109 | Basic Fuel Tuning - AEM Infinity
In this webinar we will focus on how to correctly calibrate the VE table in an AEM Infinity. We will look at how to configure the VE table break points and how to use the input from the Infinity's onboard wideband sensors to help speed up the tuning process.
It's Andre from the High-Performance Academy welcome to this webinar where we're going to be taking a very basic look at the process of fuel tuning using the AEM Infinity ECU. Now are going to be looking at this obviously here on our Nissan 350 Z but of course the lessons that we're going to be looking at will be applicable to any platform running on the AEM Infinity ECU. So this will be a high-level overview of how the AEM Infinity looks at the fueling system and what we need to do as tuners to manipulate and change the fueling being provided. One of the key aspects with the Infinity ECU that we need to understand is that it works on the VE or volumetric efficiency fuel model basis and this is quite different from any injection time based fuel model and it's really a critical aspect that the tuner needs to understand how the VE fuel model works in order to correctly tune it. Unlike an injection time based ECU there's a few critical steps that we need to go through in the right order and if we fail to go through the steps in the right order we're going to really struggle to get the right results out of our tuning exercise.
So what we need to do is start by understanding this VE based fuel model and really what we need to understand here is that when we're looking at our VE or volumetric efficiency table this really isn't a fuel table, it's a table that's defining the actual engine volumetric efficiency. And the term volumetric efficiency, Essentially when we break it down what were looking at here is how completely the engine is filling it's cylinders at any particular combination of load and RPM. So what were really doing here is telling the ECU at each combination of load and RPM what mass of air is entering the engine. Now if the ECU knows what mass of air is entering the engine and it also knows what size fuel injectors are fitted to the engine and then finally it knows what our target air-fuel ratio lambda is it becomes incredibly simple for the ECU to calculate the required injector pulse width to output to the injectors in order to achieve whatever our desired air-fuel ratio or lambda target is. So that's the main basis there very very different because in a conventional ECU our fuel table is essentially exactly that it's a fuel table.
We're directly telling the ECU how long to open the injectors for at each combination of load and RPM. And a VE based fuel model on the other hand we're not doing that, we're telling the ECU how much air what mass of air is entering the engine and then the ECU can make the required calculations in the background. Now before we get into the actual tuning a question I quite often get is what are the advantages of a volumetric efficiency-based fuel model? And really when it comes down to it we can achieve an excellent quality tune on either an injection time or VE based fuel model. A VE-based fuel model does not necessarily mean that were going to end up with a more stable or more accurate tune, however there are some subtle advantages to the VE based fuel model that I think is important to highlight now. One of the first advantages is that our target air-fuel ratio is completely separate to our VE or volumetric efficiency table and what I mean by this is if we want to change our air-fuel ratio or lambda target that the engine is running at after the engine has been tuned we can simply do that on a VE based fuel model by accessing our lambda or air-fuel ratio target table and making the appropriate changes.
Provided everything has been configured and tuned correctly you'll find now that the engine will run at our new lambda or air-fuel ratio target. Likewise, and probably one of the biggest advantages, that we see with the VE based fuel model, is if we are changing from one set of injectors to another we can enter the injector data for our new injectors and the ECU can recalibrate the pulse width being sent to the injectors and essentially this means that we don't need to go through and retune the entire fuel table as we may need to do with an injection time-based ECU. So really all we need to do is change over our injectors into the new injector data and we should find that our air-fuel ratio is still very very close to our target. Yes we may still need to do a small amount of work in the VE table, but it would generally be much quicker and much easier to get the right results and back on track than in an injection time based ECU. Now just go back to my last point as well about changing our target air-fuel ratio.
With an injection time-based ECU in most instances the numbers in our main fuel table are accounting for our desired air-fuel ratio. In other words if we want to richen or lean our mixture at a particular point we actually need to go through and adjust the numbers in our main field table so that's the background on the VE fuel model it's really important if we don't understand how the VE fuel model works you can really come unstuck quite quickly when it comes to actually tuning a VE based fuel model. Okay, let's just jump into my laptop now and what were going to do is go through the set up operations for an AEM Infinity. Now I've obviously already got this particular ECU set up and running on our 350 Z but what were going to do is go over the basics and generally the steps in order that we need to complete. Now AEM makes our job really simple because they include what's known as a setup wizard and we can access this here in the drop-down menu.
And what I'm going to do, I'm not going to go through every step of the set up all I'm going to do is look at the critical ones here from the VE fuel models basis. Alright so let's have a look for a start at our engine tab. Unfortunately in the software my laptop pointer software doesn't work so you're going to have to bear with me it's not going to be quite as obvious perhaps exactly where my pointer is. But we're currently on our engine tab and for the VE based fuel model to work it's essential that it correctly knows our engine displacement. Which you can see here is our top parameter.
Now this may sound fairly self-explanatory obviously going to enter the capacity of our engine in this case a 3.5 litre V-6, however an area you can come unstuck here is on some of the engines that are regularly fitted with stroker kits it may not be quite as apparent to you as a tuner what the actual engine capacity is. So this is an area you really need to lock down and the reason the ECU needs to know the engine capacity as this is a large bearing on the amount of air that is being moved during an engine cycle. So in this case with our 3.5 litre V-6 if we were operating at 100% volumetric efficiency this would mean that we were displacing three and a half litres of air per engine cycle so in other words if we had fitted a four litre stroker kit to our VQ35 we would need to change this otherwise the ECU would be incorrectly calculating the mass airflow per engine cycle. Likewise we need our number of cylinders to be correct this is a fairly basic parameter, again hopefully a little bit less likely to get this wrong and we also have her engine cycle types in this case we're running a four stroke engine, obviously. Now if we move down we have our airflow calculation method you can see that's selected to volumetric efficiency or VE we actually have the ability to run a mass airflow sensor into the Infinity in my own experience probably recently unlikely in the aftermarket that we would want to do that.
Our next option is our load axis selection for both our ignition tables or our spark tables and our volumetric efficiency table and in most instances here we would be using manifold absolute pressure although if we look at the drop down menu we have a few more options there in particular our throttle position would be a common selection if we were tuning an engine running individual throttle bodies or ITBs so that would be known as Alpha N again for a very simple explanation here or example here were going to be focusing on manifold absolute pressure So this has now covered our engine basics at this point our ECU knows how many cylinders there are and it knows what capacity each of the cylinders are so that's one of the key aspects to the VE fuel model. Now what were going to move down to is our injector set up and this is another really key aspect of the volumetric efficiency fuel model. Remember that the ECU needs to know what size injectors are fitted to the engine or in other words what mass of fuel will be supplied for a given pulse width so it can correctly calculate what pulse width to supply to the injectors in order to achieve our target air-fuel ratio. Before we move to that though we have first of all here our primary fuel regulator reference and you can see if we open that drop down menu we can select either atmospheric reference or manifold vacuum reference. Now this is again a really key point basically the difference here is whether we're running a returnless system which we are on our 350 Z where the fuel pressure will remain fixed this is really common on many later model cars where there is no engine bay mounted regulator and no return line running back to the fuel tank or alternatively if we're manifold vacuum reference what this means is that we have a fuel pressure regulator fitted in the engine bay which maintains a constant differential fuel pressure across our injector as our manifold pressure climbs or drops away.
So again we are atmospheric referenced here. Okay our next option here is our actual primary injector flow wizard selection. And you can see here we have our Injector Dynamics ID 1000 injectors selected. So this makes it really easy if you're using any of the common injectors that are already preconfigured here. Looking through our drop-down menu you can see there's a fairly wide range.
Now this takes a lot of the hard work out of it for us this as tuners because AEM have preconfigured all the injector characteristics to suit. Now if you want to do something custom and you have the required data then you can do so but under most circumstances you going to be able to select from this drop-down list. And it really is critical again just like our engine capacity, it is critical for our ECU to know with the injectors fitted are so that it can correctly calculate the required injector pulse width. Moving down we have our number of injectors and we also have our primary fuel type. So this is really important again on a volumetric efficiency based fuel system because the ECU needs to know in particular what the stoichiometric air fuel ratio is for the particular fuel that were running on.
This is quite critical because this will define what mass of fuel the ECU needs to supply to mix with a specific mass of air. So for example if were running on normal pump gasoline than the stoichiometric air fuel ratio is 14.7 to 1. Which means that for 14.7 grammes of air the ECU needs to supply one gramme of fuel in order to run at that stoichiometric air fuel ratio. If however we were swapping to E 85 for example stoichiometric air fuel ratio now changes from 14.7 to 1 through to 9.8 to 1. So this means that now we need to add one gramme of fuel for every 9.8 grammes of air.
So again it's really important obviously when we're looking at this, the ECUs looking at this to have the correct fuel characteristics defined. So obviously here were running just on plain gasoline so that's as far as we need to go. Okay so we've talked about our engine characteristics we've talked about the axis for our fuel or volumetric efficiency table and we've also talked about our injector and fuel characteristics. Now what we can do is have a look in some of our actual tuning tables. So in this particular worksheet here we're looking at our volumetric efficiency or VE tables.
Before we get into that though, what I'm going to do is jump across to our lambda target table. Now again as I've kind of touched on, this is one of the critical tables in a volumetric efficiency-based fuel system fuel model. And what we have to do is make sure that before we actually start tuning that we actually have some realistic values entered into this table. So we want the actual lambda targets that we really want the engine running at. So you can see here as we would normally set up an engine such as our VQ 35 we have are manifold absolute pressure on our vertical axis and engine RPM on our horizontal axis and we can see that the numbers entered into this table through our cruise area out to about 4000 RPM were targeting lambda one, so a stoichiometric air fuel ratio of 14.7 to 1.
As we move up the table to full load, wide open throttle, we can see at lower RPM we're targeting 0.87 lambda and then as we move up to higher RPM up to above 6000 RPM we drop that a little bit richer, and sit at 0.85 lambda. Likewise I've chosen to sit my higher RPM low load regions of the lambda target tables to a slightly richer than stoic value of 0.95. So this is really before we get into our tuning one of the first places we need to visit and we need to make sure that the targets we've chosen are sensible. Now this doesn't necessarily mean that we can't come back and readdress this table at a later point if we want to, but it's important to at least start with our sensible lambda targets because what were going to be doing, our process of tuning a VE based fuel table is to adjust the VE numbers in our table until our measured lambda matches our target. And when we've done that and we've got are measured and target lambda matching that means that the numbers in our VE fuel table are accurate.
For now were going to leave our lambda target table and we're going to jump back to our VE table and were going to have a look at what the numbers in our VE table actually mean. Again our VE table, we got manifold absolute pressure that we selected on our vertical axis we get our engine RPM on our horizontal axis the numbers inside our VE table represent unsurprisingly our engine volumetric efficiency and then too, right here we have a graphical representation of our engine's VE. So essentially if we see a number of 100% in our volumetric efficiency table this means that the engine is displacing 3.5 litres of air per engine cycle. Now a few tricks here that I just want to touch on as well before we get into the actual tuning because by default your VE table may not quite be displayed like this in particular you can see that the numbers in my VE table here are displayed to one decimal point and I'm not sure if it's still the case I'm pretty sure that the default view actually gives no decimal places and if we right-click on the bar up here and we go down to number format options we can see here we have this value here listed as precision and this is going to define how the VE table is displayed. So if I put in a precision value of two, this is simply going to mean that our VE numbers are displayed to two decimal points.
Probably a little bit more accurate than we necessarily need, but personally, and this really is personal preference, I like to see a tenth of a percent, it allows me to see exactly where we are and be a little more precise, a little more accurate with the changes that were making to the volumetric efficiency numbers. Likewise if we right-click on the axis again, we go down further, if we see the need to add or remove some columns we can do that here, or alternatively we can add and remove rows. And again by default we'll find there are a lot of zones added into the default VE table in my own personal preference, and again this is just personal preference, is that I don't necessarily like to have very very tight zoning in any of my tables. For the most part, unless tuning a very very aggressive engine that has very large VE changes across a narrow RPM range, we're going to find that excessive numbers of zones just simply are necessary. All that's going to do is make more work for the tuner and instead, provided that my VE table is relatively smooth and consistent as I'd expect, I'm going to provide tuning zones perhaps every 500 RPM, and perhaps every 10 or 20 KPA and I'm going to allow the interpolation inside the ECU to take care of everything else.
Okay now I'm just going to explain what we've got here on our text grid values on the left-hand side. There's only a few key aspects that we want to look at here. First of all we have lambda one. Now I do have dual lambda here just simply to avoid confusion. We're only going to be looking at lambda one for our example here.
So lambda one this is actually measured lambda value coming from a wideband lambda sensor fitted to the exhaust. Likewise below this we have our lambda target which you can see at the moment is lambda one. This is the value simply coming from our lambda target table that we previously looked at. Below this we have our lambda feedback injector one and injector six, so essentially this is separated bank to bank on our VQ 35. And what we've got here you can see injector one lambda feedback is sitting at 0.04.
This simply means that the ECU is adding 4% fuel in order to get the measured lambda to match our target. And you can see that this is moving around a little bit pretty much as we'd expect. So this give some feedback straightaway on what the state of our tune is like how close our VE table is to being accurate. And I'll just add here, if we just go back to our setup wizard, I'll just go through this particular process here, maybe I won't. Nope.
That looks like it's having a nightmare. we'll try one more time does not like me today. All right, simply what I was going to show you, was the set up there for the closed loop lambda control. Doesn't really matter, we can either choose to have our closed-loop lambda control on or off. Now for this particular example what I've done is I've set the closed-loop lambda control to be functional up to a load value of 80 KPA.
So what that's going to mean, is that when I do go to full throttle, and we'll look at a little bit of wide open throttle ramp tuning shortly, when I do go to full throttle, we're going to be ignoring the measured lambda, and were going to simply focusing on the values in the VE table. So the ECU essentially won't try and correct any errors in our lambda. All right, now we've had enough talking. Let's actually get our engine running here on our mainline dyno. And what I'm going to do for a start, we'll just get into fourth gear and I'm just going to start our logger as well and what I want to do is come up to about 2000 RPM so we'll just do that now.
Just get us up to 2000 RPM. Were pretty close to it. Okay. So right now were operating at 2000 RPM and were operating at 40 KPA. So were absolutely central in the cell that I've got highlighted at the moment.
And what we can see is that our lambda one value is sitting on our target. Which is what we'd hope, but if we actually look at our lambda one feedback we can see that in order to achieve this were actually, the ECU is adding 2 to 4% fuel. So what this means is that our VE number and our volumetric efficiency table is actually a little bit low. And this is really exactly the same as an injection time based ECU. If our mixtures lean, our lambda is lean, we simply need to increase the values in our VE table.
So there's a few ways we can go about this, and I'm going to go through each of them. So the first way were going to look at is what I like to call the trial and error method. So we know that we've get a value of 61.5 in there and we know that were a little bit lean so why don't we just enter a value of 63 and press enter. We're going to see what exactly that gives us, and I'll just bring us back up to operating in the centre of that cell. We can see that we're still sitting around about 2% lean.
So the value that I added in there wasn't quite enough. I'll just get back up to 40 KPA. Wasn't quite enough, so were still around about 2% too lean. So the other way that we can do this is we can use the plus and minus keys on our keyboard to add or subtract fuel in 1% increments. So we can see that I've just taken that up to 65.6% We can see that now we're operating right on our target and our lambda feedback at this point is sitting at zero so this means that our VE table with a value of 65.6% and that cell is now correct.
So you've seen me use two different techniques there. We used the trial and error method where I've simply guessed at a value. And then we've used the plus and minus keys on the keyboard. Now if you ever want an idea of what the keyboard shortcuts are if we right-click on this particular cell this opens up our drop down menu here and we can see that plus or minus can be used to add or subtract 1%. If we want to make larger changes we can add or subtract 10% by holding down control plus or minus and likewise we can increase the value by a value of one, not 1% this time, just add one to our VE number, by using the U and D keys or ten increase or decrease by using the control U and D keys.
When you're doing a lot of tuning on the Infinity these will all become second nature, but if you do ever forget what those shortcuts are you can always right-click and that's going to tell you. And what I'm going to do, is I'm just going to jump up here and make a change so when we move into our next cell our fueling hopefully will be wrong. So what I'm going to do now is move up to our 50 KPA site. We'll have a look at what our fueling is like at 50 KPA so we can see that at 50 KPA we're still sitting at lambda one but remember that's because our closed-loop feedback is operational you can see that this time our closed-loop feedback is removing 4% fuel so what we can do is we can make an appropriate change to our site. And we can do that on a couple of ways.
We've already looked at using the plus and minus keys and we know that if we want to remove 4% fuel we can just simply hit the minus key four times. Whoops. And that should get us pretty much on track that's actually gone the wrong way okay so that's got us to our target the other way we can make changes using the keyboard is to directly make math changes. so in this case let's just take our value to 75%. So in this case we can see that are closed-loop feedback is removing 5 or 6% from our fueling.
And if we enter zed, This is going to allow us to make a math function change to the number in our cell. So in this case what I'm going to do is I'm going to make a 5% reduction in the VE number and I'm going to do that by multiplying by 0.95. So what I've done here is I've entered zed than the asterisk symbol for multiply and then I've multiplied by 0.95 and 0.95 that's going to remove 5% from our fueling. Now we haven't quite got to where we needed to be but that's got us a lot closer. So we can use a combination of our direct entry we can use a combination of our hotkeys and we can use a combination of our multiplication sorry, our math functions in order to zero in on the correct volumetric efficiency numbers.
Are aim here is to adjust the numbers in our VE table until our measured air fuel ratio and measured lambda matches our target. Now we've come back down to idle and there's one more thing I wanted to show you here before we go and do some full throttle ramp tuning as well and we can see here if we look on the left hand side at our text grid values, we have two values here we have our VE number our VE percent. This is simply the interpolated curent VE number that the Infinity is using for its fuel calculations. Right here sitting at about 770 RPM and 29 KPA the volumetric efficiency value it's using is 44.6% give or take. Obviously it's moving around a little bit.
Now if we look at our lambda, we can see our lambda is on our target as it should be. But we can also see that our feedback is adding 2%. This means that our volumetric efficiency number at that particular site is just a little bit smaller than it needs to be. Now this value if we go all the way to the bottom we see we have a parameter called NewVE and what this is doing is calculating what the VE number should be in order to correct the error that we have with our closed-loop feedback. So in this case you can say we are hovering around about 1% higher than what our current VE value is.
So this new VE value is quite useful to help us with fast tracking our fuel tuning. Now we're going to have a look at a quick example of that using a bit of a ramp run here. What I'm going to do first of all is add the NewVE value into our time graph at our plot below. I'll just stop our data logging And if we right-click again on our top bar here and we click on set channels, this is going to show you on the right-hand side the parameters or channels that are currently being data logged. What I'm going to do is search for NewVE.
I can double-click on that and that's going to add that perimeter at the bottom of my list there. Now I'm not quite done yet because every time we had a new parameter we need to still do a little bit of work in order to display this properly. So I've right-clicked and gone into properties what I'm going to do is go down to my NewVE value and I'm going to choose an appropriate colour for that. Let's go with red. Actually let's go with orange, something that's not being used already.
And the other thing we need to do is set our vertical range. so in this case we want to go between a value of zero and 100% so now what we see is our NewVE value is being displayed in our data log and this is going to be useful because as we move around the data log what we see is our NewVE and VE values are both displayed at the same time we can see on our VE table exactly whereabouts in the VE table the ECU was accessing at that particular point in the data log. So that makes it very easy for us to make quite accurate changes to our VE table. All right, let's have a look at that in a ramp run situation. So what I'm going to do is just get our engine running in fourth gear.
We'll go across to our ramp run settings. And we're just going to do a little run out to about 5000 RPM. In this case we will just get rid of our lambda plot because it's not being used. Right, so let's get our first run underway. Just go through to full throttle, and start our run.
okay, so that's our first run complete. We'll just stop the logger and in this case I'm not really interested in saving my log file, don't really worry too much about this for what I'm doing here. But what we can see is the two parameters that are of interest here. our purple line here represents our current volumetric efficiency number our orange line represents our NewVE value which is our calculated representation. At the same time we also have, and this is really where this data is coming from, we have, if we look at the bottom, we have our white line which is our lambda target and we have our green line which is our measured air fuel ratio, or measured lambda so we can see the effect of this particular point here.
Let's just zoom in a little bit. At this particular point here we can see that our air fuel ratio lambda has gone a lot richer than our target. So this particular point our lambda one value is sitting at 0.83 and our target is 0.89 so if we look at our VE values you can see our current calculated VE value from our table is sitting at 82.6% hopefully you can read that there and the ECU is telling us that our NewVE value should in effect be 76.8% in order to get our air fuel ratio onto our target. So what we can do is simply go through our table here and make the appropriate changes based on what our NewVE value is. So what I've done is cycled through to about 2000 RPM and we can see were actually interpolating here we're at quite high altitudes, so were not actually quite reaching 100 KPA.
Were not quite reaching the wide open throttle area of our table. So were actually sitting at 94 KPA at wide open throttle. But we can see that our NewVE value at this particular point is 80.5% we can also see it represented here. So what I'm going to do is just simply change our values here to 80.5 and just to interpolate down I'm also going to change the value at 80 KPA. So that's taken us to 2000 RPM.
We're going to cycle through and we've got another zone here at 2500 RPM and we can see that our NewVE value at this particular point is 78.5 let's just make those changes. We'll try one quick round of changes here and just see how much closer that gets us. Our next particular point is at 3000 RPM I'm not going to go through the entire table here but we'll get the idea pretty quickly of how this all works. Okay and we'll go through, we've got a zone here actually at 3250 RPM. I'll just fill that in.
76 and 74. And I'll go through a little bit further. And, go through to 3500 RPM. We can see that that's sitting at 73 and 71 so all it's doing here is really simplifying the calculation that we can do based on our measured air fuel ratio and our target air fuel ratio. So remember we can calculate a correction factor to apply if we take our measured air fuel ratio and divide it by our desired or target air fuel ratio.
That will give us a correction factor to apply. So in this case if I bring up my calculator we're looking at a measured lamba of 0.83 divided by a target lambda of 0.88 so it gives us a correction factor of 0.94 essentially what we want to do is remove around about five and a half or six percent fuel. And that's essentially what the NewVE calculation is doing for us. We'll go through and will make one more correction and while I'm doing this I'll just mention we are going to move into some questions and answers shortly. So if you do have anything you'd like me to explain please ask that now in the chat, and I'll do my best to answer very shortly.
79.5 and 77.5 so for any of you who are wondering what I'm doing with this 80 KPA row, right now all I'm doing for the case of this particular example is just making an across-the-board change I'm adjusting my 100 KPA zone because that is where I'm closest to operating we're at sort of 94 and 93 KPA. So we're very heavily biassed towards that 100 KPA zone. Just to make sure we're not really changing up our VE numbers too greatly. Because were interpolating slightly down towards this 80 KPA row. I'm just simply taking the numbers I'm putting into 100 KPA and I'm reducing them by 2% Now that might not be quite right as it turns out but we're simply looking at a quick example of how we can use this particular function.
All right, so we'll just start our logger again and we'll get our engine running. Back into fourth gear. And hopefully, now it's not going to be perfect. But hopefully what we'll find is that our VE and our measured air fuel ratio should be a little bit closer this time around. Let's do our second run.
all right, well fortunately that worked exactly how I was hoping. And what we can see is that while our volumetric efficiency numbers aren't exactly correct, particularly if we look through here, sort of our area from about 3200 RPM through to about 4000 RPM where we stopped making changes. But you can see in this area that our NewVE and our VE numbers match exactly. We can also see that our lambda and measured lambda also matches our target lambda. Now we do have the spikes here, we've got an area where we're a little bit lean another area where where a little bit lean here and an area where were a little bit rich.
We're never quite going to get these sorted out in one single adjustment. The other important thing to notice here is if we look at the area in our VE table where these lean and rich areas are let's just focus for a moment on this particular zone here we can see that this is actually occurring, this lean point is actually occurring, at 3200 RPM. Now we have a zone here at 3250, so this is the area where we need to make our further adjustment. Let's go back down and we'll have a look at this particular point here. So we have our lean area here is at 2300 RPM.
At this point we're actually starting to operate between two of the breakpoints in our table. We have a VE breakpoint at 2000 and another one at 2500 RPM. And if we jump back down we see that at 2000 RPM we're actually pretty close to our target. And if we go through to 2500 RPM, we're also pretty close to our target. So in this case to correct this lean area here, what we'd need to do is to actually go back and add in a breakpoint at 2250 RPM.
So in order to do that we can right-click and we can go add column and we can add a column at 2250 RPM. This will now allow us to make more accurate changes, in the areas where we actually need to correct our fueling. So hopefully that's given you some insight into how the fuel system or the VE fuel model works on the AEM Infinity ECU. Just to recap before we move into questions and answers, remember that the process here, the VE table, what were doing is really telling the ECU how much air is entering the engine. Remember that all of our setup parameters need to be correct before we actually start tuning our VE table, and the process of tuning the VE table is simply a case of adjusting the number in the VE table until our measured lambda matches our target lambda.
And we've looked at the parameters that we can use to guide us with that. Now during that example, when I was doing the steady stack tuning, we had our closed loop lambda control turned on so I was using the feedback from the closed-loop lambda control system to guide me in the magnitude of the changes that I needed to make. Obviously the closed-loop lambda control system is there to correct any errors, so in other words we should see our measured lambda match our target, but we need to pay attention to what the closed-loop control system is doing in order to achieve that target. The less work that we can leave the closed-loop control system to do the better the results of our tune and the more accurate our air fuel ratio is going to be. And in general I like to see my closed loop control within about plus or minus 2%.
If I can get there that means I've done my job properly. So a lot of tuners will rely on the closed-loop fuel control system to make up for a shoddy tune, that's not what it's there for. We still need to do our job properly in the first place, the closed loop control system is there to help speed up our tuning and it's also there to account for any fluctuations in air fuel ratio that we may see on a day-to-day basis. Relating to atmospheric conditions, temperatures, etc. That we can't adequately account for in our tune.
Now when we went through to wide open throttle, under those circumstances our closed-loop control as I mentioned was disabled, anytime I was above 80 KPA, closed-loop control was disabled. And under those circumstances obviously we can't look at the feedback, it's doing nothing. So we can look at two parameters, we can either look at the error between our measured and our target lambda, and we can use our correction factor, so are measured over desired air fuel ratio or lambda that will give us a correction factor that we can then apply in percentage form to our VE table. Alternatively we can also use the NewVE parameter in the AEM Infinity to help guide us to what the correct volumetric efficiency number for a particular cell will be. Again as we saw with our example there it is important to understand the power of interpolation particularly when we're operating at wide open throttle.
With our VQ 35 example there you noticed that we weren't exactly operating in our 100 KPA zone due to our altitude here in Queenstown. This had us interpolating between the 100 KPA and the 80 KPA zone. So it's really important to understand that if you're not directly in the centre of the cell you're adjusting the accuracy of that result or the result that you put in is going to be affected by the numbers in the surrounding cells. So anytime we're making adjustments, we always want to be as close to central in the cell that were adjusting as we can possibly be. That's going to give us the greatest amount of accuracy.
When it actually comes to making those tuning changes we have a number of options available to us. We can directly enter the values in the trial and error method I talked about. We can also use the plus and minus keys, or the U and D keys to either make percentage changes or singular value number changes to whatever cell we're adjusting. Or by entering the zed key and then a math function we can actually add, subtract, or multiply to make percentage changes. So we've got a lot of flexibility in how we actually make tuning changes in the main fuel or VE table in the Infinity.
All right will go through to our questions and see if we've got any. We've got a couple here. CypherMikes asks: "Do you consider VE tuning to be superior to injection time-based models due to ease of hitting your lambda target?" that's a great question and that's a valid question and honestly in my opinion it's really just about a coin toss. The reason I say that is after 15 years of tuning I can't think of too many instances where I've been out on the road or the racetrack and sort of come up with the idea, hey, you know what, I'd really like to see what the car is going to do if instead of targeting 0.90 lambda at 6000 RPM and wide open throttle I try 0.94. It's generally a change I'm going to be making on the dyno so at that point whether I make my change through the VE table, sorry, through the target lambda table and a VE based fuel model, or I make my change through the main fuel table, and an injection time-based fuel model I'm not really too concerned.
For the actual tuning, it's probably just about a coin toss, one nice thing is with the VE based fuel model, you are getting a true indication of the shape of the engine's volumetric efficiency curve so were not influencing that curve by adding numbers at higher load to artificially target a richer air fuel ratio so we should be seeing a true indication of the engines VE, shape of the VE curve. The other real advantages come down to when we want to change injectors, provided the fuel model is correctly accounting for the fuel injector characteristics, then yes that does become much simpler. Also, my personal preference, when I'm tuning a flex fuel based system, there are some advantages to using VE based fuel models, the reason being that if were properly accounting for the changing fuel characteristics, as we move from gasoline to ethanol, then this leaves us in the perfect world with almost no work to do to the VE table. As opposed to an injection time-based table where we have two very distinct and very different fuel tables between 100% ethanol and 100% pump fuel and we're interpolating between them. But realistically I think it comes down to personal preference as much as anything.
More importantly, with the current crop of ECUs, VE based fuel tuning has become a bit of a key concept or its currently a hot topic out in the aftermarket ECU tuner world, so are seeing more and more manufacturers employ VE based fuel tuning. So it's really important for tuners to understand how that works so you don't get tripped up if you come across a VE based fuel table, sorry, VE based ECU, and you're not aware of how that VE based fuel tuning works. BathurstBully’s asked: If you wanted to add a few percent to a car that's running hot and O2 controllers active and constantly chasing your lambda target, what's the best approach? Do you deactivate O2 control and run purely off base map enrichment water temperature table? Really good question there. Generally with a VE based fuel system, what you're going to have is a range of compensation tables for your lambda target. So in this case to achieve what you're talking about what you would do is use a compensation table for your lambda target based on your engine coolant temperature, and you would target a richer lambda as the engine coolant temperature richens.
Now this means that the VE based fuel system can still calculate whether you're using closed-loop control or not. Also obviously if you are employing closed-loop control then the closed-loop control system can do its job correctly. DCL4213 has asked: "How his force induction accounted for in the VE table, is it above 100%?" yeah, so basically there's no difference here. With forced induction systems we typically will see numbers exceeding 100%. But probably what I didn't get into really is the physics behind the volumetric efficiency based fuel system.
What we're really doing there is looking at the ideal gas law that the ECU is calculating. The mass of air entering the engine, using the ideal gas law, and what it's doing is looking at the volume of air flowing into the engine and it's also looking at the pressure. That's one of the inputs to the ideal gas law in other words if all other things are equal and we double the air pressure then we're going to be doubling the air density So I've got twice as much mass of air in the cylinder if we double the pressure hence we need to supply twice as much fuel to achieve our static air fuel ratio target. So yeah, forced induction engines, it's common to see the volumetric efficiency exceed 100% and numbers in the 110 and 120% vicinity are not uncommon even with naturally aspirated engines. Though a well-developed naturally aspirated engine can still see VE numbers climb and exceed 100%.
All right, before I finish off, just carrying on from that topic, it's probably worth also discussing, this is a question that comes up quite frequently. If you tune a particular engine on several different platforms that incorporate VE based fuel models, it's not unrealistic to expect, that at a given combination of load and RPM, we should expect to see the VE number be consistent across different platforms and often that isn't the case. A few tuners I've had ask me why is this the case. It's important to understand that while all of the ECUs are basing their calculations on the ideal gas law, there is no magic there. What you've got to understand is that were looking at universal standalone ECUs that are expected to work on a really wide range of engine types.
And of course there are some compromises when we try and employ a standard ECU and make it work on all these different engine types. This is where we start to see some discrepancies creep into the absolute values in the VE table. And the important thing I think to note here is, we're not really so interested in the absolute value. It doesn't necessarily mean because our AEM Infinity said that our VE value was 98.3% that that is exactly what the VE is at that particular point. It's more a case of understanding that what were trying to do or what were trying to achieve with tuning the VE table and that again is, just to reiterate, is to adjust the VE numbers until our measured lambda matches our target lambda.
Another aspect that can influence the accuracy of the VE numbers as well is how thoroughly the particular ECU decides to deal with injector characterization. On the very simple basis, we could look at injector characterizations simply as a static flow rate, and a particular test pressure that the injector was tested at. More complex ECUs also bring into the equation short pulse width adders, and ultimately while we've assumed that our fuel pressure has remain fixed in our 350 Z, it's quite likely that we will see discrepancies in our fuel pressure. All of this goes through to affect the actual flow our injector will be supplying. So in sophisticated ECUs, all of these aspects can be accounted for.
This gives us potentially a more accurate VE based fuel model. The important thing to understand here is that the more accurate we want our ECU to be, the more work is going to be required from the tuner the more complex the ECU becomes. So it becomes a balancing act, if you like, as to how accurate do we want our ECU to be versus how easy do we want it to be for the end-user to tune? Some ECU manufacturers veer on the side of being much more user-friendly to set up, others veer on the side of much more technically accurate. Doesn't mean one's more right than the other. Okay that's brought us to the end of the webinar.
Thanks for joining us, and I hope you have enjoyed that webinar. Hope you've learned something. As usual if you do have more questions, please ask them in the forum, and I'll be happy to answer them there. Otherwise I'll see you all next week. Thanks.