Summary

If you're using Link's 'modelled' fuel equation mode (volumetric efficiency), then you'll also need to understand how to calibrate the charge cooling coefficient, as this will affect the ability of the ECU to accurately track target lambda changes. In this short webinar we'll explain what the charge cooling coefficient is, how it works, and see how to properly calibrate it on the dyno. For this webinar we will be demonstrating the process on a Mitsubishi EVO 9 using a Link G5 Voodoo Neo 6.

00:00 Hey team, it's Andre from High Performance Academy here, welcome along to another one of our webinars.
00:06 And in this webinar we are going to be talking about the charge cooling coefficient in the Link range of ECUs.
00:12 This is something that's only relevant if you are using their modelled fuel equation which, for want of a better term, is Link's lingo for a volumetric efficiency fuel model.
00:22 And I think it's a parameter that probably it's fair to say most people don't fully understand and I would expect that a lot of tuners simply are not touching it because they don't know how to correctly set it up.
00:35 That being said, there is of course a very complete help file that tells you exactly what to do but let's be honest, people don't read the manual.
00:43 So, what is this and what does it do? Actually I will preface this with just mentioning that this is going to be a reasonably short webinar, this is a very simple topic once you understand it.
00:53 We'll have a demonstration of it on our dyno so you can see exactly how to dial this in.
00:58 But it is going to be a short webinar so if there's any questions on this, I suggest that you do try and get those into the chat pretty quickly so that I don't end up missing out on them.
01:08 Alright, with that being said, charge cooling gate, what is it? We'll start by talking about what a volumetric efficiency fuel model is and how that differs from traditional and Link's speak which is essentially injection time based.
01:21 Two different ways of skinning the same cat realistically and I mean most ECUs these days are running volumetric efficiency based fuel models but really it's not a big deal, you can tune on either, you've got to get the same job done.
01:35 There are a few subtle advantages with a VE based fuel model if everything is dialed in correctly.
01:41 Such as the ability to swap to a different set of injectors, enter the new characterisation data for those injectors and if everything has been dialed in correctly.
01:50 The engine should essentially start and run without needing any changes to our fueling.
01:54 I personally think where the biggest gain comes from with VE fuel models is when you're starting to run flex fuel because within reason the volumetric efficiency of the engine doesn't change as we move between pump fuel and E85 or E98.
02:09 Yes I know to be completely picky and fussy, yes there can be a small improvement in volumetric efficiency on ethanol thanks to the cooling effect but we're still starting to really get down to very very small differences.
02:24 So, that makes just the tuning a little bit easier.
02:27 OK what is charge cooling gain? Well what it is is a parameter essentially that defines how much the fuel being injected will cool the intake charge temperature as it goes through a phase change from liquid to vapour.
02:41 So, this is again a very subtle aspect but it is important to understand it and to get it dialed in.
02:48 So, let's jump into our laptop software and we'll have a quick introduction.
02:51 So, at the moment I've just been sitting here waiting for this webinar to start and I've actually got closed loop control turned on for our fueling.
02:57 It's really important that we actually disable that so before we go too far, let's do exactly that.
03:03 We'll come into our closed loop lambda and we will turn that just momentarily to off and we will store that change.
03:13 Right so now we have no closed loops so the ECU isn't making any changes in order to correct errors.
03:19 Let's just come to our fuel main.
03:22 And this is where all of our settings for our fuel setup are.
03:25 I've got the engine running so some of these are locked out like our injection mode and our fuel equation mode which is the main one here, you can see that that is set to modelled.
03:34 So, basically we've got traditional which is injection time based, in which case what we're going to learn about here is irrelevant, we've got modelled and then we've got modelled multi fuel.
03:44 I'm not going to cover everything here realistically, the one that we're interested in is this one here, fuel charge cooling coefficient.
03:52 And it's currently set to 10° which is the default value.
03:55 So, as I mentioned, press F1, you're going to get the help file pop up and I would say this is one of the things that Link do an exceptional job of is the function based help is always there and generally it's always very complete.
04:10 But like I said, people don't tend to read the instruction manual so I'm going to just go through it and show you.
04:16 The idea behind this is what we're going to do is select an area of the map where we can hold a little bit of load and a little bit of RPM.
04:23 We don't want to be doing this at idle because we're going to be affected by heat soak, we're going to be affected potentially by being in the short pulse width adder area or the non linear area of the injector's operation which can make things a little bit tricky so we want to be in the middle of the map somewhere where we are in a linear area of injector operation.
04:44 That being said as well, this really is only going to work if we do have good injector data and if we come to our injector setup here, this is all the data around the injectors that we will need.
04:56 This is actually a little bit interesting because we don't have proper characterization data for the stock injectors in a Mitsubishi Evo 9, very hard to get that information rock solid.
05:06 Normally though if you're buying aftermarket injectors from a reputable source, maybe Injector Dynamics for example, they'll be able to give you the data in a drop in format.
05:15 So, what we want to do here is start with our injector flow at the rated pressure and of course what that rated pressure is.
05:22 And that's one piece of data we do have for these injectors, 560 cc's at 300 kPa, 3 bar, 43 .5 psi.
05:29 And we also have our dead time table, this needs to be set up correctly.
05:33 And then if you have the data, let's just go back to our last screen.
05:39 If you have the data, also a short pulse width adder.
05:42 So, we want the injector characterization data set up properly.
05:45 That is the first place we want to start.
05:49 Once we've got that done, what we want to do is come into our air fuel ratio lambda target table, make sure that this is set to lambda one in the area that we're going to be operating.
05:59 So, I'm going to probably be doing this somewhere around about here.
06:01 So, we're already set to lambda one, what we can do is just make sure these ones are all set to lambda one and we're good to go.
06:09 So, I'll explain the process first and then we'll go through and actually see it live.
06:13 So, what we're going to do is bring the car up to a steady state operation in this particular cell here, 60 kPa, 3000 RPM.
06:22 Then we're going to go back to our VE map here and we're going to make sure that this particular cell that we're operating in is properly dialled in so we are actually achieving lambda one.
06:33 Once we've got it consistently sitting at lambda one and we're going to be nice and smooth on the throttle so we're not moving around and bringing in transient enrichment et cetera, we're going to go back to our AFR target table and we're going to ask for a target lambda change from lambda one to lambda 0.9.
06:50 Now, if everything's working perfectly, then the ECU should be able to track that new target perfectly, in other words we should just go straight to our new target.
06:58 But chances are that's probably not going to be the case initially.
07:02 And then we're going to need to adjust our charge cooling coefficient until we get there.
07:06 Really easy to just keep in mind, if we go too rich, so we go richer than 0.9 lambda, maybe we end up at 0.87 or something like that, in order to correct that what we want to do is reduce the number in the charge cooling coefficient.
07:20 So, too rich, we want to reduce.
07:23 Conversely of course if we end up lean, leaner than our target, we're going to increase that charge cooling coefficient.
07:28 So, let's have a look at how that works now.
07:31 So, we'll just first of all bring ourselves up to 3000 RPM here at 60 kPa.
07:37 And just see how close we are to our target of lambda one.
07:42 Just get ourselves in the middle of this box, I'm actually just going to turn our colour gradient off at the moment so I can do a better job of seeing where exactly we're operating.
07:51 So, we're not bad, pretty good, we're about 1% too lean so what I'm going to do is just, let's get rid of our log value crosshairs for a moment just to make things nice and simple.
08:04 Yeah, I don't think we get too much closer to our target than that.
08:08 OK so let's just now go to our AFR target table.
08:20 For some reason the search box on my laptop tends to be quite laggy from time to time.
08:25 Alright, so what we're going to do is again just to avoid interpolation, we'll highlight the entire cells around as well and we'll go to lambda 0.9.
08:33 Now, before I press enter I'm just looking at my lambda, it is absolutely on target, press 0.9.
08:39 I mean it's not bad but we're maybe 1.5% too rich.
08:43 OK so we already know what we need to do here, we'll go to our fuel main.
08:48 And again everything's just operating a little bit slowly here.
08:52 Go to our fuel main and we're too rich so we want to retard it.
08:57 Now, I wouldn't suggest maybe making a change of one degree or something like that, let's start by just making a big change, let's just halve that.
09:05 Now, what we want to do, it's actually brought us onto our target so normally what I would do here though is just reverse that change.
09:11 So, we'll go to lambda 1.
09:14 And we're right on our target.
09:16 Normally the reason we want, no I haven't actually mentioned this, we want to do this before we start dialling in our VE table because the problem being is that if we make adjustments to our charge cooling coefficient after we've tuned the VE table, this often, or usually, in this case it's not appearing to, but usually it will actually throw out the accuracy of our VE table.
09:37 So, we want to make that our first port of call, just chuck some numbers in that are at least close enough to get us up to a point like 3000 RPM part throttle load and then we can go ahead and optimise our, now we're a little bit out, so 0.98, 0.99, so let's just go to our fuel table.
09:54 Go to that cell there and we'll just remove percent.
09:57 OK right we're on our target.
09:59 Now, it took a little bit of time but it did prove my point.
10:02 OK so let's try again and see how we get on with our 5 degree instead of our 10 degree charge cooling coefficient.
10:11 We'll again enter 0.90, press enter.
10:17 Bang on our target, maybe, maybe if I was being really fussy, possibly a touch too lean but we're there or thereabouts.
10:26 So, we can obviously start getting a little bit more granular with this but that one change, halving the charge cooling coefficient from 10 degrees to 1 degree has got us where we want to be.
10:37 So, obviously once we've done that, we also want to make sure that we are setting all of our lambda targets back to where they were previously.
10:46 Otherwise that could end up being a little bit ugly later on.
10:49 Now, as I mentioned, it's going to be a short webinar, that's basically it.
10:53 If you do have any questions, please make sure you've got those into the chat, or jump into the chat in a second and see, or jump into our document in a second and see what we've got in there.
11:01 But I just want to reiterate the process here.
11:04 So, remembering that the importance of the charge cooling coefficient is so that the ECU's able to do a more accurate job of tracking requested lambda or AFR target changes.
11:15 A lot of this comes into some of the background calculations that are going on with the ECU for corrections as well.
11:21 So, we want this to be right so that the ECU can do its job properly.
11:25 But we also need to make sure that in order for this to be effective, we need a good set of known injectors with proper characterisation data, and we want to make sure that that data is properly entered into the ECU.
11:39 From there we want to run the engine at under light load and moderate RPM so that we're outside of that non linear region of the injector's operation.
11:48 Tune our engine to achieve a target of lambda one, then change our lambda target from one to 0 .9 and look at how accurately it tracks that change.
11:57 If it overcompensates and we end up with our new lambda too rich or richer than our target, we're simply going to reduce that charge cooling coefficient and try again.
12:05 If we end up too lean, we increase the charge cooling coefficient and go again.
12:09 And remember again this must be done before we tune the VE table because it can affect the accuracy of the VE table.
12:17 So, yeah really simple little step there but as I mentioned, I know it's something that a lot of people don't understand and a lot of people are probably just ignoring and leaving with that default value because they don't know what to do with it.
12:30 We'll get into our questions now and see what we've got in there.
12:34 Looks like we've only got one from Mike who's asked," I found when I changed the lambda target, the lambda stayed pretty well on target even with the closed loopers off, is the 10°C rate likely to be pretty good in most cases?".
12:45 Look, it's probably going to be there or thereabouts, it's probably going to be close enough which is why probably most people aren't touching it and it's not going to mean that your tune's going to be completely out of the ballpark.
12:57 Remembering as well that in most instances we aren't necessarily going into the AFR target table and making changes there once the engine is tuned.
13:07 So, it doesn't really affect us making a change like that but it does affect, as I mentioned just before, some of the compensations and background calculations that the ECU is doing.
13:18 So, hey if it's close enough, not really a big deal, 10°C is probably going to be in the ballpark and as you notice there with that first change I made with the default value of 10°C, I mean we're probably 1% off.
13:30 It's kind of within the margin of error anyway.
13:33 Alright, we'll leave it there, no more questions and remember if you are watching this webinar at a later point in our archive, if you've got questions on it, please ask those in the forum and I'll be happy to answer them there.
13:43 Thanks for watching.