Practical Reflash Tuning: Step 4: MAF/Injector Scaling

Back to course

Watch This Course

$229 USD $114.50 USD

-OR-

Or 8 easy payments of only $14.31 USD Instant access. Easy checkout. No fees. Learn more

Enroll Now
8 easy payments of only $14.31 USD

Course Access for Life

60 day money back guarantee

Step 4: MAF/Injector Scaling

25.35

00:00	The next step of our process is to calibrate our mass air flow sensor and or our fuel injectors and fuel system.
00:06	Obviously, again we know that this is a stock engine, so not, well shouldn't be really any work here to do, as I've mentioned I have basically applied a bit of a fudge factor to our MAF scaling just, so we can demonstrate the process anyway.
00:19	As far as the fuel system goes, because this is a lot simpler, I've already covered this off, but remember we can't just go and choose any parts and fit them.
00:26	If we come to our modifications window, remember you can choose your fuel system, fuel pump and injectors from the modifications window, tick the boxes to suit the parts you've chosen and that will do all of the background characterisation information to get you up and running.
00:44	OK, so what we're going to be doing is using the feedback from our short term and our long term fuel trims which we can see here active while we've got the car sitting at idle.
00:53	Despite me actually making a change to that mass sensor calibration, this is all looking pretty good.
00:58	The key parameter that we're actually going to look at though is this one here, T trim or total trim, simply the combination of the short term and long term fuel trims.
01:07	What we're aiming to do here is to get our total trim within plus or minus 5% of zero.
01:13	Generally, if I can get a little closer to that, I will, 2-3% would be great, but don't go chasing a tail, this is going to vary from day to day a little bit, it is why we have a closed loop there, so we're going to leverage that and utilise it.
01:25	While we're focusing on our display at the bottom here, we can also see we've got our commanded air fuel ratio and our measured air fuel ratio, obviously those are sitting pretty much one on top of the other, remember again we're running closed loop control all of the time with this particular ECU.
01:39	Now, we can bring up our graph display here and start data logging and if we come to our closed loop set of graphs here, we've already got basically everything we need here, we've got our air fuel ratio, our short term, our long term and our total trim values being logged here.
01:55	So, this will populate as we go, however it is difficult to analyse a huge amount of data using this technique and do anything meaningful with it.
02:03	So, instead what we're going to be doing is going to our window menu here and going to advanced graphs and XY graph.
02:09	And what I've done here is I've set this up with the airflow metre frequency on the horizontal axis and our total trim on the vertical axis.
02:17	And this is just going to basically plot data and give us a sense of what the average of our trim is doing over time.
02:26	We'll see this populate as we get into this particular module.
02:30	So, it's going to allow us to gather a lot of data and we're also going to be able to do a better job of deciphering the trends throughout that data.
02:38	So, the idea here is to, in both steady state and ramp run conditions, we're going to log as much data as we can across the range of frequency of our mass airflow sensor, looking for any trends and then applying that.
02:50	So, before we do that though, we still haven't actually flashed our map into the ECU, let's just close that down for a moment.
02:57	What we'll do is we will stop our data logger, just going to shut the engine down here.
03:03	Need to go back to key it onto the run position.
03:05	And all we need to do here is just select our little upload tool, just make sure we actually, are online here.
03:14	And we click that and then the upload process or the flashing process will complete.
03:19	This is quite a time consuming process on this vehicle, it can take between two and about six or seven minutes.
03:24	So, fortunately we can make a lot of our tuning changes live as we've already discussed.
03:29	But at this stage we do need to at least initially flash this map into the ECU.
03:34	Now, once the write has been completed we will see this upload finish box pops up asking us to switch the ignition off and of course we're going to do exactly as we're told.
03:43	And once that's reset, we are then free to restart the car.
03:47	One thing we will need to do before we can run the car on the dyno is essentially disable the stability control system and similarly turning the traction control off on its own is unfortunately not enough, still won't run on the dyno.
03:59	So, there's a couple of options, there is a pedal dance, but on the Type R, it's much easier just to put it into R mode and then if we hold down our traction control disable button, for around about five or six seconds we'll hear a beep which is just done, double beep that confirms that our traction control or stability control is now, completely disabled and we can now drive the car on the dyno.
04:20	Alright, so now that we've got our new file into the ECU, let's just go live with it and we'll just again check our trims.
04:29	I mentioned earlier about the fact that our trims were zero, but of course I hadn't actually, sent the file through.
04:35	So, having said that, this is all looking pretty good at the moment, we'll see how this pans out though, hopefully there is a little bit of work for us to do.
04:43	Should have mentioned as well before we perform that reflash, always advisable just to make sure that the battery health of the car is sufficient, haven't got our headlights on full beam or anything like that, our aircon is off, our laptop is connected to mains power, so that we're not going to end up with something going flat mid flash that could end up bricking the ECU.
05:04	Alright, so now that we've got ourselves ready to run, let's get ourselves up and running on the dyno, we're going to use fourth gear here and we're going to initially just let the engine temperatures stabilise to get rid of some heat soak before we start actually logging some data.
05:20	So, let's get up and running now.
05:26	Alright, we're up and running and now straight away you can see our total trims actually, kicked into gear.
05:31	What I didn't notice before is we actually weren't in closed loop mode and our air fuel ratio measured was sitting at a constant 14.69.
05:40	We can actually see our fuel status, whether we're in closed loop mode or not, with our fuel status value here, when we are in closed loop mode, this should be reading two and it was reading one.
05:51	So, not in closed loop, so that makes a little bit more sense and we can see that we do have a bit of error in our total trim, so sitting around about minus 11 or minus 12%.
05:59	So, what we're going to do now is just essentially use a combination of our foot pedal position and the RPM set point on the dyno, we're going to gather as much data as we can under steady state conditions and then we'll analyse that and see what general trends we can make out for our mass airflow sensor calibration.
06:18	So, let's go ahead now and we'll just bring our engine RPM up a little bit, we're at 2000 at the moment, 2500, so I'm basically going anywhere from basically as low in the load as I can to keep the dyno spinning all the way through to two thirds, three quarters throttle.
06:36	Just nice and smooth movements on our throttle, we don't want to induce any acceleration enrichment that's going to affect our results.
06:45	Again we'll just bring our RPM up a little bit more.
06:48	So, again nice and smooth on the throttle.
06:51	Also wanting to monitor our engine coolant temperature and our intake air temperature, just making sure that we are staying in reasonable bounds.
06:58	So, we'll do this up to about 4000 RPM.
07:04	Trim's still staying quite negative there.
07:07	Nothing necessarily wrong with them, their closed lip control system does such a good job that we could probably just leave this, but of course we aren't going to be taking shortcuts on this calibration.
07:19	Up to 4000 RPM now, just back out of the throttle slowly.
07:29	Alright, let's come back to idle.
07:32	We'll stop our logger and we'll have a look at that, the results that we've just got.
07:37	Alright, so looking in our graphs to start with, if we just click through here, we can see that our trim is always negative, it does move around a little bit.
07:43	Best of about minus 6% there, worst of maybe down here.
07:48	Maybe minus 15% but again it's difficult to really sort of take too much notice of this from the graph chart logger.
07:56	Now, I did originally say that I've set this up to be logging airflow meter frequency on the X axis and we can see we've got no data at all.
08:04	That's because I've still got a bit of a hangover from tuning the last car I've done which was the 2017 Civic Si.
08:12	Very similar operating system, but in this case we could log frequency.
08:17	We can't here, but that's OK.
08:19	We can just use AFM which is the output from that airflow meter table, it's not ideal, but this is essentially an analogue for frequency given that the frequency defines the mass airflow in grams per second.
08:33	So, looking at this, we can now see how the XY graph looks or logs.
08:38	We can see we've got little X points for all of the frequency data that we have or the mass airflow data that we have and we've got this blue trend line that's been drawn in.
08:48	We can see areas where we really haven't got too much data, areas where we obviously, gathered a lot more data.
08:54	The more data we get, obviously the more accurate we can be with this table.
08:58	I'd say at a start, probably a really good place to get started with making changes would simply be to pull let's say maybe 8% or 10% fuel or airflow, I should say mass airflow out of the entire table.
09:12	I'd probably start on the more low side of that, remember we can make these changes in real time as well.
09:20	So, if we come up to our calibration, we'll go to our sensors and our airflow meter and we'll just click here, it'll highlight the entire table, then Control J and let's start by pulling, yeah let's pull 8% out.
09:33	Now, we can see in real time we were at about minus 18%, we can see that's dropped to 10% so it's basically given us some confirmation that we are actually making changes here.
09:46	Let's get ourselves up and running, we'll start our datalogger again and we'll gather some more data and have another look and see if we've got an improvement.
09:53	Alright, so we're back up and running here.
09:55	I've got a little bit of heat soak with our intake air temperature sitting quite high to get rid of, but already we can see that our total trim's much much closer to the mark.
10:03	Actually, as I'm torquing, it's actually cranking up to minus 16% so it's making a complete liar out of me, but we will let that stabilise and we'll just bring our throttle in a little bit and now we can see we're sitting a bit closer to the mark, minus 2, minus 3%.
10:17	So, what I'll do again, I'll just allow this heat soak to dissipate, I'll actually turn my logger off, give it a second for our air temperature to drop down a bit more and then we can start our logger.
10:28	Always want to basically make sure we've got rid of any heat soak before we start trying to gather data.
10:34	Let's start that now anyway, we'll go through this process.
10:36	So, again just a case of stepping up our throttle opening here nice and smoothly as well as our engine RPM, gathering data over as much of this table as we can.
10:46	Remember at this point we're only obviously doing this in steady state conditions, we will be looking at some ramp runs as well to really fill that table in.
10:54	So, I can also monitor the airflow meter output here in grams per second, we're just coming up to 40, 42 grams per second.
11:00	So, again just nice and smooth, basically covering as much of this table as I can.
11:04	So, we'll jump ahead here and have a look at it once we've gathered the rest of our data.
11:08	Alright, we've got our data collected under steady state conditions there and we got up to I think just under 200 grams per second airflow.
11:16	So, by watching the airflow we can do a pretty good job of making sure we're covering as much of the table as possible.
11:22	Let's have a look at our XY graph now and get a sense of what we're dealing with.
11:26	So, what we can see is much much better at higher airflow, so from around about 14 grams per second up, we're probably sitting around about minus five which isn't too bad.
11:37	However, we can see right down in the bottom end here, we've definitely got some work to do, so this is sitting down at around zero to five grams per second.
11:47	I'd say we need about 10% out and then the rest of the table I'm going to maybe take 3% out, so let's go ahead and do that now.
11:54	Alright, so back into our mass airflow sensor calibration, we can see we've got all these negative values here, obviously we're not operating in that region, so what I'm going to do is just highlight everything out here until we get out to round about that sort of five grams per second that we wanted to make a change to.
12:11	OK, so we're up to five grams per second there.
12:13	Control J and let's just start, we can watch this in real time.
12:19	So, let's take out, let's say minus 10 at the moment, we'll commit that change.
12:23	We wanna be careful when making changes like this because we don't want to introduce a big step into the table.
12:30	So, obviously we've got an improvement there, we're sitting at minus seven total, it will take a little time for that long term trim to update.
12:38	So, Control J and let's take out another 4% there at idle.
12:43	OK, we'll just let that change take effect now and we're gonna come out to our next zone here which is 2840 hertz and what I'm gonna do is just basically smooth some of these changes here.
12:57	So, we'll come all the way through to the end of the table first.
13:01	I said I wanted to take maybe 2% or 3% out, so Control J here, let's take out 3% to start with there.
13:08	So, that will fix that up, hopefully for the most part of our table, but we do need to do some of this smoothing around here.
13:16	Let's just have a look and see how this looks graphically.
13:19	We can't really get the detail to really get a good sense of how that's looking but what we wanna do is just make sure that we don't end up introducing a big step.
13:27	So, this is the point that we made that change, 2790 hertz, we can see we've got that at 4.39 grams per second, we've got 4.1, 4.4 and then 5.14.
13:38	So, what I'm going to do is just highlight out here to let's say around about 10 grams per second, Control J and we'll take another 3% out of there and then we can also just Control J, another 2% out of there, so just a little bit of hand smoothing there.
13:54	Alright, we'll close that down and because the change is live, we can see the effect of that straight away, so what we're going to do now is get ourselves back up and running and gather some more data, see how things are looking.
14:06	Alright, we've got ourselves up and running here at about 1700 RPM and what I'm going to do is just use a combination of throttle and RPM here to basically cover as much of the mass air flow sensor frequency as we possibly can.
14:20	At the moment looking at everything, throttle trim's actually looking pretty good, minus two, minus 3%, let's just bring our RPM up and I've realised as part of this as well is this particular FL5, I can't actually log the mass air flow sensor frequency which is a shame, we'll see how we can deal with that in a moment.
14:41	But while I am talking here we can see that we've got our air flow meter output in grams per second, so we can keep an eye on this and this will give us a pretty good indication of where we are in the mass air flow sensor curve.
14:53	We want to cover as much of this as we can, we'll try and get out to about maybe 180, maybe 200 grams per second under steady state conditions.
15:00	Let's continue going ahead and gathering some data and watching our total trim value here while we're doing so.
15:06	OK, now we're at around 50 grams per second, we can see that the total trim's stepped up a little bit.
15:11	Not a bad place to be given that we are naturally a little bit rich.
15:15	So, it's pulling out 10% fuel here.
15:18	And we'll come up to around 70 grams per second here.
15:21	Just want to be nice and smooth on the throttle while I'm doing this to avoid any acceleration enrichment or anything affecting our results.
15:30	Come up to 3500 RPM, we're at 80, 90, we'll get up to 100 grams per second here.
15:35	Minus 4% trim, still pretty good shape really.
15:39	Let's continue coming up here to 4000 RPM.
15:42	Now, we're at 115, 120 grams per second.
15:46	And we'll just add a little bit more throttle, up to 150.
15:49	Minus 7% to 8% trim there, just trying to allow enough time at each of these points for everything to sort of reach equilibrium.
15:58	Up to 190 grams per second and we'll just come back to an idle now.
16:04	And clutch in and stop recording our data.
16:08	So, looking through this graph here, we can see that our trim's not looking too bad but we do have some sort of outliers here, minus 9% there.
16:18	Once we got higher into the load here, minus 9% as well.
16:21	So, definitely a little bit of work to do, but let's have a look at our X Y graph here and see what we can learn from this.
16:28	Now, I did mention that I set this up with airflow metre frequency, but as I've just said, we don't actually have that PID to log here, but we can set this up with airflow metre output which is our grams per second.
16:41	This is really an analogue for what we're looking at anyway.
16:44	And straight away you can see our trend line is a little bit all over the place but for the most part we are within our bounds of plus or minus about 5%.
16:54	We're a little bit too rich here and we're a little bit too rich here.
16:58	So, what I'll do just to start with, and while I'm talking here we can see that at idle our trim's sitting at minus 21% at the moment.
17:06	We do need to be a little bit mindful of chasing these trims at idle in particular.
17:11	We can see that right now, our intake air temperature's sitting at 62°C.
17:15	So, we're very prone to heat soak at idle, more, so that our other parts of the operation, that can affect our total trim, our closed loop trims.
17:25	So, we can end up basically making decisions based on bad data.
17:29	So, for the moment, what I'm going to do is just make an across the board change here, you can see while I've been talking, that has settled down a little bit but still quite negative.
17:38	Let's just start by taking let's say 4% out of the entire map.
17:43	So, we'll come to our calibration here.
17:46	And remember this is a live change, so we can do this reasonably quickly.
17:49	Airflow metre, we'll click here, control J and let's take 4% out.
17:55	OK, so straight away we can see when I click that, our total trim at idle, still very negative, but that has shown that that change was actually effective, it's reduced that trim.
18:06	We'll get ourselves back up and running and gather some more data now.
18:09	Alright, we're back up and running here, we are going to be chasing our tails a little bit here with this high intake air temperature.
18:15	The ambient temperature at the moment's around about 30 or 32°C, so we're sort of fighting a bit of a losing battle with our heat soak today, but we'll see what we can do with this.
18:26	So, that's probably about as low as we're going to get in our intake air temperature so, let's just start logging here.
18:31	And again we'll just try and cover off as much of the mass airflow sensor frequency range as possible, we're starting at about 17, 18 Hz here.
18:40	Sorry 17 to 18 grams per second, step this up, come through 26, 25, 26.
18:47	So, at the moment, minus 6 to minus 7% trim, come up to a little bit higher in the RPM, up to 30 grams per second here.
18:58	Still sitting at minus 8 to 9% so definitely quite a negative trim here.
19:04	Our intake air temperature now has come back to about 46°C, still up there, but a lot better than what we had.
19:10	Up to now 55, 60 grams per second.
19:15	Pretty consistently around about that minus 9, minus 10% despite that change that we just made, but not everything always works exactly how you'd like it the first time around.
19:25	Up to 90 grams per second here.
19:34	And we'll come up to 160 and call it quits there while our temperature's coming up.
19:40	Now, let's have a look at our results.
19:42	OK, just to show the variability there, we're back at idle now and we can see that our trim started actually around about zero, it's just starting to wind down now.
19:51	Looking at our graph though, we do have some trends here, essentially looks like from around about 68, 70 grams per second, we want to pull out maybe 6% to 8% and the rest of the table could probably do with about 5% being pulled out of it, maybe 6%.
20:06	Again just to keep things nice and simple for the moment, let's make one more change and what we'll do is highlight the entire table again, control J and let's take minus 5% out and we'll get ourselves up and running again and gather some more data and see if that's moved us in the right direction.
20:22	Alright, after that round of modifications we can see everything's looking pretty good.
20:27	We've got some outliers here which I'm absolutely not going to worry about for the moment.
20:32	We can see, even while I'm talking, our total trim at idle's pretty good at minus 4%.
20:36	We can see that we've got a slightly positive trend, but we're talking plus or minus a couple of percent.
20:43	Up to maybe about 120 grams per second, maybe 110 grams per second.
20:49	And then as we go past about 150 grams per second we start going very slightly negative.
20:54	I mean this would be fine just as it is, so I don't necessarily see that we need to make another round of modifications right now.
21:02	But if we were to do that, as I said probably from, I would say actually maybe 90 grams per second if I'm just sort of filling in the hollows here, probably from about this point here down.
21:15	So, 95 grams per second, let's just bring up our math calibration again.
21:22	So, what we're going to do is highlight the entire table here up to 90 grams per second.
21:35	And let's just scroll out until we get to 90.
21:49	I'll make that change out to 84 grams per second, so Control J, let's just add in, let's add 2% to that area.
21:56	We also need to see what's going on at wide open throttle as well.
21:59	So, let's perform a ramp run now and that'll allow us to gather data at the higher end of the air flow meter range.
22:23	Alright, so let's initially look at our power there and we've got 311 horsepower at the wheels.
22:30	This is going to give us a big increase, a reasonable increase in mid range, but we've actually, rolled over at the top there, so there'll be something to investigate later on but at the moment we're not chasing power, we're just calibrating our math.
22:42	So, let's have a look at our graph logger and if we click through here, this is actually not a bad way of viewing this data when we're doing a ramp run.
22:50	We can see our trim is pretty good.
22:54	Minus 2% to 3% and the top end we're sitting at zero, so actually really good.
22:59	Let's have a look at that on our XY graph though.
23:03	And again yeah looking all pretty much where we'd want it.
23:08	Going all the way out to 180 grams per second there, but right out the top end, slightly negative.
23:15	I'm not going to be too concerned about that.
23:17	If anything what I'd like is a little bit of a negative trim, it means that we're naturally, a little bit rich and the closed loop system's just pulling fuel out of that.
23:25	Alright, so at this point I don't really see the need to make any further adjustments to our mass air flow sensor calibration.
23:31	It is an iterative process, you're not going to get it done in one go, but between two and three, maybe four attempts you should have this dialed in pretty well within that plus or minus 5% bounds.
23:41	Now, I'm going to show you another way we can do this under steady state conditions as well and this is where we're actually going to be making cell by cell changes or a block of changes.
23:50	So, let's, just to demonstrate this, get ourselves up and running, we'll go back into steady state here and we're going to choose, let's say 3000 RPM should be enough.
23:59	And we'll just get ourselves up and running under steady state conditions.
24:02	Again just being aware that we do have a fair bit of heat going on here.
24:07	But looking at our total trim at the moment, we're sitting at, actually bang on, our number's 0% so not much work to be done here, but how this works is what we can do is look at our air flow meter output in grams per second, find the corresponding spot, so we're sitting at about 24 grams per second.
24:22	So, that happens to be around about here.
24:25	Maybe what we might want to do is choose three or four cells in that area.
24:29	We're now currently sitting at a positive 2% trim, so Control J, we'll add 2% and there we go, that's just dialled those cells in.
24:38	Obviously, it still moves around a little bit, but again you can maybe take a couple of cracks at this.
24:43	Once you've done that then we'll just increase the throttle or RPM to bring ourselves up, in this case let's go to 31 grams per second.
24:51	So, we're looking at that, 31, we're right there.
24:54	Just gone a little bit too far, but we'll try these cells here.
24:58	Control J, let's add 2%, there we go, error is gone.
25:02	So, there's multiple ways of dealing with the approach to tuning the mass airflow sensor, the XY plot though is just a really nice way of gathering a large amount of data, so that we can analyse it all together.
25:14	Not looking at individual data points, but rather averaging it out because things do naturally move around a little bit.
25:21	However, this is something we can still monitor as we go through and optimise the rest of our calibration, but at this point I'm pretty happy with what we've got there with our MAF calibration and we can move onto the next step of our process.