Practical Standalone Tuning: Step 10: Confirming a Tune on the Road/Track
Watch This Course
$229 USD $114.50 USD
-OR-
Or
8 easy payments of only $14.31
Instant access. Easy checkout. No fees.
Step 10: Confirming a Tune on the Road/Track
25.53
| 00:00 | - For the last step of our 10 step process we've taken our car to Highlands Motorsport Park so we can test out on the racetrack and make sure that everything we saw on the dyno still stacks up under real world conditions. |
| 00:13 | Now we're going to separate our process here out on the racetrack into a few different segments. |
| 00:20 | First of all we're going to spend some time out on the racetrack under cruise or steady state conditions. |
| 00:26 | Specifically here what we want to do is test from around about 2000 RPM up to perhaps around 3500, 4000 RPM. |
| 00:33 | And from low load where we've essentially got the throttle completely closed, up to our transition onto boost, this is going to represent one of the biggest areas that we're going to be spending time in, particularly with a road driven car, this would be steady state cruise conditions at normal open road speeds. |
| 00:53 | This will also allow us to move down into the light load areas which we couldn't get to easily on the dyno. |
| 01:00 | You'll remember that as we closed the throttle on the dyno to try and access those very light load areas in the fuelling, we end up with a situation where the engine doesn't produce enough torque to keep the dyno spinning and we just end up with the RPM slowing down. |
| 01:15 | On the road or the racetrack though the inertia of the car rolling will allow us to get lower in the load, we'll be able to access those sites and see whether our tuning or our guess or extrapolation of the tune zones into those areas has worked out. |
| 01:27 | So once we've gone through and checked those areas, we also want to have a look into the higher RPM areas in part throttle. |
| 01:36 | Remember we haven't really spenT much time looking at that area on the dyno. |
| 01:41 | And we don't need to have the accuracy under those sorts of operating conditions because we don't spend much time there but we do want to have a look and just make sure that we're in the ballpark, always want to err a little bit on the side of being richer rather than leaner in those areas. |
| 01:56 | Then we're going to move forward and we're going to have a look at our acceleration performance under wide open throttle conditions. |
| 02:03 | Here what we want to do is select a gear where we're going to be placing a reasonable amount of load on the engine. |
| 02:09 | Third or fourth works pretty well, particularly with turbocharged engines if we use a very low gear we're going to end up with insufficient load placed on the engine and our boost targets that we actually achieve aren't going to be very realistic. |
| 02:22 | Now under our wide open throttle ramp run conditions, what we're going to be primarily looking for is three things, first of all is our air/fuel ratio tracking our target? Secondly, is our engine knocking, have we got any detonation occurring? Now while I'm not going to be doing this for our worked example, normally I would expect to be using audio knock detection for this process, listening very carefully to our engine, making sure that no knock is present. |
| 02:47 | As long as our fuelling and our ignition timing line up OK we're also going to be looking at our boost and making sure that our boost is not exceeding our targets. |
| 02:56 | Now what I have done here, compared to the air/fuel ratio targets that we have on they dyno, I have actually decided to richen our air/fuel ratio through that mid range. |
| 03:06 | A little bit leaner than I'd probably want to run this engine and we've gone through and richened up the air/fuel ratio targets under our high boost areas, so where we're targeting around about that 240, 250 kPa low in the RPM range. |
| 03:22 | As the boost drops down higher in the RPM we can then taper a little bit leaner again. |
| 03:28 | Now that we know what we're going to be doing, it's also worth talking a little bit about how we're going to be making these changes. |
| 03:33 | There's a variety of options here. |
| 03:34 | Obviously it's not particularly safe or advisable to try and make tuning changes on the laptop while you're actually driving.. |
| 03:42 | You've got the option here though of using a friend to help you with the tuning. |
| 03:46 | This way one person could concentrate on the driving and the other person can use the laptop to make tuning changes. |
| 03:52 | The technique we're actually going to look at today is a bit of a hybrid here where we're going to be making use of the MegaLogViewer HD to analyse a log file from the Ecumaster EMU Black. |
| 04:04 | And the Ecumaster EMU Black has the ability to output a log file in a .csv format, meaning that it can be important straight into MegaLogViewer HD. |
| 04:13 | As we go through this step of the process we'll see exactly how I've set that up and how you can use that. |
| 04:20 | I find it's a great way to analyse a large chunk of data very quickly and again you don't need to be watching the laptop while you're performing your tuning. |
| 04:28 | Now let's head out onto the track and we'll start gathering some data. |
| 04:33 | The first thing we want to do after heading out onto the track, or for that matter if you're doing your tuning on the road, is to just allow the engine a minute or two running at normal road going speed in order to clear any heat soak that may be present. |
| 04:48 | Here we've just spent a few minutes sitting in the pits while I've been talking about the process we're going through and while I was doing that, the engine was idling. |
| 04:55 | What this can do is result in particular in an unrealistic air temperature. |
| 05:01 | And while of course there are corrections in the ECU that take that into account, what we want to do at least to start with is make sure that our air temperature as well as our coolant temperature are realistic for normal operating conditions. |
| 05:15 | So let's just continue and we'll get to the point where I'm happy with those parameters and we can start our logging. |
| 05:23 | Alright pretty comfortable with the temperatures now, we can see our coolant temperature sitting at 81° and our air temperature at 25 so we can actually start our logger and start gathering some data. |
| 05:35 | Straight away we can look at our measured lambda versus our target and this is going to give us a guide straight away of how close we are to target. |
| 05:42 | You can see while I've been talking here, that we're generally within about 1% plus or minus of our target which is about as good as we can expect. |
| 05:49 | We can also use the fuel logging graph that I've got set up there and we can see that we've got our target versus our measured lambda being graphed there as well. |
| 05:59 | So this is again giving us an instantaneous look at how close to target we are. |
| 06:04 | So what we want to do here is gather a reasonable amount of data and we want to vary both our throttle position as well as our RPM. |
| 06:12 | Spanning that range that I talked about, maybe 1800, 2000 RPM all the way up to 3500, 4000 RPM and all the way from very light load where we're barely touching the throttle, up to that transition just into boost. |
| 06:25 | So we're going to spend a couple of laps now gathering that data and we'll see what we get. |
| 07:28 | Alright I'm pretty happy with the amount of data that we've gathered there and we'lll just stop our log and we can head back to the pits. |
| 07:34 | Now of course while I have been gathering that data we can glance at the air/fuel ratio. |
| 07:40 | Particularly if you've got an air/fuel ratio meter on your dash and just make sure that you're on target. |
| 07:46 | Given the areas that we're looking at, very light load, almost closed throttle through to our transition onto boost, we're typically in those regions going to be targeting lambda one, that would be expected so we don't necessarily have to have a guide as to what our target is, we'll be able to see that on the laptop. |
| 08:03 | But straight away looking at the wideband air/fuel ratio meter, if you've got one on the dash, this will allow you to see how close you are to your target and if you've got any areas that desperately need some attention. |
| 08:15 | Alright we're back in the pits, let's have a look at our log file and we'll see what we can tell from that. |
| 08:21 | Alright what we're looking for here is our target lambda which we've got here in purple and our measured lambda which we've got here in green. |
| 08:30 | So this log file is obviously at this point quite long, you can see it's around about eight minutes in length and we've got a lot of data so we can scroll through this and get a bit of a snapshot on how we're looking. |
| 08:41 | You can see we've got one area here where we are a little bit lean, we can see that this is at 5500 RPM and 43 kPa. |
| 08:49 | Now using the graph log like this can pinpoint and allow us to see exactly where this problem is. |
| 08:56 | We can see by looking over on the volumetric efficiency table here, the blue cross hairs shows us where that point is. |
| 09:02 | And again we're looking at one of these areas where we couldn't really access on the dyno so not a huge surprise we've got some problems there. |
| 09:09 | But for the most part, if we just scroll back through this log file we can see that our air/fuel ratio is actually tracking pretty close to our target. |
| 09:17 | However you can see as well with a log file that's this large where we're transitioning through a large range of our VE table, it's quite difficult to really make a lot of sense of it. |
| 09:30 | And we may find that there's one small area that was momentarily lean or rich, it could be for a variety of different reasons and if we look at another section of a log file from that same area we might find that the air/fuel ratio actually was on point so can be a little bit misleading if we're just looking at one data point. |
| 09:47 | So what we're going to do here is we're going to export this as a csv file so we can do this using this little icon here, we'll click on that and we can give it a name, so we'll call this track test one. |
| 10:00 | Let's save that and now what we're going to do is load this up in MegaLogViewer HD. |
| 10:05 | Alright with our log file loaded up we can see the relevant information here, in particular down the bottom we've got our lambda versus our target. |
| 10:14 | And we can see at a glimpse there's a few areas that do need a bit of work. |
| 10:17 | We can see those here which is similar to what we looked at when we were scrolling through the graph. |
| 10:23 | However what we want to do is actually come over to our histogram and table generator. |
| 10:27 | Now we're going to be able to see this information in a slightly more useful format. |
| 10:31 | So what I've done here is I've set up a table that has the same axes or same break points as our volumetric efficiency table. |
| 10:39 | We've got manifold pressure on our horizontal axis, we've got RPM on our vertical axis and what we can do here is just click on this little icon and that will allow us to actually manually enter the break points that we want. |
| 10:52 | Done all of that there so we'll close that down and all we want to do essentially is come back to Ecumaster software and make sure that we are using exactly the same break points. |
| 11:03 | This is going to be important so that the data is relevant to us. |
| 11:06 | Let's head back to MegaLogViewer. |
| 11:08 | The other aspect there that's important to keep in mind is the digits or precision that we're using for the X and Y axes here because we're representing manifold pressure in kPa we have zero on both the X and the Y for manifold pressure and RPM, otherwise we're going to get some slightly strange looking break points. |
| 11:27 | The data that's coming into this here, this is the important aspect though. |
| 11:32 | And here the Z axis is defined, you can see it's called ECU M AFR. |
| 11:38 | Not particularly meaningful. |
| 11:39 | This is a user generated parameter that I've made and I'll show you how to do this. |
| 11:44 | If we come up to our calculated fields here and click on this, you can see we can set up some custom fields. |
| 11:50 | I've got a range here to suit the different ECUs I've tuned. |
| 11:53 | We'll come down to our ECU M AFR and we can see the formula here and we can see this is essentially our have over want formula. |
| 12:02 | We've got our measured lambda divided by our target lambda and then we're multiplying by 100 to represent it as a error percentage. |
| 12:10 | If you want to start from scratch though, you can come up here to add custom field, you can give it a name and then for the formula you press control and space and this will allow you to cycle through all of the available parameters that you've got in that particular graph log. |
| 12:27 | Obviously it's important that you are logging the parameters that you're going to want to display here. |
| 12:32 | So we've already gone ahead and done that of course and that's the parameter that we've got in here. |
| 12:37 | Now we can see some other information here as well, we can see that we've got a total of 16885 samples in this log file. |
| 12:45 | This is where it comes down to being quite important to make sure you've got a good amount of data in here, otherwise it can throw out our accuracy. |
| 12:53 | The colouring here is also important. |
| 12:56 | We can see the table colouring, colour is based on our hit weight which essentially just means that the more sample we've got for a particular cell, the brighter green we're going to be. |
| 13:06 | So essentially what this means is that the majority of this table here, we've got a good number of samples, we've got good solid green colouring. |
| 13:13 | These ones out here, the white parameters, these we've probably only got one or two hits so, or maybe a few hits so I wouldn't be necessarily trusting these pieces of data. |
| 13:24 | Obviously yellow is somewhere in between. |
| 13:26 | Now that we understand how the histogram's set up, let's look at what it's trying to tell us. |
| 13:30 | And we can see straight away that quite a large majority of this table is looking really good, we've got numbers in the vicinity of plus or minus about 1-2%. |
| 13:39 | Few outliers there but for the majority of this it's looking pretty good. |
| 13:43 | Remembering that when we've got a value of 0% this means we're right on our target. |
| 13:47 | And it's important to be a little bit realistic about what you're expecting here. |
| 13:50 | We're not going to be able to get a perfect table filled with zeros here. |
| 13:55 | If we have our air/fuel ratio within about maybe 1-2% of our target, even plus or minus 3%, I'd be pretty happy with that in general. |
| 14:03 | And that's essentially what we've got here but there are some areas that on face value seem a little bit concerning. |
| 14:10 | Particularly down here at light load, we can see we've got numbers from 43% through to as much as 89% and again on face value this seems a little bit of a problem. |
| 14:21 | This means that our air/fuel ratio is very very lean. |
| 14:24 | It's important to understand how to interpret this though because if we are completely off the throttle then the ECU will go into overrun fuel cut off. |
| 14:32 | This means that it essentially closes the injectors off and we end up with a very lean air/fuel ratio reading. |
| 14:38 | So we want to be able to filter those out and ignore them, any time essentially the throttle is closed. |
| 14:44 | So let's see how we can do that. |
| 14:47 | We've got a range of filters up here on the right hand side. |
| 14:49 | We can edit these filters and in particular here if we look at our throttle closed parameter, we can see that any time our throttle position is less than 10% it's going to ignore that error. |
| 15:01 | Now we may want to make that a little bit less than 10%, basically we want to just get it above the point where our throttle is completely closed. |
| 15:08 | If we bring that in, we can see that straight away that gets rid of those outliers down here and now we've got some slightly more sensible data. |
| 15:16 | The other aspect to consider here is that when we are gathering the data we want to be quite smooth on our throttle input, otherwise this will give us transient enrichment or transient enleanment for that matter which can affect our results. |
| 15:29 | So we can also bring in a transient enrichment filter here which we've got. |
| 15:34 | Not going to do this but basically it works in the same way, looking at quick changes in both RPM or throttle position. |
| 15:41 | So now what we can see is our data's looking pretty good. |
| 15:44 | We do have a few areas that probably could do with a little bit of work here and we'll see how we can correct these. |
| 15:51 | In particular let's start by looking at 3250 RPM, 50 and 60 kPa. |
| 15:56 | We'll see we've got 5.4 and 3.4%, not going to get too fussy, too granular with this right now. |
| 16:01 | But let's just transfer that trim across, so I'm just going to average that out and we'll call it 3% so let's go to our fuel table. |
| 16:10 | So we were 3250 RPM, I'll just find the right zone here, 50 and 60 kPa. |
| 16:16 | So when the trim is positive here, this means that we are lean. |
| 16:20 | This is what we actually need to apply to correct it. |
| 16:23 | So all we need to do here is 1.03 and our multiply symbol, press enter, that's going to correct that. |
| 16:30 | Let's come back to our log file and we'll look at one more area. |
| 16:33 | Now we can see here at higher RPM, and remember we didn't go this high in the RPM but we do have some quite large trims, particularly at light load, that area that we couldn't get to so let's see what we can do with that. |
| 16:45 | Let's say 5500 down to 4500, we've got 8-11%, I'm going to average that at 8% and we'll apply that so let's head back to Ecumaster now. |
| 16:58 | Just want to make this at 35 kPa, I'm also going to extrapolate this below and we were 5500 down to 4500 and what we're going to do here is make a change for a start of 1.08 so 8%, we'll multiply that out and make that change. |
| 17:13 | So essentially it's a rinse and repeat of this process and what we'll do is make the changes to our table as required from our logging, go back out and grab some more data and we might take two or three goes to get this really fine tuned. |
| 17:25 | But you can see the power of what we did with our steady state tuning on the dyno, everything already looked really really good, particularly through that area of the table that we spent the most part of our time. |
| 17:36 | Let's head back out onto the track now and we'll gather our second piece of data where we're looking at replicating our wide open throttle ramp run. |
| 17:45 | - Alright again out on track we've just spent a little bit of time getting rid of any heat soak. |
| 17:51 | Gone through a full lap of the track to do that and we're just going to come up to the longest section of straight here. |
| 17:57 | I want a piece of track where I can accelerate through to 7000 RPM in fourth gear. |
| 18:03 | So we're just coming through the last hairpin now, I want to start here with our RPM down around 1800 to 2000 RPM. |
| 18:12 | Essentially we want to really replicate exactly what we saw on the dyno and all things being equal we should be seeing exactly the same sort of results. |
| 18:21 | So we'll just come through our last corner onto the front straight here, making sure that our RPM is still sitting around about that 2000 RPM point. |
| 18:30 | Once we're on the straight, we can just smoothly go to full throttle and we'll accelerate all the way through to 7000 RPM. |
| 18:59 | Alright so with our acceleration run complete there, I was watching the boost during that run, quite happy with everything that happened on the boost gauge, let's head back to the pits and we'll have a look at our data and see what that tells us. |
| 19:14 | Alright looking at the data, everything actually looks pretty good there. |
| 19:17 | We can see that our peak boost there, sitting 245 kPa, essentially just about where we expected. |
| 19:25 | We can also see that our boost, we've dropped it down at higher RPM here, particularly just to stay within control of our injector duty cycle so we're sitting down around about 200 kPa right up at the 7000 RPM point. |
| 19:40 | The air/fuel ratio of course is what we're most interested in looking at for the moment though and we can see that everything's tracked reasonably well. |
| 19:48 | We can see there's a little lean spot in here, not too bothered by this as we ramp up onto boost, what I'm much more interested in is what our air/fuel ratio's looking like through the full power area, once we're actually at full boost. |
| 20:02 | You can see that we're tracking pretty well, we are a little bit richer than our target though, we can see around about 0.77, we're about 3% richer than our target through here, sort of from about 4500 RPM through to around about 5000 RPM. |
| 20:19 | You can see as I've tracked the air/fuel ratio target's a little richer than what we saw on the dyno here, and then of course that tapers leaner as our boost drops away. |
| 20:30 | Just again trying to keep our mixtures under control and safely rich while staying under our injector duty cycle limit as well. |
| 20:37 | So we can see that right through the top end here as well, we're around about 3% too rich. |
| 20:44 | So what we can do there is use our graph to help us pinpoint where abouts we are operating, we can see at the moment we are sitting 245 kPa, 4500 RPM and we drop down to around about 200 kPa, 7000 RPM, just at this point right here. |
| 21:03 | So let's highlight that area of the table. |
| 21:06 | 4500 RPM, we're going to come down to 200 kPa here and we'll come through to 7000 RPM. |
| 21:14 | I'm actually going to extend this out as well, all the way out to 8000 RPM. |
| 21:18 | So we're around about 3% too rich there. |
| 21:22 | I'm going to start by making a change of 2% though so what we'll do is enter 0.98 and hit the multiply symbol, press enter and that's made that change there. |
| 21:34 | Now that we've made that change it's a case of heading back out onto the track or out onto the road and testing again. |
| 21:42 | Again this is an iterative process but using our logging, it's very quick and easy to dial this in accurately. |
| 21:49 | Let's head back onto the track now and we'll make our last test which is around our transient response. |
| 21:56 | Alright we've started here by getting rid of our heat soak as normal and we want a section of track that's nice and straight and pretty long and what we want to do is test our transient response on a gearshift or alternatively as we move in and out of the throttle once we're past the RPM where we can reach full boost. |
| 22:15 | For this particular test, what I'm going to do is accelerate through third gear and once we reach full boost I'm going to shift into fourth. |
| 22:23 | I want to do a quick shift here and we want to also sense how smooth the car is, whether it picks up when we get back on the throttle nice and crisply which is what we want, let's go ahead and perform out test coming onto the front straight here. |
| 22:52 | Alright so the car actually felt really good there, nice and responsive but let's head back to the pits again and we'll have a look and see what our data tells us. |
| 23:03 | Let's start by having a look at our boost response here. |
| 23:06 | So essentially this is around about where we've gone into the throttle and we can see that our boost tracks our target pretty well here. |
| 23:14 | The part that we're most interested in though is what happens on our shift. |
| 23:18 | So we can see we get out of the throttle here and what we're looking at is to make sure that we don't have any overboost when we get back on the throttle. |
| 23:25 | We can see that's exactly the case here. |
| 23:27 | Our boost comes up, we reach full boost around about 4700 RPM there and we don't have any overboost, we're actually sitting right on our target so everything's working nicely there. |
| 23:38 | We'll move over to our fuel tab now and we can have a look at how our fuelling did during those transients. |
| 23:44 | So we can see our fuelling now through our wide open throttle full boost area tracking really nicely on our target after that small adjustment we made. |
| 23:53 | We can see that it also tracks nicely after our gear shift. |
| 23:56 | What we're looking at here is on our transient and we can see straight away here that as we get off the throttle, we move quite rich, we've gone down to 0.67 lambda so definitely too rich there and we can see the area, looking at our cross hairs on our VE table we are accessing around about this area here. |
| 24:15 | Now the problem here is almost certainly not to do with our VE table itself and instead this will be a transient enleanment issue. |
| 24:23 | So what we want to actually do is remove a little bit more fuel during the tip out or as we move out of the throttle. |
| 24:33 | Looking at what happens as we get back into the throttle, that happens at this point here, we can see the throttle increasing. |
| 24:39 | And we can see there's a small lean area here. |
| 24:43 | Now I actually didn't feel this, it felt nice and crisp, everything responded quite nicely. |
| 24:48 | We can see that it tracks down to our target as we've already discussed. |
| 24:51 | But we have got a slight lean area here. |
| 24:53 | When we are dealing with transient enrichment there response of the engine in terms of how it feels is actually more important than getting a perfect smooth line for our air/fuel ratio on our graph. |
| 25:05 | However, this is an area that particularly with that rich area on tip out, probably does require a little bit more work. |
| 25:13 | We're not going to dive into our transient enrichment here in this worked example but if you do want to learn more about the transient enrichment tuning on the Ecumaster, you'll be able to learn about that in our member's webinars, check the archive there. |
| 25:27 | So at this point we've got our tune complete, we're happy with the performance on the dyno, we're happy with the performance on the track and we've seen how we can use the datalogging both within the Ecumaster and with that csv file in MegaLogViewer to quickly dial in and optimise our tune. |
| 25:44 | If you've got any questions on this worked example, please ask them in the forum and I'll be happy to answer them there. |
