Road Tuning: Step 8 - Full Power Tuning
Watch This Course
$229 USD
-OR-
Or
8 easy payments of only $28.63
Instant access. Easy checkout. No fees.
Step 8 - Full Power Tuning
19.47
| 00:00 | - Once we've done the midrange part of the mapping, the next step is to do our full throttle tuning. |
| 00:05 | Now, I'm going to show you how I do this, relying really heavily on the data logging so that I don't actually have to be watching the laptop keyboard. |
| 00:13 | I can concentrate on actually driving the car. |
| 00:16 | I'll be glancing at the wideband AFR meter that's on the windscreen at the same time, but I can sort of see that out of my peripheral vision, so I can keep good control over the car. |
| 00:26 | Now before we do that, what we want to do is make two changes. |
| 00:29 | First of all on the fuel map I'll just add a little bit of additional fuel to the two rows of the map, so I'll multiply that by 1.03 to add three percent, so the aim here is just to be slightly richer ideally as we start tuning. |
| 00:44 | We'll swap to the ignition map and again I want to be safely conservative on the ignition timing, so we're going to retard the top two rows by two degrees. |
| 00:54 | So if everything works out well, basically we should find that we've got no detonation, we're safe on the ignition timing, and hopefully we're going to be just a little bit rich on the fuel side. |
| 01:06 | Now the engine is already mapped pretty well out to 5000 RPM, so we know that if we do a wide open throttle pull to 5000 RPM, we shouldn't have anything to worry about. |
| 01:16 | So I'll just press F8 on the keyboard and start data logging. |
| 01:20 | I like to do this sort of run in second gear, so I'll just get in second gear at about 2000 RPM and then just go to wide open throttle. |
| 01:33 | Now while I'm doing that, I'm watching the wideband meter and I just want to back off at around about 5000 RPM, which is where we've already tuned to. |
| 01:47 | So just pull over to the side of the road so we don't have to actually make these changes live, and we'll have a look at the data logging and see what that run looked like. |
| 01:57 | So if we highlight the area of the run, we've got our air/fuel ratio here and you can see that for the most part it's pretty good. |
| 02:06 | We've got a lean spot here, 0.927, so we're about three percent lean at 3500 RPM. |
| 02:14 | The rest of the run looks pretty good. |
| 02:16 | We're slightly richer than ideal there at about 4200, and then right at the end, I actually went out to 5500 RPM, so we got to an area we hadn't actually tuned, and you can see we're about six or seven percent lean. |
| 02:29 | Now one of the things to watch while you're doing this is you want to be watching if the wideband does start to drift lean like this, get off the throttle, and make some changes to your map. |
| 02:40 | So don't hope that the air/fuel ratio will come right. |
| 02:44 | It does risk damaging the engine. |
| 02:46 | So based on this what I'm going to do is make a few changes. |
| 02:49 | So first of all let's start at the start. |
| 02:52 | We've got a slightly rich spot at around about 2000 RPM, 2250. |
| 03:00 | We come back to being about right. |
| 03:03 | The first change I'm actually going to make is at that 3500 RPM spot. |
| 03:06 | So if we go back to our tuning, you can see that we've actually got a nice little cursor that shows us where in the map the ECU was accessing at the time. |
| 03:16 | So we'll just highlight, again we're going to make the changes to the top two rows. |
| 03:20 | We want to add three percent, so that's 1.03 at 3500 RPM We go back to our logging. |
| 03:29 | We have a look. |
| 03:30 | We were a little bit rich at 4000 through to about 4500. |
| 03:39 | Actually 4500 is pretty good, so at 4000 RPM, I'm just going to take two percent out. |
| 03:46 | So we multiply that by 0.98. |
| 03:49 | Then the rest of this part is looking really good up until 5500 RPM, where we can see we go to 0.97 lambda. |
| 04:01 | So we'll just make a change there at 5500. |
| 04:08 | We want to add seven percent, we'll add eight percent just to be safely rich, and what I'll do at the same time is I would add that same eight percent right through the top end. |
| 04:22 | So now what we can do is basically repeat our run and see how the changes affected our lambda. |
| 04:37 | So again we want to select second gear, get to around 2000 RPM, and go to wide open throttle. |
| 04:50 | So this time I went through to 6000 RPM. |
| 04:53 | Again we're going to just pull over and see how our air/fuel ratio worked out. |
| 05:03 | So if we look at our logging, and we can again highlight the area of the map we're accessing. |
| 05:09 | So you can see we've still got a slightly rich spot at 2300. |
| 05:14 | The rest of the map is looking really good. |
| 05:17 | Now when we're targeting 0.90 lambda, you don't get an exact number on the air/fuel ratio meter, and I would expect to see a difference of about plus or minus 0.01, 0.02 on our lambda, and I would be happy with that sort of a fluctuation. |
| 05:36 | So for the most part, that's exactly what we've got here. |
| 05:39 | You can see at 5250 RPM, we're 0.936, so we're a little lean there, but the point to note here is that by 5750 RPM, we're actually back exactly where we want to be and that continues. |
| 05:57 | Up to 6500 we're actually pretty good. |
| 06:00 | Now this is an area where we may want to start making changes to the axes, and there's two points I'm going to change, so at 5250 we really want to add some additional fuel, and if we look at 2250, where we're quite rich, we want to take some fuel out. |
| 06:19 | So we're going to make that change now to the fuel map just to show you how easy that is. |
| 06:24 | So I'm going to add 2250 to the RPM range and I'm going add 5250. |
| 06:35 | Okay, so this just lets us account for changes in the fueling due to quite big changes in volumetric efficiency over a narrow range. |
| 06:45 | Now we used the interpolate function there in the G4 Plus to just make sure that those rows that we added are already where they should be. |
| 06:55 | Now that we've added those rows, at 2250 we want to remove around about five or six percent fuel there, so we'll times that by 0.94. |
| 07:07 | That takes six percent out. |
| 07:09 | Again at 5250 we want to add four percent, so we'll do that. |
| 07:19 | Okay, so that should correct for those two areas, and we just want to try that again. |
| 07:23 | The rest of our run right up to 6500 RPM was pretty good. |
| 07:28 | Now we're actually targeting a little bit richer in the top end, so I'm going to add a little bit of fuel at 6500 RPM and above. |
| 07:38 | That's just another two percent there. |
| 07:40 | We'll store those changes, and we're ready to do another run and see how that responds. |
| 07:46 | (engine revving) Okay, so on that run we went all the way up to the rev limiter there, 7000 RPM, and by basically everything's looking really close to our target. |
| 08:13 | So we'll just have a look at the data logging and see how that's come out. |
| 08:18 | Again we'll just zoom in. |
| 08:20 | So what you can see is that change I made at 2250 still, we've got a slightly rich spot there, so we still need to make a bigger change to our fueling in that area. |
| 08:33 | The rest of our run though, is looking really, really good. |
| 08:37 | Now at 5250 where we added that zone, we still need to add a little bit of fuel as well, but you can see that the lambda now isn't quite as lean as it was, and we're pretty much matching our target at 7000 RPM. |
| 08:56 | Probably want to add one more percent. |
| 09:01 | This is how easy it is to use data logging to dial in your tune and get your air/fuel ratio really accurate and really quickly. |
| 09:10 | So once we've go the fueling side of things tuned properly, the next step is to go back and do the same thing with the ignition timing. |
| 09:20 | So for our first run, we're going to put the two degrees of ignition timing that we removed before we started doing the wide open throttle runs. |
| 09:28 | We're going to put that two degrees back into the top two rows of the map. |
| 09:32 | So we'll do that now, and then we're going to perform a full-power run in second gear and we're going to confirm whether or not that's resulted in any detonation. |
| 09:45 | So let's do that now. |
| 09:51 | (engine revving) Okay so audibly from that test, what I found is through most of the rev range, there was no detonation. |
| 10:24 | The engine was nice and quiet. |
| 10:25 | Right at the top of the run though, we did get some detonation that was apparent, and you should have been able to hear that in the audio. |
| 10:32 | Because it's quite hard to distinguish when you're learning, between normal engine noise and detonation, I've also set this particular ECU up with knock control through the factory ECU, and what we can see here in the data log file is I've got the knock levels from all four cylinders showing down here. |
| 10:53 | You can see we've got a pinky line, which is our knock threshold, so basically any time the knock level goes above this threshold, that's the engine detonating, and you can see right in the top end at 7000 RPM, we had knock from three of the cylinders. |
| 11:11 | Through the most of that run though, right up to about 6500 RPM, we had no knock. |
| 11:16 | I would go back now to our timing map and remove ignition timing from the area that we did have knock, so from 6500 RPM up, I would take two degrees out of the entire map. |
| 11:29 | Now because we didn't have any knock in the bottom end, what I would try doing, just to see if we are still a little way away from our knock threshold, is add some timing into that area, and what I'm going to do is add two degrees between 2000 RPM and 5500 RPM, and you can see I'm adding that into the entire table, not just those top two rows. |
| 11:55 | So now what we'll do is another run and see how the engine is responding. |
| 12:07 | (engine revving) Okay, so with that particular run, we still had no detonation through the lower RPM part of the run. |
| 12:45 | If we look at the data log file again, you can see that we've just had one little hint of detonation here on what is cylinder three. |
| 12:55 | Now because we have tested this already at that particular timing, at 7000 RPM, and we haven't had any problems with it, I would reconfirm that and make sure that it wasn't just a one-off event and make sure it wasn't due to heat soak. |
| 13:12 | The way we're testing the car for the purposes of this demonstration is not particularly normal. |
| 13:19 | We would normally let the car run and get up to a normal operating temperature with air flow through the engine bay before we did a run, and the heat soak we're likely to get testing the way we're doing is more prone to cause detonation. |
| 13:32 | So that shows no detonation through the midrange of the map and we can audibly also notice that the engine is still clean. |
| 13:41 | We've got no detonation. |
| 13:42 | Just to show how far we can go though, what I'm going to do is add another four degrees to the midrange part that we've just changed, and we'll do another run and see how that affects everything. |
| 13:56 | (engine revving) Now in that particular run, we are starting to get a little bit of noise occurring and if we look at the log file here of the detonation, you can see while the noise profile hasn't actually exceeded the knock threshold, we are getting right to that very limit, particularly on this number two cylinder here. |
| 14:44 | So what that would mean to me is that the engine is starting to get a little bit noisy and we're right on that knock threshold. |
| 14:52 | We're right at the limit, so I would go back and remove that timing. |
| 14:56 | Now I only did that just to show you what that causes as we start getting close to that knock threshold. |
| 15:05 | So by going through this procedure, particularly if we do have an engine management system that gives us knock control and knock monitoring like this, it does make it nice and easy. |
| 15:17 | By going through this system using either the knock protection in the ECU or our audio knock detection and slowly advancing the timing up until we just get that onset of detonation then removing a couple of degrees of timing to give us a safety margin, we can quite quickly build up a table of our ignition timing. |
| 15:35 | Now one thing I did notice in that run there at 6800 RPM, which is the same point we had in that last run, we do still have one cylinder that is exceeding the knock threshold, so what I would do there is pull out a further two degrees of timing in that top end and just safeguard the engine further and make sure we've got no detonation. |
| 15:59 | We'll now do one final run and just confirm that the detonation that we saw has been removed. |
| 16:07 | (engine revving) Okay, so we've just done our last run there, just to confirm everything, and you can see now all of our noise levels for knock are sitting below the threshold. |
| 16:36 | So we've got no detonation happening anywhere. |
| 16:39 | Basically this is what we're looking for. |
| 16:41 | We're looking for an engine where audibly you've got no detonation throughout the rev range and you can see that backed up visually if you've got a knock control system, which does make life a little bit easier. |
| 16:53 | This knock control system still needs to be calibrated though, so I'm still relying solely and predominantly on my ears and audibly be able to pick out detonation. |
| 17:03 | So once we've done this wide open throttle section of the ignition map, now what I would do would be add in a little bit of timing into the part throttle area through here where we added an additional two degrees in, and the engine took that quite happily. |
| 17:21 | What I'd then do is go back and just check that at steady state and just audibly make sure that I'm seeing or hearing any detonation at that part throttle, and that just reconfirms that the midrange part of our mapping is all okay. |
| 17:38 | So we're just going to quickly do that now. |
| 17:44 | (engine revving) So that's the basics of what we're doing when we're tuning the ignition table and the ECU. |
| 18:56 | As you can see, it is a little bit more time consuming and certainly a little bit more fiddly than doing it on a dyno where we have a nice clear, accurate feedback on what the engine torque is doing, and as I said right at the start, doing the ignition timing on the road rather than on a dyno straightaway means you are going to be making some compromises, and there's a better than average chance that your final result and torque is not quite going to be as good as what you'll get on a dyno. |
| 19:24 | If you approach it carefully though, take your time, and are very sensitive to listening for detonation, you're going to get a really good result that will probably come within a few percent of what you'd get on a dyno, and you're going to get a nice, safely-tuned engine that's not going to suffer from any detonation. |
