Practical Reflash Tuning: Step 4: MAF/Injector Scaling
Watch This Course
$229 USD $114.50 USD
-OR-
Or
8 easy payments of only $14.31
Instant access. Easy checkout. No fees.
Step 4: MAF/Injector Scaling
34.29
| 00:00 | - Now that we've got our base ROM configured to suit our supercharger installation, the next step in our process is to look at our MAF scaling, and our injector scaling as applicable. |
| 00:11 | Now, we already know that we're running the factory injectors, and we're running the factory MAF sensor in the factory housing. |
| 00:18 | So on the face of it, it doesn't appear like there's much work to do here. |
| 00:22 | It is still, however, a really good idea before we start doing any tuning, to just check and make sure that our MAF scaling is correct. |
| 00:31 | We're going to have a look at how we can do that using both a wide-band oxygen sensor as well as the closed loop fuel trims, both the long-term and short-term fuel trims. |
| 00:42 | So why would we maybe have some inaccuracies in here? Well, anytime we're making changes to the intake tract, this can affect the sensor calibration, the MAF sensor calibration. |
| 00:54 | When the engineers are calibrating this MAF sensor, they'll be doing so with the entire intake tract, so anything in there that we modify, be it the air filter, or the intake plumbing after the air flow metre, this can affect the accuracy of the calibration. |
| 01:11 | So we're going to look at that and see if we need to do any work. |
| 01:14 | Now before we do that, let's talk about why we may need to correct our MAF sensor, why we need to have this accurate in the first place. |
| 01:23 | A lot of people think that the only error we're going to see creep in if our MAF scaling is wrong, is that our fueling might not be quite right, and we can simply adjust this and correct it, work around that in our AFR target table. |
| 01:38 | However, the ECU is calculating load based on that MAF sensor output, so if the MAF sensor isn't correct, it's going to affect the load signal the ECU is receiving and that can be the input, the load axis input for multiple tables, so it's going to affect the entire running of our engine. |
| 01:58 | The other thing to consider as well is, normally when the ECU moves from closed loop into open loop operation, it will take the current long-term fuel trims with it and apply those in open loop. |
| 02:12 | Now, on the dyno, that might not present a problem. |
| 02:15 | It takes some time for the ECU to learn these long-term fuel trims. |
| 02:20 | So what we can get into is a situation where, if our MAF scaling's quite inaccurate, we can still get the right air/fuel ratio under open loop conditions on the dyno, but over time, perhaps the next few days or even a couple of weeks, the ECU will learn and apply these changes to the long-term fuel trims, and then these will be applied in open loop and this can result in quite dramatically different air/fuel ratios than what we saw when we were on the dyno. |
| 02:49 | Okay, that's what we're trying to achieve and now you understand why we're trying to achieve it. |
| 02:54 | Let's go through the process now of flashing our base map into the ECU and we'll have a look and see how close our MAF scaling is to start with. |
| 03:04 | So we can do this by clicking on the tools menu, and we can click on detect vehicle. |
| 03:11 | So this will detect the ECU, and then we can go through the programming process. |
| 03:18 | So once that's detected the ECU, we want to click on the program engine ECU option and click okay. |
| 03:25 | Also open up our programming tools box, and you can see here at the top we have our ROM file to program. |
| 03:32 | So this is the ROM file that is already loaded and ready to be programmed into the ECU. |
| 03:38 | It corresponds to the ROM file that we have open over here at the moment. |
| 03:42 | Now of course if we want to choose a different ROM file, we can simply click choose ROM file, and choose the ROM that we want from all of the ROMs available. |
| 03:52 | In this case, this is the ROM we want to program into the ECU, so we simply click the program ECU button, and the ROM file will be flashed into the ECU. |
| 04:02 | This does take around about two minutes, so you do need to be a little bit patient, and simply follow the on-screen prompts because at the end of the flashing process, we do need to key the ignition off and back on again to reset the ECU. |
| 04:16 | Once our ECU is re-flashed, we're free to start the engine up again, and we want to allow it to idle and just reach a normal operating temperature. |
| 04:27 | So what we're going to do, is we're going to have a look at some logging from the ECU while this is occurring. |
| 04:34 | And we can do this by simply clicking on the live data tab here, and this is going to show us the logging parameters that we're currently going to be looking at, and you can see that these are listed here. |
| 04:47 | If I click on the map access tab here, this will start logging live data to our screen, and if we want to actually log this to a file so we can view it later, we need to click on the log to file button. |
| 05:01 | Now when we click on the log to file button, you can see that the status bar here turns green to show that the ECU is currently, the ECU data is currently being logged to a file and when we click on log to file again it will go back to its normal colour. |
| 05:19 | Let's have a look at these parameters and just discuss what we have available here. |
| 05:23 | We have a huge number of parameters that are available to log, and if we click on the all parameters tab here, the full range of options we have, parameters available, will be shown. |
| 05:38 | Now, it's always worth logging a relatively minor, or minimal data-set so that this allows the data to be logged more frequently, and it also means we're only looking at the parameters that are of actual interest to us at the time. |
| 05:53 | So let's look at what we've actually got here. |
| 05:56 | So first of all, we have our fuel system status, so this tells us simply whether the ECU is in open loop or closed loop mode and that's going to be important for this process. |
| 06:06 | Next we have our fuel trim short term and fuel trim long term. |
| 06:11 | This is the current feedback values from the built-in factory wide-band oxygen sensor, or Lambda sensor that helps the ECU adjust the fueling in order to reach our target air/fuel ratio. |
| 06:25 | We've got accelerator angle, or throttle position, which we're going to again be looking at later. |
| 06:31 | Next we have our equivalence ratio commanded, so essentially this is our commanded air/fuel ratio, or what the ECU is requesting. |
| 06:40 | And you can see at the moment this is sitting at 14.66 which is our stoichiometric air/fuel ratio. |
| 06:47 | Next we have our AFR value, this comes from the on-board Lambda sensor, and while it is a wide-band sensor, it's only really accurate again around the stoichiometric air/fuel ratio. |
| 07:02 | It can only read as rich as 12.17 to one, and if we compare it around those air/fuel ratios to a proper wide-band sensor, we'll find that the accuracy is a little bit lacking so we really can't rely on that for our tuning, and for that purpose you can see I've inputted our Innovate LM-2 wide-band into the logging here and that particular parameter is an accurate wide-band air/fuel ratio input. |
| 07:31 | Next we have our mass air flow in grammes per second and a little bit further down we also have our mass air flow output in voltage so we can see exactly what the ECU is receiving in terms of our mass air flow input and what voltage the mass air flow sensor is working at. |
| 07:48 | We have our engine load and engine speed, and finally we have intake air temp and coolant temp. |
| 07:55 | Now these parameters are important because when we're logging this sort of data, we want to make sure that we are logging under normal operating conditions. |
| 08:05 | What I mean by this is, normal engine coolant temperature that we're likely to see when we're actually driving the car out on the road, and likewise with our intake air temperature. |
| 08:15 | So particularly if the engine is cold, we may have additional enrichments being requested, and likewise if the intake air temperature is very hot, perhaps the engine is heat-soaked, that's going to affect the result of our closed loop fuel control. |
| 08:30 | So we're generally operating around about 88 to 92 degrees Centigrade, when the engine is at normal temperature, so we're about there now so let's have a look at what our short term and long term fuel trims are doing. |
| 08:45 | Now remember, the overall trim that is being delivered is the combination of the short term and long term. |
| 08:52 | And right now you can see we're idling with a positive trim in our long term of about 12%. |
| 08:58 | This means it's adding 12% additional fuel. |
| 09:02 | And what this straight away means is that our MAF scaling is likely to be somewhat inaccurate. |
| 09:09 | Now let's just confirm that, and what I'm going to do is bring our engine speed up a little bit on the dyno, we're going to put it in fourth gear and we're gonna drive the car at about two and a half thousand RPM. |
| 09:21 | This is going to replicate where we'd sort of be driving the car out on the open road or on the freeway, motorway, and we're just using a very light amount of throttle and we want to see what our fuel trims are doing under this sort of operation. |
| 09:37 | And you can see now we have a short term fuel trim of about six to seven percent. |
| 09:43 | It does move around a little bit, but that's the sort of magnitude of the error we've got. |
| 09:49 | So from that very brief test, what that means is we do have some error in our MAF curve and we're going to need to address that. |
| 09:58 | So what I'm going to do now is, we're going to go and make some changes to our MAF scaling and initially, I'm going to make one simple overall trim right through the entire operating range of our MAF scaling curve, and then we're going to make some other changes to our tune that are going to allow us to dial that scaling in more accurately. |
| 10:21 | So now that the engine's back to idle, and you can see that our overall trims are sitting at around about minus nine percent. |
| 10:29 | Sorry, positive nine percent. |
| 10:32 | So remember, we had a trim in our cruise area of around about seven percent and we've got around about a nine percent positive trim at idle. |
| 10:41 | What I'm going to do is simply go back to our maps here, and I'm going to go into our sensor scaling, and our MAF sensor scaling, and this is where we can actually see our MAF transfer function, our MAF calibration, and I'm going to highlight this entire MAF curve for the moment, and I'm just going to press control-P to make a percentage change, and I'm going to increase the entire MAF transfer function by seven percent, and that's going to add seven percent so that's hopefully going to correct a lot of our error. |
| 11:17 | So now we're going to close that down, we'll save it. |
| 11:21 | Now the other thing we're going to do, our closed loop trim's our, short term or long term fuel trims are only useful to us, scaling that MAF calibration when the engine is in closed loop. |
| 11:33 | When we're in open loop we're going to be relying on the feedback from our wide-band oxygen sensor, our Innovate LM-2 that's been input into our logging data. |
| 11:43 | So what we want to do is make our job a little bit easier. |
| 11:46 | We want to aim for a consistent and stable air/fuel ratio in open loop, so we're going to go into our fueling tab, and we're going to go into fuel map mode one. |
| 11:56 | And you'll remember that we had previously set up some broad targets for our air/fuel ratio in here. |
| 12:03 | What I'm going to do, just for the purposes of scaling this MAF, I'm going to select everything from one gramme per revolution and above, right through our entire RPM range and by pressing control-F, we can enter a target air/fuel ratio of 11 to one. |
| 12:20 | So all I'm doing here is choosing an air/fuel ratio that I know is going to be safe for me to operate the engine at wide-open throttle for the purposes of this testing. |
| 12:30 | Right, we're going to close that table down now and we will save that. |
| 12:34 | We're going to make one more change, and if we go into our air/fuel correction menu here, you can see we have an option here for our fuel trim long term minimum and maximum. |
| 12:45 | Now if we open these up, we'll find that from factory, the default values here are positive and negative 0.4 and this controls the range of long term fuel trims that are applied when the engine operates, transitions into open loop. |
| 13:01 | Now as we've already discussed, we don't really want these long term fuel trims being applied in open loop. |
| 13:06 | So what I want to do is go ahead and set both of these values here to zero. |
| 13:12 | So that will simply eliminate the long term fuel trims from being active when we are in our open loop operation. |
| 13:23 | Okay, so we're going to save that ROM file now and we're going to flash it into the ECU. |
| 13:30 | Okay so we've flashed our ECU, we've got it back up and running, I've allowed it to reach a normal operating temperature and you can see that we are targeting a stoichiometric air/fuel ratio. |
| 13:40 | So the first thing we can notice is straight away at idle, our combined short term and long term fuel trims are now very, very close to zero. |
| 13:48 | I'm typically trying to get my long term and short term fuel trims to within plus or minus five percent, and preferably closer to about two or three percent, and you can see that's exactly where we are here at idle right now, so that's a really positive result. |
| 14:03 | Let's get the engine running now on the dyno, and what we'll do is we'll get ourselves into fourth gear and we'll get to about two and a half thousand RPM, and we'll have a look at how the closed loop trims are working as we increase the air flow by opening the throttle further. |
| 14:19 | So I'm gonna start by getting myself to a point where the engine's running and I'll close the throttle down so I can get the minimum amount of air flow possible. |
| 14:28 | So I'm just simply looking here at my mass air flow value, and you can see I can get down to around about nine or 10 grammes per, grammes per second. |
| 14:38 | If I go much further the, the engine simply slows down on the dyno. |
| 14:44 | So now what we wanna do is have a look again at our trims, and you can see at this point, we're sitting about negative three percent so again we're still pretty close. |
| 14:53 | I'm quite happy with that result. |
| 14:55 | I'm just going to open the throttle slowly now, and it's important when we are doing this sort of testing to make sure that we make very smooth and consistent changes to our throttle opening. |
| 15:08 | If we do anything aggressively, then it's going to bring in transient enrichments which we don't want. |
| 15:13 | I'm just gonna log this to file, so that we can view this data later. |
| 15:17 | Okay, so we're going to open the throttle and just watch what happens to our trim values here. |
| 15:24 | At the same time we want to be watching, we want to also be watching our fuel system status. |
| 15:31 | Remember when that jumps from two to four, that means we have entered open loop operation and then our fuel trims aren't going to be active so we can't use those to help us tune the MAF scaling. |
| 15:45 | Okay, so we're still seeing, now we're up to 26, 27 grammes per second, and we're still seeing those trims sitting very, very close to zero. |
| 15:53 | We'll continue, we're up to 35 grammes per second now, again still very, very close to zero. |
| 16:00 | Now we're up to 40 grammes per second and we're probably just about at a point where we're going to transition to open loop and we've done that right now. |
| 16:09 | So you can see now our fuel system status has jumped to four. |
| 16:14 | Okay so, first of all, just visually watching those trims while we did that test, we could see that they were always within that range that I've been talking about, plus or minus five percent, and in general they were probably closer to plus or minus three percent. |
| 16:29 | So that's really positive for one quick change. |
| 16:33 | Now let's have a look at our log file. |
| 16:35 | We'll set this up and we'll see what we've got to view. |
| 16:57 | Okay so on our screen here I've just set up the information that we're actually interested in. |
| 17:01 | At the top for start, you can see my accelerator position. |
| 17:04 | So this is what I was doing during this test. |
| 17:06 | You can see me just smoothly opening the throttle. |
| 17:09 | Next down, we have our fuel system status. |
| 17:11 | So this is just so we can confirm that the ECU is in fact in closed loop mode, and you can see near the end of the log file, this is where it transitioned to our open loop. |
| 17:23 | Now the parts we're really interested in here, are our two fuel trims. |
| 17:27 | Our long term is shown here, and our short term is shown below it. |
| 17:32 | Now if we look at what's going on through this data, we can click at a particular point in the table and the current trims at that point are shown over here on the right. |
| 17:43 | So we can see that we have at this point a negative three percent trim, and if we move through, we can see that in this point our short term fuel trim now has moved up to closer to zero, and matter of fact our long term fuel trim has dropped to negative three. |
| 17:57 | So essentially we have the same result. |
| 17:59 | Now, a lot of people will go through and make very precise changes to our MAF curve at individual voltage break points in the curve, trying to chase down in getting an absolute perfect zero percent trim throughout the curve. |
| 18:12 | Now while on the face of it this seems sensible, trying to make such precise changes to our MAF scaling really isn't going to actually help us and in fact sometimes it can make our life harder. |
| 18:24 | Now remember that we looked at the shape of the MAF calibration curve initially and it's a nice, smooth, exponential shape and we really want to keep that shape. |
| 18:33 | When we're making changes, we want to make smooth changes to the curve rather than trying to make erratic changes where the MAF scaling starts moving up or down. |
| 18:43 | It's going to introduce problems in drivability later on and again, make our life harder. |
| 18:48 | So what I'm looking at here is, generally from the start of our file here, we're about 10.8 grammes per second to around about this point here in our table where we have negative 1.6, negative 2% trim. |
| 19:07 | We've got a relatively smooth trim of about minus three, maybe minus two percent and that goes out to, as we can see here, to 30 grammes per second. |
| 19:19 | Now above 30 grammes per second, we can see that our trims now are actually very close to zero. |
| 19:26 | We have a small area here where we have negative 1.6 but I'm not going to be too worried about that at this point. |
| 19:32 | So we can see in the closed loop area of our curve, from around about 10 grammes per second up to about 30 grammes per second. |
| 19:40 | We need to take around about two percent out of our MAF scaling. |
| 19:44 | So let's close down our log file. |
| 19:46 | We'll open up our maps and we'll make that change now. |
| 19:51 | So again we can just simply come down to our MAF scaling, and we were looking to make that change between around about 10 grammes per second, and I'm just going to go a little bit lower than that to nine grammes, and out to around about 30, and if we press control-P, we can take two percent out by entering a value of 98. |
| 20:14 | Now what I'm going to do as well is just smooth that change out. |
| 20:18 | So I'm going to now highlight a section, remember we were idling at around about 2.3 to 2.5 grammes per second, and we were pretty good there. |
| 20:27 | So what I'm going to do is smooth this change by making it a one percent change below that. |
| 20:32 | Again, we're all about trying to keep a smooth, consistent shape to our MAF calibration. |
| 20:38 | We'll close that down for the moment but we are going to come back to it. |
| 20:41 | Now what we're going to do is start our data logging again, and now what we're going to do is look at a wide open throttle ramp run on our dyno, and we're going to now look at our scaling in open loop. |
| 20:55 | So here what we're going to be doing is looking at the input from our LM-2, which we can see being displayed here. |
| 21:02 | We're going to compare that to our commanded air/fuel ratio. |
| 21:07 | So let's perform a ramp run now, and we will log this result to file so we can view it later. |
| 21:32 | Okay so we've completed that run there. |
| 21:34 | We've got a power figure on our dyno but at this point we're not interested in the power at all. |
| 21:39 | All we're looking at is getting our MAF scaling dialled in. |
| 21:42 | It's important to watch our air/fuel ratio during this process because if it is a long way off, we may need to abort the run and this is another reason why we start with a safely rich air/fuel ratio target. |
| 21:53 | So let's open our log file now, we'll get that set up and we'll have a look at the data and see what that tells and what we need to do. |
| 22:01 | Okay so the first things we want to look at, we've got our engine speed up here on our top graph, and you can see that the area of the actual run is demonstrated where our RPM climbs, understandably. |
| 22:14 | So this is the are we're interested in. |
| 22:16 | The first thing we want to do is make sure that our fuel system status, here, has in fact gone to open loop during our ramp run, and it has, so that's great. |
| 22:28 | We don't need to worry about our fuel system status anymore. |
| 22:32 | Next, you can see that I've also retained our fuel trims both long term and short term, and what I'm looking for here is that during our wide open throttle operation, when we are in open loop, I want to make sure that both of our fuel trims are sitting at zero, so we're not influencing our air/fuel ratio. |
| 22:50 | Those are both zero, so we're happy with that. |
| 22:53 | Let's remove those just so we're not getting anything confused. |
| 22:57 | Now, the two aspects we're really interested in here are our equivalence ratio commanded, remember this is just the air/fuel ratio the ECU is asking for, and below this we have our LM-2 air/fuel ratio so this is from our wide-band. |
| 23:13 | And you can see that our equivalence ratio commanded, as we would hope, has been sitting at 11.0 right through our ramp run, so that's great, that's what we're asking for. |
| 23:23 | I'm just gonna zoom in on the actual area of the ramp run so we can be a little bit more specific. |
| 23:29 | Now let's look at our LM-2 air/fuel ratio, and if we click at various points on the table, you can see that we are in fact still a little bit leaner than our target, particularly at a point here, you can see that our measured air/fuel ratio is sitting at 11.8 so we definitely still have some work to do with our MAF scaling. |
| 23:53 | At the top end though, at 7,200 RPM here, you can see we're getting a little bit closer so, we're at about 11.2 so what we can do here is make some averages to the error. |
| 24:07 | So what we're really looking at is the error between our measured air/fuel ratio and our commanded air/fuel ratio, and I don't, again, I don't want to go through this making individual changes to a range of break points in my table. |
| 24:22 | What I want to do is look for a trend to this curve because if I've done my job correctly, I'm expecting the overall shape of that MAF calibration to remain consistent. |
| 24:33 | And you can see that here from around about, around about 3.2 volts, which corresponds to 71.3 grammes per second, and right to the top of our run, which we can see here is 200 grammes per second, we're sitting at around about 11.25 to one air/fuel ratio. |
| 24:59 | So this means we're a little bit leaner. |
| 25:01 | Yes, the air/fuel ratio does move up and down slightly within that range, but that's a relatively good average. |
| 25:07 | Okay, so let's go and make some changes to our MAF scaling based on this. |
| 25:12 | We're going to come back after that and we're going to address this lower area where we have that lean spike that we've already talked about. |
| 25:20 | So what we're looking at at the moment is anything from 70 grammes per second, upwards. |
| 25:26 | Okay, let's move back, we'll just minimise our log file, and we're going to go back to our maps again and our MAF sensor scaling. |
| 25:36 | So this time, what we want to do is calculate a correction factor to apply. |
| 25:40 | So I'm going to bring up the calculator function, and remember we had a measured air/fuel ratio of 11.25, so this is our measured air/fuel ratio. |
| 25:52 | Our desired air/fuel ratio was 11. |
| 25:55 | So this gives us a correction factor that we can apply to our scaling. |
| 25:59 | So in this case, the correction factor is 1.02 so this means we need to actually add two percent fuel in order to achieve our desired air/fuel ratio. |
| 26:10 | We can achieve this by adding two percent to our MAF scaling. |
| 26:14 | So, we want to add that from 70 grammes per second and above, so we're going to highlight that area of our calibration curve, and again I'm just going to use the percentage change function, and I'm going to enter a value of 102, that's going to add two percent. |
| 26:37 | Right, let's have a look at our logging again. |
| 26:40 | And we'll address this area a little bit lower down. |
| 26:44 | So from the start of our run, if we see the point at 1,500, 1,600 RPM, where the engine was settled on the dyno, again we were sitting at around about 30, 31 grammes per second and we're sitting at about 11.5 to one AFR. |
| 26:59 | The point I'm really worried about here is, or more interested in, is we were at about 11.8 which happens at 46 grammes per second. |
| 27:09 | So we're going to make a change at 30 grammes per second and again up to 46 grammes per second. |
| 27:15 | So let's close down, minimise our logger, and we'll have a look at that. |
| 27:20 | So, 46 grammes per second. |
| 27:22 | This is in this area here. |
| 27:25 | Now we can calculate the required change that we need to make at this point in our table, so let's bring up our calculator again and this time if we enter our measured air/fuel ratio at 11.8, and we divide it by 11.0 this is gonna tell us that we need to add seven percent to this area on our MAF calibration curve. |
| 27:46 | However, you can see, through this area here, we've currently got a nice, consistent shape to that curve. |
| 27:53 | Now we don't really want to, as I mentioned, we don't want erratic changes to this curve. |
| 27:59 | So if I add seven percent, that's going to end up with a really haphazard looking result here, which is probably going to be, give us a negative effect when we're talking about drivability. |
| 28:11 | So what I'm going to do instead, is I'm going to make a smaller change. |
| 28:15 | Remember, we were also lean down at about 30 grammes per second, so I'm going to highlight the entire area there, that we've just looked at so everything from 29 grammes per second up to 54 grammes per second, and I'm going to add another three percent to that area. |
| 28:35 | Now that's going to help correct that lean area that we just saw but it's not going to end up with a really dramatic change to our MAF transfer function. |
| 28:45 | Right, let's close that down now, and we can save it, and we can program that into our ECU and see if that's corrected our error. |
| 28:55 | Let's do that now. |
| 28:56 | So we've got our engine back idling, we've got it targeting a stoichiometric air/fuel ratio and again, we can see that our short term and long term fuel trims are really close to zero. |
| 29:07 | What we're going to do now is just repeat the process, so we'll get the engine running in fourth gear and we'll have a look. |
| 29:14 | As we move through the throttle range and increase the airflow, we're going to make sure that our closed loop trims are still very close to zero as we saw previously. |
| 29:26 | So let's log this to a file so we can view it later if we need to, and again I'm just closing the throttle down until I can get down to around about 10 grammes per second air flow, and you can see we're there now. |
| 29:41 | Again, our short term and long term fuel trims are sitting within plus or minus two percent. |
| 29:46 | I'm just gonna slowly open the throttle here. |
| 29:49 | I'm up to 15 grammes per second, 20 grammes per second. |
| 29:53 | So far about the maximum trim I've seen is about minus one, so we're very, very close to zero. |
| 30:01 | Move up to 25 grammes per second, again, still very close to zero. |
| 30:06 | We're up to 30 grammes per second now, and we'll continue up until the point now we've transitioned into open loop at about 35 grammes per second. |
| 30:18 | So about the biggest change we saw there was a two percent trim so I'm really happy with that result. |
| 30:23 | Let's now see how close we are if we do a wide open throttle ramp run. |
| 30:51 | Alright, we've completed that run and I'm watching the air/fuel ratio reading while we're doing that run, and I know already that it is very close. |
| 30:59 | Let's open up our logging and we'll set that up and see exactly what happened during the log file. |
| 31:05 | So again, we'll just start by initially making sure that our fuel trims in open loop were in fact zero. |
| 31:12 | You can see that they are. |
| 31:15 | You can also see that our equivalence ratio commanded, our commanded air/fuel ratio, was 11.0, so that's great, we'll get rid of our fuel trims now so that they're not cluttering up our log file and now we can see our measured air/fuel ratio from our LM-2 and if we go through the file here, you can see that now we are very, very close. |
| 31:39 | We've still got this lean area here which corresponds to that 45 grammes per second that we've previously talked about. |
| 31:48 | However, through the majority of the run, you can see that we are so close to our target now, we were always within about 0.1 of our AFR target. |
| 32:00 | Now, that's about as close as we need to get. |
| 32:04 | The reason for this, again, is we want to maintain a smooth shape to that MAF scaling, that MAF calibration curve. |
| 32:12 | The other thing is, we're going to struggle to get much more accurate than we are right now by changing our MAF scaling, and we're always going to potentially find small areas such as this particular one here at 45 grammes per second where the actual air/fuel ratio deviates slightly from our target. |
| 32:34 | Now we could try and correct this in the MAF scaling but from my experience, that's probably more likely to create problems in other areas. |
| 32:43 | So now that we're this close, what we're going to do is, if we've got any remaining error, we can now correct that in our AFR target table, our main fuel map. |
| 32:54 | So the process is really to get our MAF scaling to a point where our error is as close to zero as possible. |
| 33:02 | Worst case I would accept would be plus or minus five percent, but really more often than not as we can see here, we should be able to get our error to within plus or minus about two to three percent. |
| 33:13 | It's going to mean that we're going to be very accurate when we start putting numbers into our AFR target map and then, if we've got any remaining error between our air/fuel ratio target and our measured air/fuel ratio, we're actually going to correct that in our AFR target map, rather than in our MAF calibration. |
| 33:33 | So the key process here is, we're going to be looking in closed loop mode, at the combined short term and long term fuel trims, and we're going to log those as we vary the air flow across the entire range we can get to in open loop. |
| 33:50 | Once we have done that, and corrected that area of our MAF scaling, we're going to do the same under wide open throttle in our open loop range. |
| 33:58 | And remember, we're not trying to chase every individual error that we see in those curves. |
| 34:04 | We're looking for trends to the error that we can then apply to our MAF scaling that's going to give us an accurate result without ending up with an erratic shape to our MAF calibration, that's going to cause us more trouble later on. |
| 34:18 | Okay, now we've got our MAF scaling calibrated smoothly and correctly, we can move on and actually look at optimising our tune. |
