Practical Standalone Tuning: Step 10: Confirming the 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 the Tune on the Road/Track
16.19
| 00:00 | - Alright for the last step of our 10 step tuning process, we're here at Highlands Motorsport Park and we're going to essentially put the car through its paces out on the racetrack and ensure that everything we saw on the dyno stacks up under real world conditions. |
| 00:14 | And the reason of course we do this, as explained in the body of the course is because it's very hard even with a well developed dyno cell to accurately replicate the sort of temperatures and airflows we'll see at speed out on the racetrack or under real world conditions. |
| 00:29 | So it's quite possible that we can expect to see some variation in terms of our air/fuel ratios, maybe our boost pressure will not quite be the same as what we saw on the dyno and it's possible that we could even have knock events occur out on the road or track that we weren't seeing on the dyno. |
| 00:45 | Of course with the rotary engine we're going to know about that very quickly because the engine is most likely going to end up damaged. |
| 00:52 | We've got conservative timing in here, I don't expect to see any issue with that but if you're applying this tuning technique on the Adaptronic to a piston engine, it is very important to monitor for knock when you're out on the racetrack or doing your road tuning. |
| 01:06 | So what we're going to do is go through this in two stages and really we're replicating our dyno process here on the racetrack. |
| 01:12 | We're going to start by gathering some data under low load, low RPM steady state conditions so this is the cruise and transition areas of our map. |
| 01:22 | Ideally here I want to cover the areas between about 2000 RPM and maybe 3500 RPM and probably between about 30 kPa, maybe 100, 125 kPa. |
| 01:34 | These are the areas that we're going to be spending the majority of our time, particularly for a street car and we want to make sure that our air/fuel ratios stack up and we're seeing essentially what we did on the dyno. |
| 01:44 | From there, we're going to then go out and gather some data doing a wide open throttle acceleration pull and use here third gear so we're really loading the engine up and we're not risking wheel spin in those lower gears. And we're going to use datalogging so that we don't need to monitor that laptop while we're on the track and I'll show you some tricks as well that can really speed up the tuning process from a large dataset. |
| 02:09 | So let's get our engine up and running and we'll head out on the track now. |
| 02:13 | First thing we want to do when we head out on the racetrack or on the road for that matter, is make sure that we allow our operating conditions to reach normal temperatures. |
| 02:22 | So I'm talking here about our engine coolant temperature and our manifold temperature or intake air temperature. |
| 02:28 | So particularly there being in the pits while I was talking at the start of this module, our air temperature in the manifold was sitting at about 48° so that's a lot higher due to heat soak and that's going to give us some unrealistic operating conditions so generally a couple of minutes of driving at speed will allow those temperatures to come back down to more reasonable, more normal operating conditions and this is just going to mean that if some of our compensations are off, with very high manifold temperature that could be affecting the air/fuel ratio that we're seeing so we want to be under normal operating conditions while we're gathering this data so what we're going to do now is just continue for around about a lap here and let those temperatures stabilise. |
| 03:12 | Alright we've got rid of our heat soak, our operating conditions are normal now so this is the time when we want to start our logger so I've just started the logger now and what we're going to do now is continue to gather data. |
| 03:24 | What we're trying to do here is be as smooth as we can on our throttle inputs and we want to try and stay away from gearshifts where possible. |
| 03:32 | It's easier to do on the racetrack than it is if we are on the road, sometimes that's going to be unavoidable but we'll see how we can deal with that as we move through this module. |
| 03:42 | And what we want to do is basically be as smooth as we can on the throttle so we avoid transient enrichment events or enleanment events for that matter and we want to cover as much of that range that I talked about as we can. |
| 03:54 | 2000 to 3500, 4000 RPM, 20 kPa, 30 kPa up to about 125 and we want to go through as many of those cells as many times as we can so we get good quality data. |
| 04:06 | The more data points we've got the better the quality of that data. |
| 04:10 | Now to do this we can use some of our road tuning techniques from the course, left foot braking here to apply a little bit of load, that's going to allow us to basically hold the engine with a little bit more load than normal without the engine RPM increasing so basically want to gather a couple of laps of data in this way. |
| 04:28 | Once we've got that, we can stop our data logger, and we'll head back to the pits and have a look and see what that's showing us so let's carry on now. |
| 04:54 | Alright we're back in the pits, I've downloaded the data from that track session and in order to see what's going on here in a little bit more detail, rather than using Adaptronic's built in log viewer, I've actually converted the log file to a .CSV, allowing me to then analyse it in MegaLogViewer HD. |
| 05:11 | Let's have a look at what we've got here. |
| 05:13 | And if we're looking at our log viewer, this is very much like what it will look like in the Adaptronic log viewer. |
| 05:19 | It's a little bit hard to really know exactly what to make from this. |
| 05:23 | We've got three pieces of data being shown, or three graphs being shown. |
| 05:26 | In the top graph we've got our RPM, in the middle graph here we've got our inlet manifold pressure and our throttle position, throttle position being the yellow trace and the bottom graph here we've got our lambda and our target lambda. |
| 05:40 | So really what we want to do is look at our lambda versus our target lambda. |
| 05:44 | The green plot here is our measured lambda, yellow is our target and we can actually see everything is looking pretty good, it's doing a reasonably good job here of tracking. |
| 05:54 | Important to understand here, we're always going to see some error in our air/fuel ratio, it's never going to perfectly overlay with our target so if I'm within about +/- 1-2% I'm pretty happy with that. |
| 06:06 | We will also be enabling closed loop fuel control once the tune is complete but we want to get our VE table as accurate as we can so that we've got as little work to do with our closed loop as possible. |
| 06:18 | So a few areas do pop up. |
| 06:19 | You can see that in this area here we are a little bit leaner than our target. |
| 06:24 | We can click on that and we can see exactly what's happening there. |
| 06:27 | We have a measured air/fuel ratio in this case the green one, 0.993, let's call it 0.99, target 0.97. |
| 06:36 | Still within that 2% but it does pop out as a area that's a discrepancy. |
| 06:40 | Likewise we've got an area here where we are a little bit rich. |
| 06:43 | You can see we're 0.83 vs 0.88 so about 5% richer than our target. |
| 06:48 | Everything else is looking pretty good, we've got the odd slight right area here. |
| 06:52 | We've got this big rich area here but we need to understand what's happened there, that was actually a gearshift and we can see that the yellow trace up here, the throttle is completely closed. |
| 07:03 | So how can we actually make use of this data because going through and making specific cell changes off this would be very difficult. |
| 07:10 | What we're going to do is move over to our histogram and table generator tab and we'll look at that instead. |
| 07:17 | So the histogram basically allows us to log a large sample set of data and put it into individual bins. |
| 07:25 | This means it can average the data it's collecting for each of those bins. |
| 07:28 | What I've done here is I've set up a table that's got the same break points as our VE table. |
| 07:34 | Vertical axis here, inlet manifold pressure, horizontal axis, RPM. |
| 07:38 | There is one little caveat here, just due to the way MegaLogViewer is set up here, our manifold pressure goes form low to high as we move from the bottom up to the top and if we go back to our Eugene software, we can see our VE table actually goes the other way. |
| 07:55 | So little bit confusing but as long as we're aware of this it's really not an issue. |
| 07:59 | So let's head back to MegaLogViewer HD and see what this is telling us. |
| 08:04 | So what we've got here is our X axis and Y axis setup, no problem there. |
| 08:09 | Then we've got the Z axis so this is the actual value that is being logged into the table. |
| 08:15 | You can see I've got this called Adaptronic lambda error. |
| 08:18 | And essentially this is just a user math channel, we can do this by going to our calculated fields, we'll go down to our custom fields, let's have a look at how I've set this up, so we'll click on edit. |
| 08:28 | So we can see the formula that we are going to be using here and essentially we're doing our measured air/fuel ratio divided by our target air/fuel ratio and we're representing this as a percentage error or correction so that's all we're doing there. |
| 08:43 | You can take note of that formula if you want to make that yourself. |
| 08:47 | So that's being logged into this histogram here and what we're looking at here, numbers that are positive mean that we are actually lean and we need to add a little bit of fuel so we do have a few of those here but you can see again they're all pretty close to zero. |
| 09:05 | Numbers that are negative mean that we're too rich and we need to remove a little bit of fuel and again we've got a few numbers here that are just a little bit richer than target. |
| 09:14 | We do have this area down here at 20 kPa though which is looking pretty ugly. |
| 09:18 | We can see we've got values of -16 to -19. |
| 09:22 | We need to be a little bit mindful of what notice we take of these. |
| 09:26 | So this is where using some filters can come in handy. |
| 09:29 | Got data filters up here on the top right hand corner and one of the ones we'll just use, just for simplicity here is our throttle closed. |
| 09:38 | So I've set this up basically any time the throttle's below 5% it's going to basically stop logging, stop logging samples so we can see straight away that cleans everything up quite nicely. |
| 09:49 | We do have still a couple of outliers, we've got -9, -3.7%. |
| 09:54 | Now we also need to understand what we can take from these numbers based on the colours and this is based on the number of hits we've had in an individual cell. |
| 10:04 | So over here on the left hand side we can see that we've got a total of 2800 data samples and we've filtered out 250 of those. |
| 10:13 | If we hover over one of these cells we're going to see that it's going to tell us exactly how many samples were taken in that cell, in this case 779 inside of that particular cell that I'm hovering over. |
| 10:25 | If we go over this one here, we can see that we only got 24 hits so the white cells we don't really have very many data points in them so I wouldn't take too much notice of them, we want to use the bright green coloured cells, that's some good quality data. |
| 10:38 | So now, all we're going to do is basically take individual note of the numbers that we've got in these cells and then we can go over, back into our Eugene software, back into the same cells on our VE table and we can make that same correction, we do this remember, we choose a cell here and we want to make a percentage change, maybe in that cell we were a percent too rich, press the P key and then -1 that will make that change, that'll remove 1% fuel so we can very quickly fine tune our VE table based on that histogram data. |
| 11:10 | Now it may take a couple of iterations to get it absolutely perfect or as close as we can but it is a very very fast way of gathering a lot of data, getting good quality data and being able to fine tune our table very accurately. |
| 11:23 | So once we've got our VE table dialled in and we're within that 1-2% which as you can see, we already were, we're going to go out now, we're going to gather one more piece of data which will be our wide open throttle ramp run, going to do that in third gear and we're going to essentially go through the same process, getting rid of our heat soak, we use the longest section of track here so we don't need to worry about corners or anything like that, just going to do a wide open throttle acceleration run and log that and we'll come back to the pits and have a look at it so let's head out on track and get that data now. |
| 12:03 | Alright we're back in the pits, we've got our data from that wide open throttle acceleration pull and what we've got is gone through the same process here, we've exported that as a .CSV, we could actually analyse this in the Adaptronic log viewer but for simplicity I like to stick to the same package and we've got it loaded up in MegaLogViewer so let's have a look. |
| 12:24 | This time we're going to be using our log viewer because we really only are interested in the sample through that acceleration run. |
| 12:31 | So we've got the same parameters we've been looking at, we've got RPM, the middle trace, our inlet manifold pressure versus TPS and then down the bottom of course our lambda versus our target. |
| 12:42 | So the first thing we want to really look at here is was our boost under control? So our green line here, we can see that green line coming up and we're hitting full boost there at about 4000 RPM. |
| 12:52 | We do have a little area here where we could possible add a little bit of duty cycle and bring our boost up further but important thing here is that our peak boost is sitting at about 197 kPa so we haven't gone over our target, we're basically seeing very similar boost to what we saw on the dyno so nothing to worry about here which is excellent, that's exactly what we wanted. |
| 13:13 | Now we're going to analyse our air/fuel ratio so let's have a look at our plot here. |
| 13:17 | Straight away we can see we've got this area here where we've got quite a big discrepancy, we are significantly leaner than our target. |
| 13:25 | Now we do need to be mindful of how to analyse that data because we can see that also coincides with the area where I was accelerating or going through to full throttle so this doesn't necessarily mean we've got a problem with our fuel table, our VE table, this is actually an indication that we need some work on our transient enrichment. |
| 13:43 | We're not going to cover transient enrichment within the worked example, we'll cover that in a member's webinar. |
| 13:48 | So we can ignore that, basically from the time that the transient event sorts itself out, we can see the area where we are ramping up onto boost and our air/fuel ratio target's dropping richer, we're pretty good, we're always within that 1%, we've got a tiny lean spot here but not really too worried about that. |
| 14:05 | We can see once we actually get up to the full boost area, from 4000 RPM, we can see that initially we are a little bit richer than our target. |
| 14:14 | That's not necessarily a bad thing, as we've said through the worked example, always safest to start rich and move lean but this is definitely richer than what we saw on the dyno so this is why we do check the tune on the road. |
| 14:27 | You can see at this point, we're about 0.74 lambda on a 0.77 target so around about 3% richer than our target. |
| 14:34 | Right up at the top of the run there, 6800, 7000 RPM we can see that we get a little bit closer, we're within about 1% up there. |
| 14:42 | But this does give us a little bit of work to do. |
| 14:45 | So what we're going to do here is just take note of where these changes need to be made. |
| 14:50 | In this case we can see that basically from about 4000 RPM up to around 6000, 6500 RPM, I've be taking about 2% out of that table and we know that the area that we're in there is from around about 180 kPa up to 200 kPa, I'd actually extend that out so what we can do is come into our VE table here and from 4000 RPM and 175 kPa we'll go out to, in this case what we could do is be a little bit more precise, we know that we're about 3% rich to start with so could make that change there between 4000 and 5000 RPM so P and -3% there. |
| 15:30 | And then 5500 and 6000 RPM, actually out to 6500 RPM, P we'll take -2 out of that area. |
| 15:39 | So the next step here would be to head back out on track, gather some more data and see if those changes have got us the result that we're looking for. |
| 15:48 | Like everything, this is an iterative process and it may take a couple of goes to get our air/fuel ratios where we want them to be. |
| 15:55 | Once we've got that tracking nicely, our tune is complete, we've got our air/fuel ratio under control, we've got a car that drives nicely and is nice and smooth and our boost is under control. |
| 16:07 | So at this point our worked example is complete. |
| 16:10 | If you do have any questions on any of this, please ask those in the forum and I'll be happy to answer them there. |
