Emissions Tuning Fundamentals: Fuel Tuning

Back to course

Watch This Course

$299 USD

-OR-

Or 8 easy payments of only $37.38 USD Instant access. Easy checkout. No fees. Learn more

Enroll Now
8 easy payments of only $37.38 USD

Course Access for Life

60 day money back guarantee

Fuel Tuning

25.27

00:00	Since this course assumes you've already taken the prerequisite HPA tuning courses or otherwise already possess the knowledge, we won't be covering how to scale a MAF or build a VE table.
00:10	That said, it's worth noting that the level of precision required for emission success is much greater than the level of precision required just to make something drivable.
00:19	Fuel trims of plus or minus 10% might be acceptable in light throttle areas for off road use but for emissions compliance, we'd suggest trying to get them down to plus or minus 3% and settle for 5% only if absolutely necessary.
00:34	While closed loop operation helps to ensure that your air fuel ratio is matching target, it's a reactive system, not predictive, meaning that before the closed loop control can correct an error, the error must be present in the first place.
00:47	The closer to the target we have our fuelling, the less error will be present, the less work the closed loop system has to do and therefore the less emissions we'll be producing.
00:56	Dynamic enrichment and leanment and wall wetting compensations should similarly be configured to keep fuelling on target as much as possible.
01:05	As we've discussed elsewhere, while chassis dynos with electric motors can help simulate downhill conditions where an engine can be driven at very high levels of manifold vacuum, conventional dynos can't offer this.
01:17	This can however be simulated on a road or track with a downhill portion to help you dial in those super light load areas of the map.
01:24	No matter how wisely chosen your lambda targets are, if the engine needs a significant trim to hit a target, that means it's going to be off target until the closed loop system corrects the error.
01:35	During that time, emissions likely won't be favourable.
01:38	On diesels, this isn't quite as simple since most don't calculate and allow you to monitor target lambda.
01:44	At high load, when against a rich limit, you can compare it to wideband data and tune your injection table until the monitored data matches the rich limit.
01:53	Specifically, I mean if your rich limit is 1.2 lambda and you're achieving 1.15, you'll need to reduce table values until you hit that rich limit target.
02:02	At light load, you can monitor wideband data and keep the engine near stock behaviour.
02:07	Regardless of the type of engine you're tuning, proper injector characterisation is vital and we've included this information in our practical reflash and practical standalone tuning courses.
02:17	For poured injected engines, that includes injector latency, also known as dead time or battery offset, short pulse width compensation which accounts for flow non linearity at the low pulse width range of the injector operation and injector flow rate at the differential fuel pressure being used.
02:34	For gasoline direct injection, you may only find flow rate versus pressure and a latency or offset table.
02:41	Diesel injectors are characterised a bit differently.
02:44	Generally, there's a single table and you enter the injector pulse width values required to deliver the specific fuel mass value on one axis at each fuel pressure level found on the other axis.
02:55	Starting stock and only swapping in your new injectors is a great way to see the impact they have by themselves.
03:01	Enter the characterisation data, adjust all time based tables, meaning tables where you specify fuel delivery as an amount of time and then see if the engine operates the same or not.
03:11	With great injectors and data, the engine should operate just as it did in stock form unless you pick something that flows too much to be able to control accurately at small pulse widths.
03:21	With a direct injected engine, a small increase in injector flow rate may not require changing injector timing values to maintain clean emissions.
03:29	Injectors that flow 30% over stock will require significantly more work and it can be difficult or even potentially impossible to keep emissions clean.
03:37	DI relies on precise timing of fuel delivery and if the injector flows a lot more than stock, while maintaining the same fuel pressure, the timing of fuel delivery changes.
03:47	Reducing fuel pressure to maintain start and end of injection timing may solve the timing concern but this can reduce atomisation.
03:54	This negatively impacts evaporation, air fuel mixing and combustion quality.
03:59	In direct injected systems, end of injection is generally more critical than start of injection so you can alter start of injection time to maintain a similar end of injection but that doesn't guarantee clean operation.
04:11	Remember that DI engines can do amazing things when fuel is delivered in a very specific way over a very specific period of the engine cycle.
04:20	Next, on a race vehicle that only operates off road, you may have used extra acceleration enrichment to ensure maximum response but extra enrichment on a road going vehicle can cause excess emissions.
04:31	The rule to live by on a gas engine is that lambda one is your friend.
04:36	While some tailpipe emissions increase as we move richer than stoic and others increase as we move leaner, the overall combined tailpipe emissions are generally minimised at lambda one.
04:47	At the stoic ratio, the catalyst can also perform its job properly, leaving fewer emissions remaining.
04:54	Deviation from lambda one should be avoided except when needed for component protection or NOx mitigation.
05:00	One caveat is that some manufacturers on some vehicles purposely cycle the engine slightly rich then slightly lean of stoic to cycle catalyst operation.
05:10	While slightly lean of stoic, the catalyst is able to manage NOx and store extra oxygen.
05:15	Then the ECU switches to slightly rich of stoic and that stored oxygen gets consumed catalysing hydrocarbons and carbon monoxide which the rich of stoic mixture creates.
05:25	The cycle continues back and forth over and over on vehicles using this method and this generally involves readings taken at both the primary and rear oxygen sensors.
05:35	Again, that all happens at or very close to lambda one but there are conditions that require targeting significantly richer ratios.
05:43	Moderate to high load areas encountered on the US06 drive schedule or during similar driving can cause high heat which can create NOx.
05:51	The amount of NOx permitted by current regulations is much smaller than the allowance for hydrocarbons and for that reason, bringing in some additional enrichment to cool, combustion and avoid NOx is the right move.
06:03	Measuring with a 5 gas analyser or higher end device will allow you to see when you've gone too far in either direction.
06:10	If you're tuning a diesel engine, the stoic air fuel ratio being chemically correct doesn't coincide with reasonable emissions.
06:16	In fact at lambda one you'll have significant black smoke pouring from the tailpipe and hence hydrocarbon and particulate matter emissions are completely unacceptable.
06:26	Giving a diesel that much fuel can also cause dangerously high combustion temperatures leading to the failure of parts such as pistons.
06:33	What does remain the same as a gas engine is that most of the core emissions are reduced as you go leaner.
06:39	That means avoiding overfueling solves most diesel emissions concerns but you need to be mindful of crossing the line going too lean which will create excessive NOx.
06:49	Diesel, catalyst and EGR systems are focused on NOx reduction to assist you but again, a 5 gas analyser or higher end system will show you when you've gone too far.
06:59	On both gas or diesel applications, closed loop fueling should be utilised to hit targets wherever possible.
07:05	Just be mindful that the factory calibration may switch to open loop to avoid using skewed readings caused by high exhaust manifold pressure at high engine load.
07:14	Injection timing is discussed in our diesel tuning course and port injection timing is mentioned in lessons on dialling in a big cam.
07:21	Hopefully you aren't trying to meet emissions with a big cam as that may also be an impossible task but the methods shown to optimise timing can benefit any engine.
07:31	With a mild aftermarket cam, you may need to adjust injection timing to maintain efficiency and clean operation.
07:36	Otherwise on DI engines or DI plus port injected engines, avoid significant changes to direct injection timing and the ratio of fuel delivered between the direct and port systems unless necessary to deliver enough fuel at high load.
07:52	On diesel engines, if you want to attempt improvements in engine noise and drivability using DI timing adjustment at idle or light load, understand you'll need to verify whether that degrades emissions using test equipment.
08:04	Next, because the importance of what I've covered in this module cannot be overstated, we've set up a series of tests in the emissions lab.
08:12	So, in order to demonstrate the impact of changing your air fuel target on emissions, I've set up a specific calibration which forces full time open loop operation and allows me real time adjustment of fueling through Cobb's Access Tuner Pro software.
08:28	In order to achieve that, I've run the TGVs open all the time and altered some settings which would normally keep the vehicle in closed loop at light load and idle operation.
08:38	First thing of note here is at the moment, if I look down at my laptop, we're right at about 1 lambda, 0.99 occasionally.
08:47	Here on our math scaling we can see where we are.
08:50	And the current readings are right about 0 parts per million NOx, total hydrocarbons around 17, CO at 30, and CO2 at 0.19%.
09:04	So, the first three readings are in parts per million, CO2 is in percent, so just remember that as we're talking about those values.
09:12	Now, the first thing I'm going to do is I'm going to richen things up a bit.
09:19	I'm going to make a change where I add about 5% fuel, and we'll see what this does.
09:29	There is going to be a transport delay, and the cat may be able to clean a significant amount of this up, but let's see what this does.
09:38	So, we're down to about 0.95 lambda, and I'm waiting to see what type of response we get from the total hydrocarbon value.
09:50	So, far we're sticking right around 17.
09:59	So, in this case, the cat is still able to clean things up pretty well.
10:05	There is a significant amount of oxygen stored in the catalyst, so it takes a while to consume that after you deviate from lambda 1.
10:13	And I think what we're starting to see now is that as that oxygen gets consumed, we achieve a steady state operation where we are unable to burn the additional hydrocarbons that I'm providing through overfueling.
10:29	So, let's see how high this gets when it really plateaus.
10:41	Lambda is still holding steady, right about 0 .95.
10:48	And it looks like we've hit a peak and then found an equilibrium around 24, so quite a bit more hydrocarbons than we had before.
10:55	And since we're richening things up, NOx is not increasing.
11:00	NOx will generally be present during a lean condition or high-temperature condition.
11:07	Okay, so I'm going to take that additional fuel back out, and we'll see how quickly we get back down to our more optimal total hydrocarbon value.
11:17	Also, we'll note that CO, carbon monoxide, is up around 130 parts per million.
11:23	Let's see what happens when I take that extra fuel back out.
11:40	It looks like the vehicle shifted just slightly, so now we're at 0.98 to 0.99 lambda, where before we were at 0.99 to 1.0.
11:49	So, for the sake of this demonstration, I'm just going to lean things out a little bit, get us back up right about lambda 1.
12:07	You can see the carbon monoxide has come back down.
12:10	Looks like we're down to about 35 parts per million now.
12:14	Total hydrocarbons still up at 22.
12:25	Taking just a hair more fuel out, really trying to get us right at lambda 1.
12:43	Carbon monoxide really coming down now.
12:46	Total hydrocarbons gradually coming down.
12:50	It's really important to see how long this takes.
12:54	Anytime you're doing steady state work with emissions, right now we're at a CVS flow rate of about 16 CFM, I believe.
13:03	The transport delay isn't super long, but because of the way the catalyst operates, you aren't necessarily going to see the impact of your change right away.
13:14	What you're going to see is how the catalyst performs over time.
13:24	And in this particular case, we're seeing that when we go too rich, it can take quite a while for the catalyst to be able to operate optimally again.
13:36	Alright, so our total hydrocarbon value has come back down.
13:39	CO actually right around zero right now.
13:43	So, I think we're a little bit tighter to lambda 1 than we were when we started.
13:47	And now what I'm going to do is demonstrate a lean condition.
13:54	So, I'm going to go about 5% lean.
13:58	We'll try to shoot for about lambda 1.05 and see what we get.
14:14	Now, that we've gone lean, our total hydrocarbon value might come down, but our NOx value might come up.
14:23	And let's just wait and see what happens.
14:34	One thing that can sometimes complicate this is if the engine starts to stumble enough to misfire, then you might actually have an increase in total hydrocarbons rather than a reduction.
14:52	So, far it's looking like having the engine slightly lean of lambda 1 is actually working pretty well, but we are at idle.
14:59	And if we were driving, this could potentially overheat the catalyst, so this is something you'd have to be careful of.
15:10	At this point, I'm going to go back to lambda 1 in this idle area, and then I'm going to get the car rolling.
15:17	So, what I'm going to do here is get the vehicle in third gear.
15:21	This looks like here, right about 2200 RPM.
15:27	We're relatively close to lambda 1.
15:29	I'm just going to give our math scaling a small tweak here to get us a little closer.
15:38	So, I'm going to take 1.5% out of the math scaling.
15:44	Pretty close.
15:45	I'll take another 0.5% out here.
15:55	Another 0.5%.
16:04	And that looks like a little much.
16:06	Right about lambda 1.
16:11	And what I'm going to do this time is I'm going to alter our target here.
16:20	So, we're going to go from 1.0 to 0.9.
16:28	So, right now we've got NOx quite low, total hydrocarbons again quite low, right around 6 parts per million.
16:35	And we're getting 3 to 4 just from background air, so that's really clean.
16:41	And we'll see what richening it up to a target of 0.9 lambda does.
16:52	Alright, so we've gone slightly lean of our target, so we're hitting about 0.91 lambda on a 0.9 lambda target.
17:04	But let's see the impact that has.
17:06	Looks like we had a significant spot spike in NOx.
17:15	And total hydrocarbons are coming up, CO has gone quite high.
17:21	We're beyond the limit for that actually.
17:24	I see a warning on CO, so I'm going to take that extra fuel out.
17:31	Start bringing things back to normal.
17:33	And that gives you an idea of just how sensitive these systems are.
17:40	So, obviously that wasn't super rich, and chances are if you're working with a turbocharged vehicle, under some conditions it's going to be quite a lot richer than 0.91 lambda.
17:52	But that was rich enough to cause a massive increase in CO, more than 100-fold.
17:59	So, now we're going to let things stabilize a bit.
18:02	And then I'm going to make a smaller change.
18:11	Hydrocarbons and CO have come back down.
18:14	So, I'm going to go ahead and drop us from 1 lambda to about 0.98.
18:21	It looks like we're hitting about 0.98, 0.99.
18:30	And now we're going through that transport delay, getting to the analyzers.
18:34	Plus there's also oxygen stored in the catalyst that's able to catalyze some of this additional hydrocarbons that we're providing.
18:42	But as that oxygen storage is depleted, we'll start to see an increase in hydrocarbons and CO.
18:51	It's certainly not going to be as dramatic or happen as quickly, since I'm not as far from lambda 1 as we went before.
18:57	But we will start to see a change.
19:01	So, the CO is starting to creep its way up.
19:12	And at the moment I'm again not using closed -loop fueling, so I can see that our lambda is actually teetering at 0.98, 0.99 at the moment.
19:22	So, that's part of why we're not seeing a dramatic change.
19:26	I'm going to drop us down slightly further.
19:29	We'll go to a 0.97 target.
19:33	So, we get about 0.98 lambda result.
19:43	Let's see what that does for us.
19:48	There we go.
19:49	CO is starting to creep its way up.
19:52	So, as you can see, it took a good amount of time for that to have an effect.
19:56	If we were only slightly rich of target for a brief moment, it might not have had any impact on our emissions results at all.
20:03	But if we stay at steady state, rich of lambda 1, even just a couple percent for any significant duration of time, emissions are really going to suffer.
20:13	Alright,, so as we can see, CO continues to creep up.
20:17	We're going to set that back to normal.
20:22	And then in order to go lean of target, I'm going to skew our MAF scaling.
20:30	The reason being is in this particular software, lambda 1 is as lean as you're able to target.
20:35	So, in order to go lean of lambda 1, we're just going to take some fuel out of our MAF calibration.
20:42	And that way we'll provide less fuel because we're thinking that we have less air than we really do.
20:49	So, I'm going to take 5% airflow out of the MAF calibration.
20:55	And we'll get about 5% less fuel.
20:59	And we'll be at about lambda 1.05.
21:05	And let's see what result we get while we're moving.
21:16	So, at idle we weren't seeing a significant change, but as I mentioned, while driving, it's going to be a different story.
21:22	And as you can see here, NOx is coming right up.
21:27	Not seeing enormous spikes, nothing massively concerning at this point, but this isn't something we're going to want to keep up for long.
21:33	So, we'll take that change back out.
21:40	Alright, next.
21:42	I'm going to show what happens when you add a significant amount of fuel in an area that you might encounter during a drive cycle, for example.
21:53	So, not necessarily flooring it, not trying to make great performance, but maybe you're trying to get up a moderate hill, merge onto the highway.
22:02	Not something super aggressive, but we're going to enrich the car sooner than it would traditionally be enriched.
22:10	I'm doing that here in our primary open-loop fueling section.
22:14	So, what I'm going to do is I'm going to bring in some enrichment just a bit sooner than we typically would.
22:23	Here you can see we're tracing right here in these lower calculated load areas.
22:27	And I'm going to apply some additional throttle so that we get up into this range where we go up to 0.9 and then a 0.8 lambda target a bit sooner than usual.
22:37	Now, typically you can see that we maintain lambda 1 much deeper into the table.
22:43	And then as RPM increases on this vertical axis and load increases moving left to right, the targets become richer.
22:52	So, let's try to get us into this area and see how rapidly things change on our real-time emissions feedback.
23:16	All right, so CO has gone out of the ballpark.
23:21	And that's kind of what I wanted to show you there.
23:23	Obviously, total hydrocarbons started coming up, but the CO came up very quickly and to massive effect.
23:30	So, I've brought things back to lambda 1.
23:32	We'll let the cat start cleaning things back up.
23:35	Now, that absolutely would have failed the test horrifically.
23:39	Okay, so never something that you would want to do.
23:42	And that's part of what I'm trying to illustrate through these practical examples.
23:46	So, at a certain point you definitely need to richen the vehicle up for component protection, to protect the pistons from overheating, turbo manifold flanges, things of that nature.
23:57	But you never want to bring in additional enrichment from lambda 1 until you absolutely need it.
24:02	So, now you've seen what happens when you go rich or lean from lambda 1.
24:08	And with that in mind, choose conditions where you would do so very wisely.
24:13	Now, you're probably not going to be going lean of lambda 1 on any emissions control vehicle that would be driven on the road because it's not appropriate for the catalyst.
24:22	If you're in a diesel vehicle, of course things are different.
24:24	But for right now we're talking about a gas vehicle, lambda 1 is as lean as you want to go but you also don't want to be rich of it unless you absolutely have to.
24:34	Let's now recap our fuel tuning tips.
24:36	With a gas or diesel, avoid additional enrichment except when absolutely necessary for performance, reliability or NOx control.
24:45	Battery air fuel ratio targets achieve a compromise of fuel efficiency, performance, emissions and reliability so consider which of those factors your changes may be compromising.
24:55	MAF scaling and VE tuning for emissions compliance requires a higher precision so take your time and focus on the details.
25:03	Running lean of stoic reduces multiple core emissions but tends to create NOx if you go too far.
25:09	Injection timing can be just as important as lambda target selection and hitting your targets.
25:14	At a certain point, some injectors cross a threshold where to achieve much higher flow than stock, they no longer have the precision to achieve emissions goals.