147 | Important Information about Injector Data
In the last decade we’ve seen large improvements in the technology applied to fuel injectors and aftermarket ECUs. We’re also seeing injector suppliers offer more complex data to define the operation of the injectors. If we want to achieve the best performance from this new crop of injectors we need to understand what this data means and how to apply it to the ECU. In this webinar we’ll explain what injector characterisation data means and how to utilise it.
- It's Andre from High Performance Academy, welcome along to this webinar where we're going to be learning a little bit more about the operation of our fuel injectors. We're going to learn about the characterisation data that we're often seeing supplied now with some of the aftermarket injectors that we're commonly fitting to our engines. And we're going to find out what that data actually means, as well as how we can implement that data into our ECUs. So this is something that I know a lot of tuners are struggling with. There's a lack of understanding of the basic principals of operation of fuel injectors and really when it comes down to it these are the last aspect of our engine management system.
This is what the ECU is controlling, and this is what's supplying the fuel into our engines so we really need absolutely accurate control over the injectors for the ECU to be able to do the best job possible of controlling the air fuel ratio. Now obviously there's two parts to this, first of all we do need a quality injector that's actually capable of supporting the fuel flow that we require, both at the low end, such as at idle, as well as where we're requiring maximum fuel flow to achieve the desired air fuel ratio under wide open throttle, high load operation. So those are two aspects to do with the actual, the injector that we're going to choose and going to run in our engine. But when we've got that injector fitted we also need to basically tell the ECU what the injector is, how it works, and if we can do that properly, the ECU is going to be able to more accurately control the amount of fuel that's being delivered. In modern times we've seen a really large improvement in the area of fuel delivery.
So what we've seen is the technology that's gone into the construction of the injectors has come a very long way. We're actually getting better injectors that can do a more accurate job of metering fuel more consistently. We've got injectors that can operate more quickly and we've also got a better understanding of how the injector operation affects the fuel delivery. At the same time we've also seen the ECU manufacturers do a better job of accounting for this data, and again this all adds up to allow us to get a more accurate state of tune, more stable tune into our engine. Now it does require an understanding of how the injector works and what the data means so that's what today's webinar is all about.
As usual if you do have any questions that crop up, anything that you'd like me to discuss, please ask those in the comments or in the chat, and I will deal with those at the end of the webinar. So let's begin with discussing what we actually want from the injector, what do we want the injector to be able to do? Obviously we need to be able to supply the amount of fuel that the engine is going to require in order to achieve our target air fuel ratio under maximum power conditions. And I think this is probably one of the areas that most people could easily understand. Clearly if our injector is too small, we're going to end up with our air fuel ratio beginning to lean out under wide open throttle at higher RPM. Obviously not a situation we want.
The other aspect though that is often overlooked, I think a lot of people just don't really consider this, is we also at the other end need an injector that can accurately meter the much lower volume of fuel that we're going to need in order to achieve a stoichiometric air fuel ratio at idle. So under idle conditions obviously we don't need very much fuel volume to be delivered into the engine. So we really want the injector to be able to capably span that full range, both the low fuel volume at idle, as well as what we need at the other end, maximum throttle, maximum load. Now we also want the fuel delivery to be consistent as conditions change as well, that's really important if we're going to maintain very stable air fuel ratio across changing conditions. And ideally what we want is the flow from the injector to be proportional to the pulse width that the ECU is delivering to it.
What I mean by this is the ECU is going to expect a liner relationship between the pulse width that it delivers to the injector and the amount of fuel that the injector is going to deliver into the engine. In other words if we double the pulse width, the ECU is going to be expecting to double the flow. So this is one of the aspects that the ECU is using to calculate the required pulse width that the injector is going to need. Now it's also important to understand the relationship between pulse width and flow, 'cause again this is an area that a lot of people get confused about. There's a lot of talk in the aftermarket injector suppliers out there, lot of marketing information put out about the ability for their injectors to support incredibly low pulse widths while providing accurate fuel metering, and while yes that is beneficial to a point, it's not the actual pulse width that matters.
What we really want to consider is the amount of fuel that is being supplied into the engine, that's what we really care about. And the pulse width that we will need to achieve a certain fuel volume is going to depend on the size of the injector. So what I mean by this is if we've got a massive injector, a really really big injector, maybe a 2200, 2400 cc per minute injector or a 5000 cc per minute injector, which also happens to have amazing linearity and control down to very very short pulses, maybe half a millisecond or something in that range. What we'll find is that if we've got that injector fitted to a small capacity engine, even though the injector has incredible low pulse width control, it's still supplying way too much fuel at those small pulse widths, half a millisecond, to actually allow us to achieve a stoichiometric idle. Essentially the air fuel ratio is going to be way too rich because we're still supplying more fuel than the engine needs.
The flipside of that is if we've got a really really tiny injector, really small flowing injector, that has absolutely horrible linearity at low pulse widths, that may still not be an issue because on our engine, we need so much fuel at idle that the injector pulse width at idle may be in the region of three to five milliseconds. So we're going to talk a little bit more about the linearity that I've just discussed in detail shortly, so if you don't understand what that is right now, don't worry, we will get into it. So that's essentially what we want from an injector anyway just in a nutshell. We'll move on now and we want to talk a little bit about the basics of injector operation, 'cause we need to understand those in order to be able to understand the data that comes with these injectors. Now the first point is really really simple but it's easy to overlook.
The injector is a mechanical device. It's being delivered a signal from the ECU in order to open and close, and when it opens it flows fuel into the engine, when it closes obviously it shuts off. However because it is a mechanical device, we have a valve or a pintle inside the injector that physically needs to be lifted when the injector is energised before flow can begin. So there is always going to be some lag between the signal that the ECU sends out to the injector and the injector physically being able to move off it's seat, the valve off it's seat, open and start flowing fuel. Now this is generally referred to as injector dead time but there's a range of different terms for it, also maybe referred to as injector latency, offset, or battery compensation, I think that's probably a list of the most common names that I've heard it referred to as.
Now to give you an idea of what that looks like, I've just found a really poor quality image, I'm sorry, on the internet, but it's going to do a pretty good job of explaining things. So if we look up the top here, we've got the drive current that is being sent out from the ECU to the injector. And then on the bottom we've got the pintle position. So essentially what the injector has actually done. So first of all if we look right here, we can see that this is the point where the injector drive has signalled the injector to open.
This is where the injector's going to begin to become energised and the pintle is going to begin to lift to actually open and begin flowing fuel. But we can see that the pintle doesn't actually open for some amount of time. So this is very dependent on a range of things which we're going to talk about shortly. But you can see there is a physical delay between that signal reaching the injector and fuel actually beginning to flow through the injector. Now this is what most people refer to as injector dead time.
It's actually a little bit more complex than that. This is the opening time delay. But there's also a similar delay when the injector is signalled to close. So let's have a quick look at that, if we go to the other side of this graph, we can see this is the point here where the injector drive has signalled to close, it's gone from high to low or on to off. And we can see that again there is a small delay, much smaller this time on the closing side, between that signal dropping away to close, to off, and the injector pintle actually closing and shutting off fuel flow.
OK so that's the operation, that's what we've got going on and it's important to understand that. So when we're talking about the dead time, the injector latency or whatever terminology you want to use, what that is is the combination of the opening time delay, as well as the closing time delay. So what I like to consider this as, or a simple way of explaining it in a nutshell, is that the injector dead time is the difference between the time that the ECU has requested the injector to be open for and the time that the injector is physically open and fuel is flowing. So that's I find just a really nice easy way of dumbing it down in order to explain it. Alright we'll just head back across to my notes now.
Now the injector latency or dead time is affected by a number of parameters, so primarily it's going to be affected first of all by the battery voltage. So the reason for this is when we have low battery voltage there's less electrical energy there to physically open the injector. It's harder for the injector to be opened, so hence it takes longer for the injector to open. So what we see is that as battery voltage drops away, the injector dead time will increase. Likewise if we increase the battery voltage, we're going to find that the dead time becomes lower because there is a lot more electrical energy there to quickly open that injector, get that pintle off the seat and get fuel flowing.
Now the other aspect that's really important to our injector dead time is the differential fuel pressure. Now we're going to discuss this in a little bit more detail soon so if you don't understand what the term differential pressure is, don't worry about it too much because we will go into detail. So taking this into account, let's just head across to my laptop screen for the moment, and here I'm on Injector Dynamics' website. This isn't a paid advert for Injector Dynamics, it just happens that we do use a lot of Injector Dynamics product. Simply I find that Injector Dynamics' Paul Yaw and Tony Palo are well ahead of the curve in understanding the injectors, they were probably the first aftermarket injector manufacturer to actually provide the aftermarket tuning industry with solid characterisation data and a better understanding of how our injectors work.
And what we've seen now is a lot of the other aftermarket players have been forced to play catchup. So certainly this sort of data is now available from a really wide range of sources, But here I'm on the tab for the Injector Dynamics ID1000 injector which is their slope and offset versus pressure. So this is the sort of information we need here. So in this case the ID terminology they're using is offset. Essentially the same as our dead time data.
So what we can see is on the left hand side of the screen here we have the fuel pressure. Now this is differential fuel pressure, again we'll talk about that in a little bit more detail shortly. Then we've got our offset, this is in microseconds. Often we'll find that when we actually use this data inside an aftermarket ECU, we'll need to express it in milliseconds, but it doesn't matter, it's pretty easy to convert from microseconds to milliseconds. So then we've got our battery voltage break points here.
And we can see the offset or dead time data for the various combinations of battery voltage, and differential pressure. Now interestingly as well you can see here, if we're down at eight volts and we're running really high differential pressure, so up above 75 psi, you can see that there's no data in this and that's because under these conditions, remember as the pressure increases, it's harder for the injector to be open. That injector when the ECU signals it to open, it actually has to open against the fuel pressure. So the higher the pressure is, the harder it becomes for the injector to open, and at eight volts we see here that there physically isn't enough electrical energy to lift that pintle off its seat. So if we get to a situation where we're running 75, 80 psi of differential pressure, the battery voltage drops down to eight volts, the injector physically will not open.
Now that might seem like an irrelevancy seeing most of our cars run at 13.8 to about 14.2 volts once they're up and running. But a consideration there is that we are quite likely to see our battery voltage drop during cranking. So while we're actually trying to start the car. So if you're in a situation where you've got very high fuel pressure and your battery voltage happens to drop down to eight volts, then you may find that the injectors physically won't open, you can't start the car. Probably not a scenario that's very likely but I just thought I'll mention that, that's why we have these NA values in this table.
Then on the right hand side we have our slope in cc's per minute. So this is essentially the static flow of the injector, this is the flow relative to our differential pressure. So we can see that as the pressure increases, the flow that the injector provides also increases. OK so that's the sort of data we need there in terms of dead time, or that's what the data may look like. We'll just head back across to my notes now.
So that's just a description I guess of those two parameters that are affecting the injector dead time, what the completed table looks like as well as how the flow varies. Of course the dead time data will also vary based on the construction of our injectors, the size of our injectors et cetera. There are a bunch of parameters that will affect things. So what happens if we have an ECU where we aren't either accounting for injector dead time at all, or maybe we don't have the correct numbers in our injector dead time table. With all of the talk about the importance of injector characterisation, I'm seeing a lot of tuners these days get scared to the point that they feel if they don't have exactly the right data inside these tables, then the engine's simply not going to start or it's going to blow up as soon as they get near a dyno.
And that's not necessarily the case. So let's understand what this data actually means and what the affect will be if we don't have it correct. So the first one which should be reasonably obvious is that if our injector dead time is incorrect, we're going to see a situation where the air fuel ratio will vary, it won't become stable, it won't stay stable, if we see our battery voltage fluctuate. So again on a modern car this is often not that big of a concern because we should have an alternator that's providing relatively stable battery voltage. But of course it's not always the case as well.
We've got a lot of electrical load on the system, we may see the battery voltage drop down. We're also gonna see that battery voltage drop during cranking as we've already discussed. So if we do see any variations in our battery voltage, this is going to affect the amount of fuel being delivered into the engine because the ECU won't be able to correctly compensate for that, and this can see our air fuel ratio vary. Likewise the air fuel ratio will vary with changes in our differential pressure. Again that should be pretty straightforward, we've seen that three dimensional table that we just looked at where our dead time varies with both battery voltage and differential pressure.
So if any of those things vary, we're going to see our dead time vary and our ECU won't be able to account for that. The other one that's a little bit less obvious is the corrections that are being made by the ECU also won't be accurate if our injector dead time isn't being accounted for. So I'm talking here about compensations for changes in inlet air temperature for example. That's just one that we can easily look at. So to make it really clear, let's look at two examples.
We're going to take one ECU that has proper injector dead time characterisation tables inside it, and we'll look at another ECU where injector dead time is just completely ignored. So for simplicity let's say that to achieve our target air fuel ratio at whatever cell we're operating our engine in, we need to supply a two millisecond pulse width of fuel from our injector, and let's also say to keep the maths really nice and easy, that the dead time for the injector is one millisecond. OK so let's assume that if we've got our ECU that has no dead time compensation table, we're physically going to have to command three milliseconds in order to get our injector to open for two milliseconds. So remember that's the two milliseconds of fuel flow that we want plus the one millisecond of injector dead time. OK now if we're on the other hand, we've got our ECU that is compensating for injector dead time, it knows that the injector physically is only open for two milliseconds, the final pulse width is the two milliseconds of fuel that we want, plus the one millisecond of injector dead time.
So both ECUs are achieving the correct air fuel ratio. Both are giving a final pulse width to the injector of three milliseconds. OK that's all pretty self explanatory I hope. Now what we're going to do is request for whatever reason a 10% change in our fuelling. Now if we are dealing with that 10% change and we're looking at the ECU with proper dead time compensation, what we're going to end up with is increasing just the pulse width that's delivering fuel, so remember that's two milliseconds.
We're going to increase that by 10%. In other words we're going to end up with 2.2 milliseconds. We're still adding our dead time back onto that but that part hasn't changed. So the final pulse width being delivered to the injector becomes 3.2 milliseconds. If that same 10% compensation or change though was applied to an ECU that's ignoring dead time completely, it's going to apply that 10% change to the entire three millisecond pulse width, so we're going to end up with a fuel pulse width being supplied of 2.3 milliseconds.
So that's once we take the one millisecond injector dead time away from it. So we should've got 2.2 milliseconds of fuel delivery, instead we've to 2.3. So the actual compensation that was applied was 50% larger than what it should've been so that's something that's really important to understand that I think is really often overlooked. Our compensations will not work. So any of the compensations that the ECU is applying for whatever reason won't work correctly if our injector dead time is inaccurate.
The other important point to understand with injector dead time is that the affect of any error in our injector dead time is always going to be greatest at idle. So at idle this is essentially where we've got the smallest pulse width being supplied to the injectors so this means that the injector dead time becomes a really large component of the injector's opening time. So in other words if we're requesting maybe 1.5 milliseconds of fuel delivery at idle and we have an injector dead time of one millisecond, you can see that the proportion of the injector dead time is much larger than if we still had a one millisecond dead time but we're now operating at wide open throttle, higher RPM and we might have maybe 12 milliseconds of fuel being delivered to the engine. So we're going to see that affect there. Now also just so we can actually see what this looks like, if we can jump across to my laptop screen again for the moment, we're gonna be having a look at this exact diagram again shortly, again this comes from Injector Dynamics.
What we can see is this is a diagram of the injector flow. So we've got the volumetric flow rate. So this is actually the amount of fuel in micro litres per injection pulse on the vertical axis, and we've got the pulse width being delivered to the injector on the horizontal axis. So it's just a nice way of seeing the injector dead time or offset in operation. What we see is that essentially we get all the way up to somewhere around about 0.8 milliseconds and there's absolutely no flow being delivered.
So essentially up until the point where we get past about 0.8 milliseconds, nothing's happening. The injector physically isn't opening and no fuel is being supplied. Once we get past that point we get to the area where the injector is now starting to supply fuel. OK we'll just head back across to my notes now. OK so I mentioned previously the term differential pressure.
And I wanna just spend a little bit of time going over that and what that means. We saw the differential pressure was one of the axis on the Injector Dynamics ID1000 injector offset and slope data sheet that we've looked at. But what exactly is differential pressure? Most people, most tuners I think have by now a pretty good understanding that the flow and maybe even the injector dead time will be affected by the fuel pressure. And they look at this as the pressure that's in the fuel rail. But it actually goes a little bit deeper than that.
And let's just see, I've got a photo here, let's jump across to my laptop screen. This is actually a diagram that I use to explain another phenomena which is wall wetting, but it'll also work quite nicely for our demonstration here. So here we've got our fuel injector, it's fitted into the intake runner, and obviously we've got our intake valve sitting down here. What isn't drawn in here is obviously we're going to have a fuel rail running across the injector, please excuse my terrible artwork but that red blob we've now got there is our fuel rail. So we've physically got fuel in that fuel rail and we have that fuel sitting at some pressure.
Maybe it's three bar, maybe it's four bar, it doesn't really matter. But we have some amount of pressure acting on the top of the injector. Now the part that's easy to overlook though is we also have pressure in the inlet manifold, in the inlet port. This is all assuming here that our injectors are fitted post throttle body, which is going to be the case in most instances. So the pressure in the inlet manifold is acting on the other side of the injector, on the actual pintle side of the injector.
And what we really need to worry about, or what the injector really is interested in, is the difference in pressure between the fuel pressure in the rail and the pressure in the inlet manifold. So essentially what we're looking at is the difference between the fuel pressure and the inlet manifold pressure. And that is what's going to affect the operation of the injector. So if we look at this in a relatively extreme example. When we have a turbo charged engine, let's say for example we are running a fuel system with three bar of rail pressure.
So we've got three bar of fuel pressure up in the rail. We've got a turbo charged engine, maybe it's running at atmospheric pressure, and that gives us a differential pressure of three bar across the injector. So when the injector opens, the pressure in the fuel rail is higher than the pressure in the inlet manifold so fuel will flow through the injector. Now let's increase the boost pressure in our turbo charged application to three bar as well, three bar gauge so three bar above atmospheric pressure. Now what happens when our injector opens is that no fuel will flow.
And the reason for this is that the pressure in the fuel rail is identical to the pressure in the inlet manifold. So when that injector opens, the pressure on each side of the injector is equal. So nothing will flow. If we actually could go further, we could increase the boost pressure above the fuel pressure and we're actually gonna find that the boost pressure will run through the injector, so it will actually go completely the opposite way. Obviously that's not a scenario we're going to get because our engine isn't going to run with no fuel flow.
But it's that differential pressure that we need to understand. So that's a really important concept and that's why if we head back to our Injector Dynamics, let me just get back to that. If we head back to our Injector Dynamics sheet here we can see that the fuel pressure is listed as psid, which stands for psi differential. So it's a really important aspect to understand there. Alright let me just head back to my notes again.
So why this is important is obviously it's going to affect the flow as well as our injector dead time. We saw that as we increased the differential fuel pressure, our injector is able to flow more fuel as well. So it's affecting the actual flow for a given pulse width that's coming out of the injector, and this is going to affect our air fuel ratio. And this aspect, where this becomes important, is if we are running an engine with a returnless fuel system versus an engine that has a return style fuel system. So modern cars we're seeing most of them come out these days with a returnless fuel system, where there's a regulator fitted inside the tank and the fuel system runs at a fixed constant fuel pressure.
Often this is four bar or 58 psi gauge pressure. So what this means is that regardless what the manifold pressure in our engine is doing, the fuel pressure in the fuel rail will stay at a fixed pressure. And this means that our differential pressure across the injector is constantly changing. So that's really important with a returnless fuel system to understand that. On the other hand if we've got a more conventional return style fuel system, with a manifold pressure referenced fuel pressure regulator, the job of that fuel pressure regulator is to maintain a constant differential pressure across the injector.
What I'm going to do is just have a quick demonstration here on our G4 Plus ECU on our Nissan 350z. So let's just jump across to my laptop screen for a moment, we'll just go online with our G4 Plus ECU. And I'll just get our logging up and running. OK so I've got some basic parameters here which we'll look at. So on the top here we've got our engine RPM.
And then below this we've got our throttle position. And then we've got three parameters being logged here. In our lower group we've got in white our differential fuel pressure, remember that's the term that the injector really cares about. We've also got our fuel pressure being measured in the fuel rail. You can see that's currently sitting at 350 kPa.
And then we've got our manifold pressure. OK so this 350z runs a returnless system. As I've said very common on late model cars. So our fuel system is running in this case at 350 kPa, 3.5 bar. What I'm going to do now is blip the throttle, so this is going to affect our manifold pressure.
When I go through to full throttle, our manifold pressure is going to go up to atmospheric. We're going to see the affect on both our differential pressure as well as our fuel pressure itself. So let's do that now. OK let's just zoom in and have a quick look at what happened there. OK so first of all we can see my throttle blip there.
And we can see the affect on our manifold pressure which is in blue. So as soon as I go and give the throttle a decent blip we see our manifold absolute pressure jump up to 95 kPa. What we can also see at the same time though is our fuel pressure being measured in the fuel rail, for all intents and purposes, that stays pretty much consistent here at 350 kPa. So we haven't really seen much movement if any on our actual fuel pressure in the rail. But because the manifold pressure, which is on the other side of the injector has changed, it's gone from 30 kPa to 95 kPa, the affect on our differential pressure has been quite dramatic.
And that's the white line that we can see here. So our differential pressure is initially sitting at 420 kPa while we're idling. What we can see is that differential pressure drops down, and we're essentially tracking right the same as our actual fuel rail pressure. So that's a really important concept to understand with any engine that runs a returnless fuel system. Our differential pressure across the injector is going to be a constantly moving target, depending on what our current manifold absolute pressure is.
So that's going to affect, as we've already talked about, that's going to affect the flow through the fuel injector and it's also going to affect our injector latency or our injector dead time. So we really need to be able to cope and control, cope with those parameters and control the injector in order to achieve stable air fuel ratio. Right we'll head back again to my notes. OK so we've talked now about the injector dead time. We've also talked about differential fuel pressure.
I also want to spend a little bit of time talking about the short pulse width performance of the injector. So this is again something that's become quite a hot topic over the last few years. Again with companies such as Injector Dynamics releasing much more detailed characterisation data and understanding better how the injectors are working. So let's understand what that term actually means. So let's head across to my laptop screen.
So again this is taken from Injector Dynamics, it's from their ID1600, it's an injector they actually don't supply anymore. We've got the axis which I've already discussed, we've got our volumetric flow rate through on our vertical axis and we've got the pulse width on our horizontal axis. So remember initially I mentioned that the ECU, if it doesn't know any better, is just simply assuming that the fuel flow through the injector is linear, and in other words it's proportional to the pulse width being delivered. In other words if we supply double the pulse width, we should get double the fuel flow. And if we look at what's going on here, let's just draw a line through this graph.
And we're gonna come all the way back down through zero as well as I can. That's pretty good. So we can see that above around about 1.5, maybe two milliseconds we can see that the actual flow does a pretty good job of representing a straight line, it is in other words linear. So above this point where the flow is linear is referred to as the linear operating range of the injector. So this is where the ECU can really easily control the injector.
If it does double the pulse width being supplied to the injector, we're going to end up with double the fuel being supplied. What we can see though is if we get down to about 1.5 milliseconds we see that the flow starts to deviate away from this ideal, or theoretical linear flow. So we've got this lower non linear area of flow. There is also an upper non linear area of flow as well but generally not quite as dramatic or as important to understand. Now we can also see that at about one millisecond here, there is this sudden drop away in our fuel flow, so this point is referred to as the knee.
And this really is the lower limit of where we can happily operate an injector and expect to get good control over the fuel delivery. If we are operating to the left of that knee, this is a point where the pintle in the injector is not opening all the way. And the problem we see is that we get a really really sharp change in our fuel delivery over an incredibly small pulse width change. So it's not possible to get really good control over the fuel injector or over the volume being delivered out of the injector if we're operating to the left of that knee there. We actually really want to be ideally operating well above the knee because the problem we can see, if we've got our injector, our engine tuned, and we're operating right on this limit here, right on the knee, we can achieve a good air fuel ratio, it can be stable.
But if we get to a situation where the ECU needs to make a small reduction in fuel flow because of a compensation coming in, what we're going to end up doing is basically moving to the left and really quickly we're going to fall off the cliff there and our flow is going to drop dramatically, with even a very minute change in our injector duty cycle. So this is going to have the engine running lean and stalling. So if we're ever running near that knee, we're going to have very very poor control over our fuel injector. OK so the point of this demonstration here is this, we wanted to talk about this lower non linear area of injector flow. Now we don't necessarily need to worry about it.
And most ECUs up until probably the last five or six years, didn't do anything about it. They completely ignored it. And that's OK to a point, we can get around this. What we're going to find is that, let's just draw this line back in here again. If we have our ECU all set up and tuned correctly, and we are getting to a situation where we're operating down maybe around about 1.3 to 1.4 milliseconds, what we're going to find is that we're going to run a little bit lean because the injector is actually supplying less fuel than the linear graph would suggest.
And what we do is compensate by this, we manually just end up putting slightly larger numbers into our fuel table to compensate for that non linearity in the injector flow. So we might end up with a fuel table in our ECU that doesn't quite look as smooth and as consistent as we might expect. But generally it's really not that big of a concern, and we can definitely do a reasonably good job of tuning our ECU with no data to help correct for this. However of course as we've seen improvements in technology, particularly this has been driven a lot by the OE manufacturers, who are needing to meet more and more stringent control over the fuel delivery in order to reduce their emissions year after year, these sort of inconsistencies actually start becoming quite a serious problem. So what we see is that now the OE manufacturers a lot of them will actually characterise this non linear area of flow.
There's a couple of ways of doing this. The most common way that I've seen is by the use of an additional table inside the ECU that's referred to as a short pulse width adder table. It's a bit of a mouthful but essentially what it is is a correction that's applied to basically account for the difference between the linear flow, the ideal linear flow and this non linear area that we've got down here below sort of in this case, about one point, it's actually up to about two milliseconds. So it accounts for the variation that we've got in there. Now if you've got an ECU that has the short pulse width adder table and we've got it correctly filled in, this is going to give us the ability to get more accurate control over the injectors and we're not going to have to bake in these errors into our actual fuel table.
And as I've said this is common place now with OE manufacturers so if you're reflashing an aftermarket ECU from the likes of GM or Ford then these ECUs have this sort of information, require this sort of information in order to properly control the injectors. So this actually becomes a pretty important part if you are upgrading to a larger set of injectors in one of these late model factory ECUs. Alright I'll head back across to my notes now. One other example I just wanted to give here, and this is probably something that a lot of the older tuners out there who have dealt with drag cars may have come across. Back when I was running our own drag car, we didn't have the selection of large injectors that we're blessed with these days, it was really difficult to find good quality large injectors.
And again there wasn't as much known about the injectors. So one of the go to injectors was the Indy Blue, or the Bosch 1600 as it's also known. And I just wanted to show you this is another graph here if we can jump across to my laptop screen. So this sort of shows you how far the technology has come. This is again a graph from Injector Dynamics that compares the ID2000 to the Bosch Indy Blue.
So the Bosch Indy Blue is this line here in blue on the screen. So there's a couple of things you can notice here. First of all there is quite a large bulge in flow, or non linearity in flow once we get up above around about eight milliseconds there. So we've got a sort of linear area that runs from about eight milliseconds down to about two milliseconds. The point I wanted to note though is you can see that there is some really ugly stuff going on down here in the low flow area.
Particularly we can see that the injector doesn't essentially open until one millisecond. And we see this big bounce here and it drops down into a trough. And the interesting thing for anyone who has tuned on one of these old Indy Blues, is that if you're running the sort of fuel or size engine where you need to be operating the injector down quite low in the injector pulse width, you find that you inevitably end up down around about 1.5 milliseconds. And the reason for that is as you can see, you're sitting right in this trough, and the unique situation that if you don't have access to this sort of data and understand what the injector's actually doing, it's really confusing because if you're sitting at 1.5 milliseconds and your air fuel ratio is still richer than your target, the obvious reaction would be to reduce the number in your fuel table. So that will reduce the pulse width being delivered to the injector.
But you can actually see that the effect of that, let's just zoom in a little bit. The effect of that is to actually increase the amount of fuel being delivered by the injector. So we get to this weird situation where we're at 1.5 milliseconds, we drop the pulse width to perhaps 1.3 milliseconds, and the engine actually goes richer. Now it's just an example really that I wanted to show of how far the injectors have come, how far the technology has come. If we compare that to the ID2000, we can see that that injector is almost linear right down to 1 millisecond.
So you compare those two injectors, you can understand how much easier it's going to be to perform your tuning with one of these later generation of injectors. Right we'll head back again to my notes now. OK so we've talked about the parameters that are important there with the injectors. We've talked about the dead time, we've talked about the reference flow value, how that is affected by battery voltage and how the reference flow and the dead time are affected by the differential pressure. We've talked about differential pressure and now we've covered off the low pulse width performance of the engine, that lower non linear area of flow.
So that's the injectors, we've learned a lot more about these thanks to the companies that have been doing this testing. And at the same time we've also seen the ECU manufacturers get on board and come a long way as well. So I said right at the start that the ECU generally expects a linear relationship between the pulse width and the flow, and we've seen that that's not necessarily the case. So we've seen the aftermarket ECUs now get on board with being able to characterise this lower non linear area of flow with a short pulse width adder table to help give us a more consistent, more accurate tune, and a more realistic shape to our fuel tables. Now dead time in itself, that's been a fairly common parameter for a number of years.
It hasn't always been the case though. I've been in the industry now for about 15 or 16 years. When I first got started there were a number of ECUs that were pretty common and did by all accounts a reasonable job, that completely ignored injector dead time. There was absolutely no compensation for injector dead time at all. And in fact there are still brands of ECU out there that still carry on with no injector dead time compensation at all.
So just because injector dead time is inaccurate or isn't considered at all, definitely doesn't mean that you can't tune the engine. Hopefully we've now discussed the effect that that's going to have on the accuracy and the consistency of your tune though. Now we're also seeing more complex characterisation come in in terms of the 3D dead time compensation tables as opposed to just two dimensional. I wanted to just briefly mention that because that might sound a little bit confusing. We've looked here at our 350z, I'm going to do another demonstration in a second, and we now know that with a returnless fuel system we're going to end up with a constantly changing differential fuel pressure as our manifold pressure varies.
This is not a consideration if we've got a return style fuel system with a manifold pressure referenced fuel pressure regulator. If everything's working correctly there, the fuel pressure regulator should maintain a consistent differential pressure, so we can get by with a two dimensional injector dead time table. Let's just head back across to our injector dead time for the ID1000, we'll see what exactly I mean by this. So first of all let's go with the simple system. We've got the manifold pressure referenced fuel pressure regulator.
Let's say we've set our fuel pressure to 43.5 psi, three bar and we're going to be operating solely in this three bar row of our injector offset table. So we can see that we've got values here at eight, 10, 12, 14 and 16 volts. We're going to be using those values for our 43.5 psi differential fuel pressure, and we'll enter those into our two dimensional injector dead time table. And again as our manifold pressure varies, if we've got a turbo charged engine, we bring that up onto positive boost pressure, the fuel pressure in the rail is going to be raised to maintain that same 43.5 psi differential across the injector hence the dead time is going to remain the same. The differential pressure isn't changing, you only need to worry about battery voltage.
However if we've got a returnless system like our 350z, things get a little bit more interesting. Now we can basically be operating anywhere in this table depending on what the manifold pressure is doing, what our fixed fuel pressure in the rail is doing, and of course our battery voltage. So this is why we need to actually develop a proper three dimensional table for our injector dead time compensation. So again I'm just going to start up our 350z. And we'll have a look at that in action.
So let's head across to our ECU's injector dead time table. What we can see here on this page, obviously some of this is going to be specific to the Link ECU but it may look a little different in different brands of ECU, but essentially the way it all works is still gonna be the same. So we've got a three dimensional table and we'll go to our injector dead time table, and we can see we've got the same axis as we've got on our Injector Dynamics dead time data. We've got battery voltage and then on our vertical axis we've got differential fuel pressure. Just to confuse matters slightly, the numbers on this are in metric units whereas Injector Dynamics use imperial.
We can actually just change that across by pressing U on the keyboard. So we've got break points on this table that are essentially exactly the same as our injector data here for the ID1000. And what we've done here is just taken these values from the Injector Dynamics data and we have planted them into this table. So what's the point of this? We can see at the moment we're sitting nice and stable in the 14 volt and 60 psi zone. You can see our dead time at that point is 1.08 milliseconds.
Watch what happens though as I give the throttle a blip. So when I blip the throttle, just like our last demonstration, that's going to affect the manifold pressure. Remember our fuel pressure is going to stay fixed, but this means that our differential pressure is going to vary. So what we're going to see is the little cursor that's in yellow is going to move around so let's just do that now. So when I blip the throttle we end up jumping up to the 50 psi 14 volt zone.
Because as I blip the throttle our manifold pressure reduces hence our differential pressure across the injector is also reduced. When the engine comes back down in revs though we actually pull a higher vacuum that what we're seeing at idle and that's why we actually saw that just before we come back down to idle, we actually move slightly down and start interpolating towards that 65.3 psi zone. So again I'll just do that demonstration and you watch that cursor move. So while it is possible to get the engine running without a three dimensional dead time table, this is going to do a much better job of controlling the injector, gonna give us a much more accurate tune if we have the potential to set up a three dimensional injector dead time table like this. If you don't have the ability to do that, what you can do or what you'd want to do, is focus on the differential pressure that you're actually seeing at idle, and base your injector dead times on this because remember I said that the injector dead time has the greatest affect at idle, and less so as our actual injector pulse width starts to increase.
Alright we'll head back to my notes now. OK so what do we actually need from our injector supplier, our injector manufacturer, what sort of data do we need? So the main points we're going to need, we're going to want our injector dead time or offset data versus battery voltage and differential pressure. That's going to allow us to set up our dead time table regardless whether it's two dimensional or three dimensional. We also need the slope and in terms of Injector Dynamic's data they refer to it as slope, reference flow's another value, essentially the open flow, the maximum flow that that injector can supply. So this is another important aspect that the ECU needs to understand.
It'll also be able to work out from that flow, the actual flow that we'll see as the differential pressure is adjusted to. Then we also need, if available, the short pulse width adder table, and we need the minimum pulse width that the injector can reliably operate at. Now that's a lot of data and the common question we get is what do we do, where do we get this data from? So the first answer I'll give you is when you are buying injectors this is something that's often overlooked and people really regret it later. When you are buying aftermarket injectors, it's always best to purchase these injectors from an aftermarket supplier that can give you proper characterisation data for those injectors. You're gonna have all of the information you need, it's gonna make your job really really easy.
There's a number of injector manufacturers now that are supplying this data. Obviously today we've looked predominantly at Injector Dynamics but those aren't the only guys out there who are able to provide this data. So it's probably one of the first questions you need to be asking when you're looking for injectors, can you provide me with the injector characterisation data? If you've already got a set of injectors and you don't have this data for the injectors, this is where it gets a little bit trickier. So you can develop your own data. This is hard, but it's not impossible, and we're seeing a lot of aftermarket ECUs these days now come with some more advanced injector test set ups so you can actually pulse the injector an accurate number of times for a very specific pulse width, and you can measure the flow of that injector into a graduated cylinder and essentially develop your own flow charts like we've already looked at for the Injector Dynamics data.
Now obviously this requires a fair bit of work on the part of the tuner and not too many tuners are going to be going down this path, but if you do have a fair bit of time on your hands and you're prepared to go through the process, it is achievable. What we'll do is just have a quick look at how that works. So if we jump back into my laptop screen here, what we'll do, this is on the PC Link, Link G4 Plus software obviously. We can see under the fuel setup menu, there is a parameter there or an option there called injector test. We'll click on that there.
So what we can do is go through and it performs what's referred to as an advanced test. If we double click on this we can see that we have the option to provide an advanced test that is only available on injector output number one. And when we go into that advanced test we have the ability to define the number of injection events. We can also define the test injection frequency, and then finally we can also define the pulse width that will be delivered to the injector. So again if we're careful, we can do a really good job of measuring the flow from the injector and develop our own injector table, so flow and dead time values.
Obviously if you wanna develop your own injector dead time tables, you're also going to need a way of varying both the fuel pressure as well as battery voltage. So there's a fair bit to this. I know that the MegaSquirt brand of ECU also provides these advanced test functionalities too. So it's not restricted just to the Link G4 Plus software. OK so as I've said that's probably a little bit beyond what most people are going to want to do.
You can possibly try and find an injector testing or cleaning company that will do these tests for you and develop the data for you. There were a few companies around a few years back that were doing this for customers, we're finding these days that that is less and less common. If you are dealing with one of the aftermarket brands of ECU though, the companies now are doing a lot of this testing for their own brand of ECU. So for example I know Adaptronic are doing this on their newer modular range of ECUs. If they don't already have the characterisation data for the injector you're running, you can send a couple of injectors in to Adaptronic and they will develop injector characterisation data that's then available to all Adaptronic users.
Likewise Motec are also doing this for the M1 range of ECUs and I think we'll see more and more ECU manufacturers jump onto this because they know it is hard for us in the aftermarket to actually get this data ourselves. Particularly if you are dealing with an injector that's factory where you've got no chance of getting proper aftermarket data from the injector manufacturer. OK so the other option is what if we simply don't have any data? And this is the question that I get quite frequently. And this is where I'm a little bit probably controversial because I feel that while any time I'm tuning a car I'm always going to go out of my way to provide all of the data for injector characterisation so that I can do the best job possible, and I'm going to look for that data when I am purchasing injectors as I've already mentioned, I still know that it is possible to do a perfectly good job of tuning the engine without this data. The engine is still going to start, it's still going to run, we're still going to be able to tune it.
The downside is we're going to need to understand that there will be some inconsistencies with our air fuel ratio, some of our compensations won't work. So if you're in that situation, you've got no way of getting data, then the first thing I would say is you can ignore any of the short pulse width adder tables in the ECU and what I would do is zero those tables out. Now what that's going to do is, it's not going to mean that you're not injecting any fuel, what it's going to do is just have the ECU operating exactly like an ECU that has no short pulse width adder table in it. So no big problem there, we've got that same issue that's going to then creep in which we've talked about where we may see some inconsistencies in the shape of our fuel tables when we are operating down in that non linear area of the injector operation down there with very small pulse widths. The harder part is the injector dead time table and this is where a lot of guys really just don't know what to do.
And again my somewhat controversial opinion here is I'm quite happy to start with a value at 14 volts, somewhere in the region of maybe 0.85 milliseconds. Obviously the actual injector dead time at 14 volts is going to vary depending on our fuel pressure and the actual injector we're operating. But we're in a situation where we have no data so one way or another we are going to be taking a guess. So let's start with 0.85 milliseconds. We'll find that that's probably a reasonably run of the mill value for a lot of the injectors we're going to be operating with.
The ID1000s that we've just looked at here in our 350z at four bar of rail pressure, are just on one millisecond so we're a little bit short of that but let's just take that as a value. We know it's wrong but that's what we're going to start with. Now we can tune our engine to run at 14 volts, and that's going to give us all of the problems that we've talked about with a potential inconsistency in our air fuel ratio. What we can do is go one step further than this though and we can then start developing our dead time table at lower voltages. So we can unplug that alternator to prevent that charging, get our voltage down to perhaps 12 volts from 14 and then what we're going to do is see our air fuel ratio drift.
What we can do is adjust the 12 volt zone in our dead time table until our air fuel ratio sits back where it was at 14 volts and then we can also go a little further, maybe add some electrical load to the ECU, to the engine, to the car I should say, by turning on headlights, maybe fans, maybe the windscreen wipers et cetera to drop the voltage down further and we can continue this process. It gets a little bit harder to increase the voltage above 14 volts but what we will tend to see is in the zones we can get to, we're going to generally see a relatively obvious trend to our injector dead time table, and generally what I'm going to do is extrapolate that out into the zones that I can't get to. One important point to note if we are going to be doing that is that we do need to be very careful of the voltage being supplied to the fuel pump. If we get to a point where our fuel pump voltage has dropped down, we're going to find that our fuel pressure may also drop so this can also be another effect on our actual air fuel ratio. So if we don't recognise that our fuel pressure has dropped away because the fuel flow out of the pump is dropping, then trying to compensate that with our injector dead time table is going to lead to some serious problems.
What I'll do now, I'll just make sure that this is going to work, yep OK if we jump across to my laptop screen here, I just mentioned the fact that we generally have a reasonably consistent shape to our injector dead time table and I just wanted to show you this. So here we've got our 3D table here in our 350z. And we can see that in both directions we've got a relatively consistent shape to our table. So this is what I'm talking about. If we've developed dead time points at 14 volts at 12 volts and at 10 volts, we can extrapolate the shape of that curve both downwards as well as upwards to the higher voltage settings.
Now a lot of tuners do believe that that is developing an accurate dead time table and I want to point out that it is not. It's going to develop a dead time table that at idle will do a reasonably good job of correcting the air fuel ratio or maintaining a stable air fuel ratio as our battery voltage fluctuates, but the actual absolute numbers inside that table are almost certainly going to be inaccurate. What we've essentially done is just develop a correct step between the different cells, but the numbers in the cells are almost certainly incorrect. OK so we'll move into some questions and answers now. And I see we've got some questions already in there so I'll get started on those.
If you do have any more questions that crop up, please ask those in the comments and I will do my best to answer them really shortly. Carl has asked, regarding the huge injectors such as the offerings from Billet Atomizer and others that do not have characterisation data available, is it even an issue on such large injectors that are primarily designed for drag racing alcohol applications where the small pulse width performance is not accounted for? And what are good practices with use of these types of injectors to get decent idle and low load performance. OK yeah you're absolutely right Carl, this is a situation where the injector manufacturers often don't provide data. I've used the Billet Atomizer injectors in a methanol powered jet sprint boat application and was a little bit surprised when I rang the manufacturer and basically when I asked for the injector dead time data or characterisation data, they looked at me or sounded pretty blank on the phone I should say, really didn't give me anything that I could use. What we are seeing though is there are companies again like Motec that are characterising these injectors in house for users of their platforms.
The point you've made though is pretty accurate. Essentially if you're getting to a point where you're using a 5000cc injector, almost certainly it's going to be in a drag application running on alcohol fuel, and really the idle quality is almost irrelevant. So even though while we would like to have injector characterisation data, it's really an injector that's primarily going to be pouring in a huge amount of fuel under wide open throttle operation, and the idle quality and the low pulse width performance probably is a secondary consideration there. Nick has asked, since you guys are on the subject of fuel, what do you do to work out the right size fuel lines in regard to pump flow and rail pressure and also the size of the return line. Probably a little bit off topic on this particular webinar, however Nick if you wanna jump into our webinar archive we've got two webinars in there that discuss in detail the fuel system design.
So we talk about pump sizing, we talk about lines and regulators in there. So probably can't really do justice to that answer inside this webinar now. Andy's asked do you use fuel pressure dampers in the fuel rail on returnless fuel systems? Yeah I do. So this is something that I've probably gone through most of my career giving little thought to but I've got two applications actually in our workshop at the same time which really made me rethink this situation. So our Toyota 86 is our other development car.
That runs a fuel pulsation damper on the factory rails as do most EFI cars. And it's quite common to simply discard these when upgrading the fuel system, fitting larger fuel rails et cetera, quite often there's no position for a fuel pulsation damper. There are some really weird little idiosyncrasies on the Toyota 86 fuel system that I found fairly early on in our testing with the Motec ECU. And when we log the fuel pressure in the port injector fuel rails at very high frequencies, we actually find some oscillations that are occuring even with those dampers in place. So the system we upgraded to was a radium fuel rail system and they actually have the ability to refit the factory pulsation dampers.
So I've gone with that. If you remove the pulsation dampers, you can in some instances get into a situation where there are harmonics at certain RPM ranges in the fuel system that can have a really large impact on the fuel delivery and give a really really strange shape to your fuel system. So our 350z is another car which is really known for these issues, a lot of people get themselves into problems when replacing the factory fuel rails with larger fuel rails, discarding the pulsation dampers, and can get particularly a bank to bank fluctuation as a result. Chad at Motorsport Wiring has asked, in a flash tune or a reflash scenario if we assume the car manufacturer has the injector data correct and we perform an injector upgrade that have an increased flow of 50% without any data available, is there a rule of thumb or any hint as to what to do with the injector dead time adjustments. OK so probably actually something I should've covered here, this is covered in detail as well in our Practical Reflash Tuning Course.
So yes there is, there's a process that we go through called scaling aftermarket injectors. This is what we'll go through in an aftermarket ECU that has a relatively simple modelling of the fuel injector which is normally considered with just a reference flow or an injector scaling value, so it's essentially the injector's flow, and then a latency or injector dead time table. Now remembering that the injector dead time has a bigger impact on the injector flow at idle or very low pulse width than it does at higher load where our injector pulse width is much higher. What we want to do is test, we're basically using our short term and long term fuel trims to show us how close we are at two points in the engine's operation. I generally do this at around 3000 RPM, light load cruise.
So that's where we've got the higher injector pulse width, and then we'll do it at idle. Now I generally start at the higher point 3000 RPM in cruise and we'll focus initially on the injector scaling. So that's gonna make the biggest difference to our flow. So what we wanna do is basically adjust the scaling, and check that our injector, our short term and long term fuel trims are pretty close to zero. Generally we wanna have our trims very very close to zero, that means that the ECU knows the size of the injector correctly and is able to supply the correct amount of fuel.
Once we've got that pretty close we can then check the performance at idle. And if our injector dead time is incorrect, what we're likely to find is that the air fuel ratio at idle will be incorrect. So our trims will be doing a lot more work at idle. So if for example our injector dead times are too low, what we're going to find is that at idle our air fuel ratio will be lean. So it's an iterative process when we need to go through, adjust our scaling, then adjust our injector dead time.
Of course when we adjust our injector dead time or latency, that's also going to affect our required scaling, so it's a process of going backwards and forwards between these two points, adjusting the scaling and the latency. Your point there is good though. If you have gone with a set of injectors that are let's say double the size, what we would start by doing, just to get our engine up and running so we can begin gathering some data and logging, would be that we would make the appropriate change to our base scaling number. So if we've doubled the injector size fitted to the engine, we would start by halving that injector scaling value. Now that's not going to be perfect but it should get us up and running to a point where we can start scanning and actually logging some data.
Before we do move on I will say here that this is, a lot of the ECU manufacturers, factory ECU manufacturers are going with a much more comprehensive model of the injector's operation. The likes of GM, the likes of Ford. And in those situations we really do need a much more comprehensive set of data for the injectors. David has asked, do you see any side effects for injectors used in an E85 fuelled race car that sits around a lot, like clogging or rusting? Yeah generally E85 is not a fuel that I'd recommend leaving in the fuel system for any period of time if the car is not going to be used. The fuel is hygroscopic, it does absorb moisture out of the atmosphere.
So particularly if you live in a very humid area, it's likely to start gaining quite a large amount of water inside the fuel system so that can potentially be damaging to a lot of the fuel system components. Some of the latest crop of injectors we're finding now in the aftermarket are designed really around E85 with full stainless internals. But you also do tend to find that E85, if it is left for a long period of time, can start forming sort of a nasty gel like substance which isn't probably a great thing to be putting through your injectors. So my rule of thumb is if I'm running an E85 car and I'm planning to not use it for more than a period of around about a week, I'll actually drain that fuel and pickle the system with pump gasoline. Superman07 has asked, if you wanted to future proof your build and let's say went with ID1300s in the place of ID1050s for later modifications, would dropping the fuel pressure from OEM specs be a sound strategy for idle and low load drivability provided you have the data for the combination? Really if you're talking about those two sets of injectors, there's probably almost no problem with running the ID1300s right from the get go.
First of all if you've got a problem with, I'm guessing what you're talking about here is the ability of the injector to supply a small enough volume of fuel to get a good quality idle. If you're going to have that at a higher fuel pressure, you're going to have that at a lower fuel pressure as well. It's not going to make a dramatic difference really there and you're only talking injector that is around about 300cc's per minute difference in flow. So probably unlikely to be an issue. I'd probably go with the larger set of injectors.
I will say there as a caveat though, obviously I've got no idea exactly what your application is in terms of the engine capacity, number of injectors, and the fuel so all of those things have an effect there. I will probably also recommend that you check out our fuel system webinars in the archive there, 'cause that gives you a much wider understanding of the whole fuel system and the sizing of the injectors for your application. Sorry for the long winded answer there but I though I probably needed to do a better job of explaining myself. Andy has asked are 1000cc injectors rated at 100% injector duty cycle even though you never want to go past say 85. Yes the 1000cc rating is a static flow rating where they are operating at 100%.
Probably with the latest crop of performance injectors, the EV14 based injectors, 85% is probably quite conservative these days. I'm generally pretty happy with running the injector up to 90%. There is a urban tuning myth, I guess you'd call it, that if we run the injectors past, or at very high injector duty cycles, this can result in the injectors becoming overheated and operating erratically. And this actually isn't the case. There is actually an endurance test that is performed where the injectors are run with no fuel though them under constant wide open use for a period of time without damage.
So it's not so much a case of damaging the injector. There's two reasons we really don't want to go very high in the injector duty cycle. First of all just like there is a non linear area of flow at the short pulse width area, we also have a non linear area right at the very high injector duty cycle so we can see again inconsistencies in our injector delivery and a funny shape to our fuel curve. The other thing as well is it doesn't give us very much head room if some compensations need to for some reason add some additional fuelling. Let's say for example our intake air temperature drops and the ECUs compensations automatically start adding some additional fuel.
Jacob has asked, while Injector Dynamics cuts off their flow data at 90 psi, are there any injector providers that give us flow data and latency data reliably at 100 psi plus fuel pressure for high boost guys? Not to my knowledge and I'm pretty confident in saying that the Injector Dynamics injectors, a lot of them are safely capable of running beyond 100 psi. I don't wanna quote on that. It's all on their website though. So the issue you've got there as well as you start increasing the fuel pressure to very high values, it does also put a lot more stress on the fuel system in terms of the pumps, and quite often particularly if you're running an electric fuel pump, you actually get into a situation where you sort of go around in circles because you're increasing the fuel pressure in order to gain additional flow out of the injector but often the pump can't keep up and then the flow from the pump drops away so you end up sort of going around in circles and not really getting any improvement in your outright fuel flow. The situation most people end up with if they're running something like 100 psi differential pressure is that they're going to be running a mechanical fuel pump.
In general my own experience with the drag racing applications that I've been involved with, most guys running really high boost pressure with mechanical pumps are running 80 psi differential pressure so I've never been in a situation myself to run beyond 100 psi differential fuel pressure, and I just wonder if you're getting yourself confused there with differential pressure versus fuel pressure itself. But yeah I've never needed to run up that high. OK that looks like it's brought us to the end of our questions so hopefully everyone's learned a little bit more about injectors, about the characterisation data, what it looks like, why we've got it and what to do with it once you do have that. As usual for our members if you do have any questions after this webinar has aired, please ask those in our forum and I'll be happy to answer them there. Thanks for joining us guys.