband with these things in mind.
So now let’s setup another scenario… you go to the race track and decide you want to mix in some race fuel for added safety. Going down the track you notice your tune that was dialed in at 12.8:1 is now running 13.8:1 on your wideband! What happened??? You’ll notice that the unleaded race fuels typically have a much different stoich value than the leaded fuels. Leaded fuels are normally quite close to the stoich of unleaded gas. Unleaded race fuels aren’t uncommon to have a stoich in the 13s, effectively leaning out your vehicle. If you have a particular fuel you are curious on the stoich of let me know and I will look it up for you.
Let’s take a final example. You have your vehicle tuned for unleaded gasoline with a stoich of 14.64:1. There are 3 gallons of this in the tank. You add in 5 gallons of Sunoco GT plus unleaded (112 octane) with a stoich of 13.8:1. Doing the math you now have a stoich value of roughly 14.11:1. Your car will be running just over half a point lean. It is not the octane that makes the difference but rather the changed stoich value. Keep this in mind.
There are still even more considerations when tuning in the fueling on your vehicle, these items just brush the surface of the larger picture but they are things I want every one of my customers to understand.
This article and the first post were written by Wes @ TPS which is a local performance shop.
A special thanks goes out to Michael Rauscher at L&M Engines for proofing and offering suggestions in the development of this article.
Wayne
Quite a bit of emphasis is put on AFR or “Air to Fuel Ratio” on the Internet. With the emergence of ethanol blends such as E85 and the increasingly common E10 around town, any discussion AFR is not as straight forward as it would seem. AFR can only be calculated if one knows the stoich of the fuel. Most individuals do not fully understand tuning with alternative fuels and how wideband readings are affected. Some popular wideband tests could be improved from a more thorough understanding of the concepts we’re about to discuss in this two-part series.
One thing we should clear up first is the difference between your vehicles stock oxygen sensors (narrowband) and an aftermarket wideband sensor and controller. Your stock narrow band sensor is essentially a switch - it can tell you if you are higher or lower than a set point (a lambda of 1.00 - read further to understand what this is). A wideband sensor - and more importantly the controller driving it - is capable of interpreting a wide range of lambda, not just which side of a single fixed point you are on.
A discussion of widebands and air to fuel ratio must begin with an understanding of lambda (λ). So on to lambda… what is it?
Lambda is quite simply a multiplier used to determine how far off from the fuel’s stoich value you are. Lambda is the language spoken by all narrow and wideband oxygen sensors. When talking about premium unleaded gasoline, a value of around 14.64 parts of air to one part fuel is accepted as the (stoichiometric) mixture at which all fuel molecules are burned/oxidized with all oxygen molecules to form CO2 and H2O. It is generally assumed that a value of 14.64 is stoich for generic unleaded fuel which is not always the case.
Here are a few common fuel types and their approximate stoich value. Each of these stoich values are also represented as a lambda of 1.00:
* Unleaded 93 octane: 14.64
* E5 93 octane: 14.35
* E10 93 octane: 14.08
* E15 93 octane: 13.79
* VP C16: 14.77
* e85: 9.76
* e100: 9.00
Now say you want a richer mixture for a supercharged car under boost – typically I will command about 0.80 lambda. To get this into an air to fuel ratio you will take the stoichimetric value of your fuel (14.64 in this example) and multiply it by the lambda – 0.80 * 14.64 = 11.712:1. In order to use your O2 sensor and meter you must know the stoich point (14.64 used in the previous equation) if converting the O2 output to A/F and one can only use a wideband sensor for output not a narrow band sensor.
Let’s take it a bit further and check out the stock fuel table for a Shelby GT500. Understand this table is only used under certain throttle conditions but this is a good example to get you thinking…
Looking at this table, you’ll see RPM across the top and TP (throttle position) down the left. At 6000RPM and 800TP (floored) the PCM commands 0.819 lambda. What is 0.819 lambda? It is 14.64 & 0.819 = 11.99:1. The Ford PCM uses 14.64 as the stoichiometric value.
Now a caveat… just because you change this table does not mean the car will magically hit the lambda you command, in fact, there are quite a few other items that will add/subtract fueling commanded in this table. One of the many things I do while tuning your car is to make the desired fueling under all conditions match what is actually seen on the wideband, keeping in mind the various lambdas that need to be commanded during wide open throttle, cruising, idling, cold start, hot start, the list goes on.
Here’s the kicker, with the ethanol blends floating around (E5, E10, E20) the stoich of the fueling will change, changing what you see on the wideband. Scary? Yes possibly but that is why I wanted to show you a bit of what goes into properly calibrating a vehicle and what is taken into consideration, this is only the tip of the iceberg.
In my next article, the rubber meets the road. I will explain how you can datalog your air to fuel ratio/lambda, things to keep in mind while using widebands and lastly the effect of the fuel change we’ve seen here in Baton Rouge going from unleaded gasoline to an ethanol blend. For the curious I’ll also touch on my experiences with converting to e85 in non flex fuel vehicles, especially as they relate to the higher octane provided.
PART 2
In the last article I left you with some questions - with the changing stoich values for fuel, how do you know what your wideband should read? I also alluded to how fuel with different stoich values can lean out your vehicle. In this article we are going to nail down how to know that you can trust your wideband (or can you) and how to protect your vehicle when mixing in race fuels at the dragstrip. Just because it is a race fuel doesn’t mean it will give you additional safety!
So let’s setup a real world example. Say your vehicle is tuned for unleaded gasoline. It has a stoich point of 14.64:1. When you are cruising (termed “Closed Loop”) your vehicle is taking readings from the O2 sensors. The PCM will run a calculation to shoot fuel based upon its built in programming. The PCM then polls the O2 sensors after a short period (some longtubes need tuning for this delay because of how the O2 is moved) and measures the results of this calculation to see how its calculation worked out. These errors are collected and learned over time and a correction is applied to the calculation to keep things in check for fuel variations, changing conditions, aging sensors etc. If you are far enough out on your tune or a sensor the car will throw a check engine light along with a rich or lean code. Typically a vehicle can learn a 15% correction either way before throwing a light. When I tune a vehicle I shoot for no more than 3-5% correction in cruising conditions.
So say that your vehicle is perfectly tuned and the trims are within a few percent of each other. Now you get a tank of E10 which requires more fueling to make that same lambda of 1.00. Essentially when you fill up with E10 you have now leaned your car out. The closed loop correction will take care of you and enrichen the mixture back to a lambda of 1. The trims will be adding roughly 4% more fuel with the E10.
So with this learning things should always be kosher. Not quite… when you nail the skinny pedal you leave the happy world of closed loop. The vehicle now requires a mixture richer than stoich and the stock narrowbands cannot accurately read outside of a small range around a lambda of 1. Because of this you get no correction to the fueling, what is programmed in is what gets squirted out of the injectors. You can see why this might be bad if your car is run with a different fuel than it was programmed for. Some new cars can learn this and apply a correction. Most cannot.
This brings us to a discussion of wideband air to fuel meters. They are capable of reading outside a lambda of 1. Not all widebands are created equal. Not even close. Most consumer widebands use the Bosch LSU-4 sensor. It is a very cost effective sensor but over time as the sensor ages the readings can drift. When I dyno tune I use a sensor/controller combo that is proven accurate and does not exhibit this drifting phenomenon. While it works fine in a daily driver type scenario to get a sense of if things are going horribly wrong, on the dyno I have to trust what I am seeing - the Bosch sensor in my experience does not give me that warm fuzzy feeling.
Ok so we have a wideband controller and display, it says we are running 12.8:1 at wide open throttle. Is this correct? Is 12.8:1 really 12.8:1? What is stoich? To understand this you must first realize that the wideband reads in lambda and then multiplies the data by a hardcoded (or programmable on some) stoich value. This is important to know when you are using a wideband. Some widebands use a stoich of 14.7, some use 14.54, some 14.64. This will skew your numbers so keep it in mind. The error is not a huge deal but it is something you need to realize when tuning or watching your gauges. On your higher end widebands you are able to program the stoich point or forgo reading in AFR alltogether and instead just see the lambda values.
The short answer is to know the stoich of your fuel, educate the tune appropriately and then interpret the readings of your wideband with these things in mind.
So now let’s setup another scenario… you go to the race track and decide you want to mix in some race fuel for added safety. Going down the track you notice your tune that was dialed in at 12.8:1 is now running 13.8:1 on your wideband! What happened??? You’ll notice that the unleaded race fuels typically have a much different stoich value than the leaded fuels. Leaded fuels are normally quite close to the stoich of unleaded gas. Unleaded race fuels aren’t uncommon to have a stoich in the 13s, effectively leaning out your vehicle. If you have a particular fuel you are curious on the stoich of let me know and I will look it up for you.
Let’s take a final example. You have your vehicle tuned for unleaded gasoline with a stoich of 14.64:1. There are 3 gallons of this in the tank. You add in 5 gallons of Sunoco GT plus unleaded (112 octane) with a stoich of 13.8:1. Doing the math you now have a stoich value of roughly 14.11:1. Your car will be running just over half a point lean. It is not the octane that makes the difference but rather the changed stoich value. Keep this in mind.
There are still even more considerations when tuning in the fueling on your vehicle, these items just brush the surface of the larger picture but they are things I want every one of my customers to understand.
This article and the first post were written by Wes @ TPS which is a local performance shop.
A special thanks goes out to Michael Rauscher at L&M Engines for proofing and offering suggestions in the development of this article.
Wayne
Last edited:
