Practical Reflash Tuning: Step 5-A: Optimising the Tune
Watch This Course
$229 USD
-OR-
Or
8 easy payments of only $28.63
Instant access. Easy checkout. No fees.
Step 5-A: Optimising the Tune
20.13
| 00:00 | - Now that we've got our mass air flow sensor calibrated correctly, along with our speed density subsystem, we can get into the next step of our process which is where we're going to actually optimise our tune or optimise our calibration on the dyno. |
| 00:13 | Now before we do get started with this, there's a couple of things we wanna cover off inside our VCM editor software, a few things we may choose to change. |
| 00:21 | So let's head there now. |
| 00:23 | For a start I'm going to look at our power enrichment wide open throttle lambda table. |
| 00:29 | Now you'll remember that in order to simplify the calibration of our mass air flow sensor, we started by setting this to a fixed value. |
| 00:37 | So our entire table is currently set to Now we may choose at this point to select some different lambda targets based on our experience or what we are expecting the car will respond to. |
| 00:50 | In our instance I'm actually going to leave this set to 0.78 for the moment. |
| 00:56 | The reason I'm going to do this is that I know this engine is quite knock sensitive on our local 95 octane fuel and just by targeting slightly richer than I normally would for this type of engine, this is going to help cool that combustion charge temperature and hopefully allow us to sneak in just a little bit more timing. |
| 01:15 | So I'm expecting we may be able to get a little bit more power with this richer target lambda. |
| 01:19 | Of course we are free to continually change and update this based on our findings while we start doing some ramp runs on the dyno. |
| 01:27 | So we're not stuck with this, we can easily alter it. |
| 01:30 | And this is really the key once we've got our mass air flow sensor calibrated correctly, we can easily make adjustments to our target lambda in this table, and the ECU will track those changes perfectly. |
| 01:41 | Next we're going to head across to our spark tab and we're looking here at our advance tab. |
| 01:47 | Now we've already discussed the fact that we are going to be doing the majority of our ignition tuning in our borderline advance tables that we've got here, and again we already know we've got one of these tables for each of our mapped points. |
| 02:00 | Now what I wanted to mention here is that there are a couple of techniques available that we can use to dial in these advanced tables. |
| 02:08 | And really this is going to come down to your own personal preference, how much modification has been made to the engine, in other words how far from standard it is, and also to a big degree, how long you're prepared to spend on the calibration. |
| 02:23 | So one of the options we've got is that we can fix the ECU to run on a single mapped point table. |
| 02:30 | Now the way we can do this, if we head back to our air flow tab, and head across to our variable camshaft tab, we have our mapped points configuration. |
| 02:40 | So let's just click into that just to refresh our memory, and we can see that the mapped points that are enabled are set to one. |
| 02:47 | So for example if we decided that we wanted to set the ECU to only run on mapped point zero, what we could do is simply select the other mapped points and set these to zero and that would have the effect of setting the ECU to only run on a single mapped point table. |
| 03:04 | Now this is going to be quite time consuming. |
| 03:07 | In our instance we've only got five tables that we're operating in, but depending on the particular engine you're tuning you may have many more tables and if you're going to go through and optimise each one individually, you're going to be spending a huge amount of time on that task. |
| 03:22 | So the way we're going to proceed with our tuning is that we're going to make our changes across all of the tables simultaneously. |
| 03:31 | And what I mean by this is that if we happen to find our ECU is operating on mapped point zero, we're at 3000 RPM and 1.2 load and we find that our engine is knocking, rather than only removing timing from that particular mapped point, we're going to remove the same amount of timing from all of our borderline advance tables. |
| 03:53 | So this just allows us to get a good result and it's going to dramatically cut down on the time we're going to need to spend. |
| 04:00 | So the reasoning behind this is that it's fair to assume that if our ignition timing is over advanced on one of our map point borderline advance tables, causing knock, it's likely that it's also over advanced on the other mapped point tables so this is why we're going to make that across the board adjustment. |
| 04:17 | It's also worth just revisiting our distance tables and just having another look at those. |
| 04:24 | You'll remember that in our ECU we have the option to run in various modes. |
| 04:29 | So in this case we can run in emission reduction, fuel economy, optimal stability or also we have optimum power modes. |
| 04:38 | Now you'll also remember that in our particular calibration we are always working in the fuel economy mode. |
| 04:44 | So this means that we are also always working on our best fuel economy distance table. |
| 04:51 | Let's just open that up again. |
| 04:52 | You'll remember that we have our load versus our engine RPM on this table. |
| 04:57 | And the numbers in this table define for any combination of load and RPM, which of our mapped points are being accessed. |
| 05:05 | You'll also notice here that any time we're above around about 0.8 load, we are operating solely on our mapped point zero table. |
| 05:15 | This is important because this is the table that we're going to be accessing during wide open throttle ramp run operation. |
| 05:22 | And this is where we're going to be spending the majority of our time during our calibration. |
| 05:27 | Alright so let's get our engine up and running and we'll perform another ramp run and have a look at our results. |
| 05:37 | Alright we've got our engine up and running and while we're just allowing the temperatures to stabilise I'll just go over a few of the aspects we've got in our scanner configuration. |
| 05:46 | First of all you'll notice that I am logging the parameter VCT schedule mode. |
| 05:52 | So you can see at the moment this is sitting on best fuel economy, and this is important particularly if you have your ECU calibration using multiple operating modes, you'll be able to see which mode is being accessed at any particular time. |
| 06:06 | An important parameter that we are going to need to log is our air load. |
| 06:10 | We can see we're logging that in our parameters list here. |
| 06:13 | So this is the load the engine is operating at. |
| 06:16 | I'm also logging this in our chart recorder here along with the cam timing that the engine is operating at. |
| 06:23 | Now another aspect that is vital to log is our mapped point waiting. |
| 06:29 | So I've got these set here, we're logging mapped point zero through to mapped point four waiting. |
| 06:36 | And what this does is it lets us see at a glance which particular mapped points are being accessed and used at any point in time. |
| 06:45 | So we can see for example here, we are operating around about 50% on mapped point two and 50% on mapped point three. |
| 06:54 | So this allows us to be really accurate when it comes to making changes to the likes of our spark advance. |
| 06:59 | We know that right now the ECU is interpolating between the value in mapped point two and mapped point three for our borderline advance tables. |
| 07:08 | And that is going to give us the final ignition timing that is being delivered to the engine. |
| 07:14 | We can see that's currently sitting at 40 degrees. |
| 07:17 | Alright so let's get a run underway and we'll see the results of the changes we've made so far. |
| 07:47 | Alright we've got our first run complete there and on the dyno we can straight away see that we've actually already picked up some power, we're a touch over 400 wheel horsepower. |
| 07:55 | Right now we're not that interested in the power though, we want to just get this tune dialled in. |
| 08:00 | So let's have a look in our scanner and we'll see what's been going on in there. |
| 08:03 | First thing we want to check is our fuel delivery, make sure that we're matching our target there. |
| 08:09 | Now if everything from our last step is on point, then obviously our fuelling should also be on point. |
| 08:16 | At the bottom here we've got our target versus measured lambda. |
| 08:19 | And we can see everything's tracking quite nicely there, as it should due to the closed loop control. |
| 08:24 | To really see how well that closed loop control is working though, or how much it's working, we need to look at our short term and long term fuel trims. |
| 08:33 | We can see particularly during the area of our wide open throttle ramp run, our closed loop trims are all sitting nicely and on zero or very close to, exactly what we'd expect given our work in our last step. |
| 08:44 | Alright so let's jump up and we'll have a look at our ignition timing. |
| 08:48 | And what we can see straight away in our chart here is that we've got some knock occurring still. |
| 08:54 | Now you'll remember we've still got a global minus two degrees in here as well, so this is still a little bit worrying, it means we've got a little bit of work left to do here. |
| 09:04 | We've got a couple of spikes of detonation, we've got knock occurring during the settle time when we've first gone to wide open throttle. |
| 09:11 | And we've got another knock occurrence happening at around about 2500 RPM, and then we've got a little bit more light knock occurring higher up it the rev range. |
| 09:21 | So we're going to need to perform some work there in our spark tables. |
| 09:25 | Now we need to also see where we were operating. |
| 09:28 | Now we've already looked at our distance table, we know we should be operating on mapped point zero, but let's just have a look at that again. |
| 09:34 | So we can see during our wide open throttle ramp run our VCT schedule mode is still sitting on fuel economy, as we know it will be. |
| 09:41 | And we can see that we are 100% weighted on mapped point zero. |
| 09:47 | So essentially everything we would be expecting there. |
| 09:50 | If we look at our cam timing we can see that the cam timing in this situation is sitting at zero degrees, again exactly as we know it should be, based on the variable cam set up in our editor. |
| 10:02 | Alright so the first change we're going to make here is to remove some more timing. |
| 10:08 | Now remembering we've got that global minus two degrees, so we're going to actually start by removing two degrees from our tables everywhere while we're under wide open throttle conditions. |
| 10:18 | And then we're going to go a little bit further. |
| 10:20 | So let's head back across to our editor and we'll see how we can do that. |
| 10:24 | Alright we're in our spark advance tab and the very first change we're going to make is we're going to zero out our global spark adder. |
| 10:32 | So we're going to set that back to zero. |
| 10:34 | Now we know that we are still seeing some knock occur, even with minus two degrees being pulled out across the board. |
| 10:41 | The knock that we are seeing though is all under high load, so there's no need to retard the timing in the cruise light load areas of the table, this is probably going to be detrimental to our engine performance. |
| 10:54 | So what we wanna do is take into account that we were seeing knock even with two degrees out of our timing tables and we're going to see how we can modify those tables. |
| 11:03 | So let's start with our mapped point zero table. |
| 11:06 | This is the one we know that we're operating in primarily when we are operating under wide open throttle. |
| 11:11 | What we're going to do is start by highlighting the entire table above a load point of 1.00 and we're going to remove four degrees from that area. |
| 11:22 | Now this takes out the two degrees that we've already had from our global spark adder, plus we're removing a further two degrees. |
| 11:29 | Now I'm also going to smooth that out, we're going to remove two degrees from our 0.7 load area. |
| 11:37 | So with that change now made, we're going to then go and apply the same changes to our other mapped point tables, let's go and do that now. |
| 12:03 | Alright we've made those changes to all of those tables. |
| 12:05 | For the time being we're going to leave our target lambda at 0.78 Let's now flash that calibration into the ECU, we'll get up and running, and we'll perform another test. |
| 12:16 | Alright we've got our engine back up and running. |
| 12:18 | So we'll perform another ramp run and see the effect of those changes. |
| 12:44 | Alright so we can see there, the results of those changes, we have lost a little bit of power, and that's really to be expected because we have retarded the timing. |
| 12:52 | Now this is an area where chasing power alone really is not our aim. |
| 12:57 | There's no point making power if our engine is suffering from knock. |
| 13:00 | So we need to get our timing under control so we're not constantly suffering from detonation. |
| 13:05 | Once we've got that, we've at least got a safe calibration even if that does sacrifice some power. |
| 13:11 | In the top end, we can see that the power is relatively unchanged. |
| 13:14 | Let's have a look in our scanner and we can see we've still got a little bit of knock activity but we can also see that this is predominantly down in the low RPM range at the start of our run. |
| 13:26 | In particular we can see right up in the top end we've got no knock activity. |
| 13:30 | And we can actually see that our knock retard has ended up adding some timing back in. |
| 13:37 | Remember that it has the ability to add timing if there is no knock active. |
| 13:43 | So what we wanna do here really is focus on getting rid of this knock that we're seeing down in particular here, 2400 RPM, as well as when we're setting the engine down at 1500 RPM So let's look at this 2400 RPM point in particular. |
| 13:58 | We can see we're at 2400 RPM and our load is about 1.0, 1.1 So that's the area we're going to need to make that change. |
| 14:07 | Let's head back to the editor and we can make that change. |
| 14:09 | We know that we are working on mapped point zero here, and the point we were looking at is around about 2400, 2500 RPM, and we were 1.0, 1.1 in our load. |
| 14:21 | So we're operating specifically around about this area here. |
| 14:25 | So what we're going to do is highlight this particular area of our table. |
| 14:29 | And we were at 2400 so I'm going to extrapolate this change back a little bit back down to 2000 and let's remove another two degrees from this area. |
| 14:38 | Now we'll also extrapolate this back up, we were also suffering from some knock right at the beginning of our run. |
| 14:46 | Now we're around about 0.8 to 0.9 in our load right at the start of the run where we're settling so we'll just make a change here, we'll take another two degrees out at our load point of 0.7 and we'll also make a further change here at 1.0 Alright so now we're made that change, we're going to make exactly the same change to our other mapped points, let's go ahead and do that now. |
| 15:25 | Alright we've made those changes now we'll flash that through to the ECU, make a further run on the dyno. |
| 15:31 | Alright we've got our engine back up and running, let's perform another run on our dyno. |
| 15:57 | Alright we can see we've essentially got the same peak number there, just on 400 wheel horsepower, but you can see that we've picked up quite a sizeable chunk of power through the mid range. |
| 16:05 | Let's jump int the scanner and we'll see why that's the case. |
| 16:09 | Now straight away we can see that we've eliminated most of that knock retard. |
| 16:13 | We've still got a little bit going on there but it's a lot better than our last run. |
| 16:18 | And particularly if we look at the point where our ramp run actually starts, our knock retard, we've had a couple of knock events there but we're only pulling out around about half a degree, so essentially nothing much going on there. |
| 16:30 | What we can see as well is because there isn't any retard going on early in that run, we've ended up with more timing through that mid range and we can see that in particular here from around about 3200 RPM and above, our knock retard is actually adding three degrees back in. |
| 16:47 | So from here we're not gonna continue this process, it just becomes iterative, we're going to continue removing timing from areas where we do have knock retard active and we're going to rinse and repeat what we've already seen. |
| 17:01 | What I would say though is looking at the data we've got now, I believe that the engine is going to want to take a little bit more timing from around about 3000 RPM and above and we're seeing that by the knock retard system adding timing back in. |
| 17:15 | So we're going to remove a little bit of timing down low and we're gonna add some timing in above 3000 RPM and we'll continue until we've got our timing optimised with minimal knock retard. |
| 17:26 | Let's jump ahead and we'll see what that looks like. |
| 17:30 | Alright so on the dyno we have our final run and this is overlaid with our base run. |
| 17:35 | So you can see our final numbers were just under 420 wheel horsepower. |
| 17:39 | And what we can actually see is despite the fact that a lot of our work has been involved in removing timing, we'll actually see that we've picked up quite a decent amount of power from around about 2500, 3000 RPM and above. |
| 17:52 | And again this is just because now that knock control strategy is doing less work and we've also reduced the delay in the ignition timing being reintroduced after a knock event. |
| 18:05 | Let's jump into our scanner and we'll have a look at the results of that ramp run. |
| 18:08 | Now for a start we'll just have a quick look at our fuelling. |
| 18:10 | So this is essentially our final run, so we have gone through and made some adjustments to our fuelling as well. |
| 18:16 | What we've ended up doing, you'll remember we started with a flat 0.78 lambda target. |
| 18:21 | We found that the rich mixture wasn't really helping our knock down low, so we've ended up leaning out our mixture at low RPM to about 0.82 lambda and you can see we've got a general sort of richening as we go through the rev range so we're still hitting our target of 0.78 at higher RPM. |
| 18:42 | So in this particular run we have still got one little section where we're showing a little bit of knock activity. |
| 18:49 | You can see that at that point here, we are pulling 1.5 degrees. |
| 18:54 | It actually looks more dramatic than it is because during most of this run, the knock control strategy has actually been adding timing back in, for example here at the start of the run we can see that we're adding three degrees in. |
| 19:06 | Likewise at the top of the run we're also adding three degrees back in. |
| 19:10 | So what we want to do here is also apply a little bit of common sense. |
| 19:14 | When we're trying to remove any knock events we wanna be sure that these knock events are actually real and are consistent. |
| 19:23 | What I mean by this is we'll probably want to do a couple of runs and just make sure that the knock that we're logging is repeatable. |
| 19:31 | If it's just a one single random knock event, we're going to ignore that. |
| 19:35 | If we're repeatedly run after run getting knock events happening in the same area then of course we want to remove some timing, we've gone through how we can do this. |
| 19:44 | So at this point we've essentially got our calibration complete under wide open throttle ramp run conditions. |
| 19:50 | We've got good control over our fuelling, we've got good control over our spark advance and we don't have any excessive knock events occurring. |
| 19:57 | We do however still need to consider the part throttle areas of our map in terms of our spark advance. |
| 20:04 | So in the next section we're going to have a look at how we can use the scanner to help us dial those areas in. |
