air/fuel and timing control (1.8 ECU tuning session!)

[Project_Legacy]

i was at the local "speed shop" today, and was asking about piggyback systems i could use to control my 1.8 ECU in an impreza. ya'lls might think i should be asking elsewhere, but i know people on here who have done this before.

basically, from what i have heard, boosting on the stock 1.8 ecu, timing is bad up top, and it runs rich? i asked the "speed shop" about what i can use to combat that, and they say i can use an S-AFC to lean it out up top, cuz it will run rich and probably bog out. NOW WHAT GETS ME, is that they said, fixing the up top rich condition, will in turn make my timing up top better. so, what they are trying to say is, the S-AFC will control my fuel and timing issues. IS THAT CORRECT??

im going crazy thinking about all the piggyback systems. smt6, emanage, and now this. i read about the S-AFC's and it doesnt say ANYTHING about timing.

on another note, i was asking them about the emanage, and they said it didnt control timing.... which is weird since the emanage has a harness for ignition timing control.

let me know what ya'lls think. thanks!

[vrg3]

Don't go back to that "speed shop."

If you boost an otherwise stock EJ18, you get too much ignition advance and lean running at the top end.

If you use an S-AFC to lean out fuel mixtures you actually generally end up with more advance at the same time. People who sell S-AFCs generally don't advertise that fact because it is often undesirable.

Finally, the S-AFC affects timing but offers you no direct control over it. The e-Manage does allow you some timing control.

[my3awds]

Yes that speed shop is incorrect. Having too much timing uptop causes you to run lean, not rich. So what you want to do is get something like the e-manage, which will do what the s-afc AND s-itc can do (adjust air:fuel and retard timing). This is all rough estimates, but you want to pull back timing up top so you max around 15-20. This will in effect cause you to run richer. Thats where either a wideband or scantool will provide useful. With timing around 15-20, you can lean out your WOT or HI settings on, say, an safc enough to get your wideband o2 sensor to read 11:1 (depending on what your goals are). A knock link or Knock lite will help keep your swap on the ej18 in check. It wouldn't hurt to have an egt gauge as well to make sure you don't melt things either. (I've seen my fuel ratio on a dyno say I'm rich, but my egt temps were skyrocketing)

Manny~

[Project_Legacy]

yea, that is what i figured. i dont see why they say they can dyno tune me on the S-AFC without doing something to the timing. maybe there are different ECU's that respond differently to it? i guess i gotta go with the emanage... hey vrg3, have you dealt with the emanage before? is it hard to tune? i checked out the perfect power software, and it is like, over my head... let me know what you think. thanks a lot!

Don

[Project_Legacy]

could you get away with running the S-ITC and an S-AFC??? tuning might be a pain to do wouldnt it? having to deal with two different devices. meh, i dunno. they actually didnt say it would run rich, i assumed it would. so running too much timing up top causes lean conditions, causing detonation?

thanks for chiming in guys. im having a hard time with all this EM BS.

[vrg3]

I've never used an e-Manage myself.

If the complexity of the Perfect Power goes over your head, you need to read up on it until you're comfortable with that level of technicality. Tuning an engine ain't simple, so you need to learn a lot before you can do it.

You could use an ITC and an AFC but as you say it can be a pain because they're both fairly crude and adjustments to the AFC affect timing.

Too much timing advance doesn't make you run lean. The two are fairly independent.

[Project_Legacy]

ahhh.... yea this whole past two weeks ive been looking around, searching, checkin out stores, and crap like that about all of this. they have books on tuning correct? where would one go to find a book like this? local book store? i would love to get into this kind of stuff.

i figured that timing and fuel would be independent... there is just so much conflicting info and people out there...

[my3awds]

Too much timing advance doesn't make you run lean. The two are fairly independent.

Then what is the 'real' cause for one to run into a leaner conditioin when you advance timing, and a richer condition when you retard timing. On 90-94 Eclipses, they have an adjustable CAS. Retarding the timing by turning the CAS richened my entire settings. On all of my fuel trims (low, mid, hi) and during openloop. True, timing advance may not be the 'direct' cause for running leaner, but it does cause it, doesn't it? Just like leaning on an safc or similar causes you to not only run leaner, but it advances your timing indirectly. Sure, the 'safc doesn't advance your timing' ...but it does, and I guess thats what I meant to say in relation to too much timing and running lean.

[vrg3]

First, let's clarify (in simple terms) what happens with an AFC:

The ECU pulses the injectors long enough to inject enough fuel to burn the amount of air that it senses with the MAF sensor. It also fires the spark plugs at a time appropriate for the current load on the engine.

So, to reduce the amount of fuel injected, an AFC reduces the signal the ECU gets from the MAF sensor. This makes the ECU think less air is entering the engine, so it injects less fuel.

However, it also makes the ECU think the engine is under less load, so it starts the burn earlier (i.e. advances the timing) to ensure the less dense mixture finishes burning in time.

To be clear: an AFC directly modifies one of the ECU's primary inputs in its spark timing calculations.

Now, to try to answer your question -- I don't know exactly why you saw richer mixtures when you turned your crank angle sensor. There are a couple possible explanations I can come up with, assuming you were testing with a wideband oxygen sensor:

1) Retarded timing generally creates higher exhaust gas temperatures and pressures. This can make an oxygen sensor read richer than it should. I wouldn't expect this to act all across the board though.

2) With timing retarded, you were likely seeing a less complete burn. As a result, some free hydrocarbons escape out the exhaust valves and head past the wideband oxygen sensor, which detects them and gives a richer reading.

Note that these only affect the perceived AFR through an oxygen sensor in the exhaust -- the actual quantity of air ingested divided by the actual quantity of fuel injected is the same. Moreover, this is completely different from what happens when you use an AFC to manipulate the airflow signal.

Does that make sense? I may be missing something, but I'm pretty sure this is a good start.

[Project_Legacy]

so, when i would adjust fuel, in turn, it would adjust timing just a lil bit because it thinks its under less load. but, the adjust of timing wouldnt be enough to compensate up top on a 1.8 ecu that is boosted. therefore, i would need something that directly controls timing enough to make sure the motor doesnt run too advanced. basically, i want to retard the timing then.

i believe i am starting to understand, so when the timing gets advanced too much (lets say, up top on a 1.8 ecu that is boosted), the car will detonate because the spark isnt sparking at the right time? so, adjusting the timing to spark at the optimum level would be the best thing to do, to ensure against detonation. can preignition happen as well in this situation? or is it because it is running too advanced, that it doesnt spark the fuel in time before it is already about to leave the cylinders?

i have a rough understanding, and i thank you very much vrg3! i would love to learn how to tune well. its something that has always been in the back of my mind, of what i wanted to do.

[vrg3]

so, when i would adjust fuel, in turn, it would adjust timing just a lil bit because it thinks its under less load.

Correct. In particular, it would advance the timing.

but, the adjust of timing wouldnt be enough to compensate up top on a 1.8 ecu that is boosted.

Not quite -- it would actually be the opposite of compensating. Boost usually requires timing retard, and pulling fuel with an AFC causes timing advance.

therefore, i would need something that directly controls timing enough to make sure the motor doesnt run too advanced. basically, i want to retard the timing then.

Correct again.

i believe i am starting to understand, so when the timing gets advanced too much (lets say, up top on a 1.8 ecu that is boosted), the car will detonate because the spark isnt sparking at the right time?

Yeah, that's basically it.

so, adjusting the timing to spark at the optimum level would be the best thing to do, to ensure against detonation.

Well, it is one necessary thing to do to prevent detonation. You also have to make sure the air/fuel ratio is appropriate, and that the engine is in good working order.

can preignition happen as well in this situation?

Runaway detonation can lead to preignition, yes.

or is it because it is running too advanced, that it doesnt spark the fuel in time before it is already about to leave the cylinders?

Let me clarify:

Advanced means earlier. Retarded means later. So a spark that happens 20 degrees before top dead center is more advanced than one that happens 15 degrees BTDC.

To fully understand why too much advance can cause detonation you need to understand how combustion works. I don't fully. But in very simple terms, more advanced timing means greater peak cylinder pressure (up to a point), which means uncontrolled combustion is more likely to happen.

i have a rough understanding, and i thank you very much vrg3! i would love to learn how to tune well. its something that has always been in the back of my mind, of what i wanted to do.

You're welcome. I wish I could recommend some books but I can't think of any off the top of my head. Maybe you could start just by Googling a lot.

[vrg3]

Oh -- another thing: if you're running low boost, you might consider installing a rising rate fuel pressure regulator (and a high-capacity high-pressure fuel pump to go with it) and leaving everything else stock. You could probably get away with leaving timing stock if you run rich enough.

[Project_Legacy]

yea, right now, all i would be running would be a vf10 probably around 8-10 psi for now.

yea i didnt understand which way advanced and retarded were. thanks for clearing that up. i had it backwards. LoL.

Oh -- another thing: if you're running low boost, you might consider installing a rising rate fuel pressure regulator (and a high-capacity high-pressure fuel pump to go with it) and leaving everything else stock. You could probably get away with leaving timing stock if you run rich enough.

so if i did this, the timing would still be advanced, but it wont get any more advanced than it would with an AFC. i dont understand how that would work though. is it because there will be enough fuel in the cylinder when the spark happens to not run lean?

i think i may understand, i just cant word it right. LoL

[vrg3]

yea, right now, all i would be running would be a vf10 probably around 8-10 psi for now.

Oh. That's not low boost. You should use some kind of timing control with that much boost.

so if i did this, the timing would still be advanced, but it wont get any more advanced than it would with an AFC.

Advance and retard are relative terms, so be careful throwing the words around.

An AFC alone does not advance or retard timing; it depends on what the settings of the device are. If it is used to reduce fueling (like is often done when big injectors are installed), it tends to advance timing. But if it is used to increase fueling (which is not done quite as often because it's still subject to the limits of the rest of the system), it tends to retard timing.

The reason I was suggesting an RRFPR is that it would allow you to increase fueling (as is necessary when a turbo is added to an engine) without affecting ignition timing at all. It's also a very simple device that requires a minimum of "tuning."

i dont understand how that would work though. is it because there will be enough fuel in the cylinder when the spark happens to not run lean?

Let me again make a clarification:

The burning of air and fuel is a chemical reaction. Like any chemical reaction, the reactants react in a certain ratio to produce the products. The word used to describe this is stoichiometry. In this case, every gram of fuel reacts with approximately 14.7 grams of air. The products are water, carbon dioxide, some other random compounds which are pollutants, and heat (which is what actually drives the engine). So 14.7:1 is the stoichiometric ratio of air to fuel for combustion.

"Rich" just means a lower ratio -- or more fuel.

"Lean" just means a higher ratio -- or more air.

In any case, the air/fuel ratio is a measure of what goes into the cylinder prior to the burn. It has nothing to do with how the burn actually happens. (This is complicated by the fact that the only good way we have to measure air/fuel ratios is really by looking at the products of the combustion and inferring what happened, and that can be influenced by how the burn happened. This is what my3awds and I are discussing.)

Sometimes people use the words in relation to the stoichiometric ratio -- air/fuel ratios above 14.7 are called "lean" and ratios below it are called "rich." Other times, however, the words are just used as relative terms. An air/fuel ratio of 13.0:1, for example, is richer than stoichiometric, but leaner than 10.0:1.

Within the vicinity of the stoichiometric ratio, richer mixtures usually mean cooler burns and leaner mixtures mean hotter burns. This is largely due to the fact that vaporizing fuel absorbs a good deal of heat (if you've ever spilled gasoline on your hand you probably noticed it felt cold).

So, running rich can help stave off detonation (and in fact turbo engines need to run fairly rich when on boost). That does not mean it is equivalent to retarding ignition timing, however.

[my3awds]

Maybe we should change the topic of this thread to "vrg3's lesson on tuning" LOL. All great information! This also clears up my misconception of the proper use of the afpr (rrfpr), since previously I only understood the need of it when running a larger fuel pump, to help prevent overrun and turn back to stock pressure. Since you had mentioned that increasing the fuel pressure helps richen settings without increasing timing, does increasing the pressure effect IDC at WOT? Perhaps increasing the fuel pressure higher then stock without a way of viewing the duty cycle isn't safe? Since I have no scantool to use on this 1.8 ecu, my only tuning tools is a wideband, egt, knock lite, and o2v readings. This makes tuning alot more difficult then what I am use to (pocketlogger).

Sorry to butt-in on your thread Project haha, but vrg3 always posts great info.

[vrg3]

Well, it seems kind of relevant... But, yeah, maybe you might want to adjust the title of the thread to reflect what we're discussing a little better, Project_Legacy?

But please don't think I'm an expert on tuning. I'm not. I'm just sharing what little I do understand (or think I understand).

my3awds, there are indeed some situations where you need to change the FPR just to keep fuel pressure the same after installing a high-pressure high-flow pump.

A little background: an injector is basically a valve that the ECU opens and closes. When it is open, it connects the fuel rail to the manifold. The pressure in the fuel rail (called fuel pressure) is kept always higher than the pressure in the manifold (what we think of as vacuum or boost), so the valve being open makes fuel flow from the rail into the manifold.

As with any valve, the actual quantity of fuel that goes through the valve is determined primarily by three factors:

1/ The size of the opening in the valve -- in this case, the injector nozzle size. This is generally not changeable, so we just accept it for any given injector.

2/ How long the valve is open. If you hold the valve open twice as long you get twice as much flow. So what we mean here is the injector pulse width (or injector duty cycle, if you divide by the actual pulse width by the maximum pulse width). The pulse width is determined by the ECU. It knows how much an injector flows and knows how much fuel it wants to inject, so it holds the injector open long enough to flow that much fuel. Mucking with its airflow input (like with an AFC) effectively mucks with injector pulse width.

3/ The amount of pressure differential across the valve -- in this case, fuel pressure minus manifold pressure. If you quadruple the pressure differential, you double the flow through the injector. In stock form, this pressure differential is held constant at all times.

Now take a quick look at this post for a quick explanation of how an FPR works:

http://bbs.legacycentral.org/viewtopic. ... 6779#46779

Also, if you have the patience, look at this thread for a discussion about rising rate fuel pressure regulators (it's an argument, so there is some misinformation there, but just pay attention to my posts in the thread and you'll be fine):

http://bbs.legacycentral.org/viewtopic.php?t=20255

Okay. So there are two separate cases that might require a change of FPR:

1/ The one you are thinking of. It is possible for a very powerful fuel pump to push so much fuel that the stock FPR is a restriction even when it's fully open. The result is abnormally high fuel pressure especially in situations where fuel pressure is supposed to be low. A high-flow FPR replacement set to the same pressure as stock can resolve this issue. As you say, this does not actually increase fueling or anything -- it's just to keep stock behavior despite a big pump.

2/ The case where you want to alter mixtures without involving the ECU. As I said above, the ECU knows (or thinks it knows) how much the injectors flow and pulses them accordingly. If you increase fuel pressure you increase fuel flow. Aside from the effects of feedback in closed-loop operation, IDCs do not change and the ECU has no idea anything is happening differently. You just get more fuel flow. An RRFPR is a pretty neat, if crude, way to do this -- it enriches the mixture as boost increases.

Does that make sense?

[Project_Legacy]

nah its kool this is all great information. there is so much to learn just to make the boosted 1.8 ecu run right. LoL.

ok im getting it now. the terms and how i use them are just off a little bit.

so how would i be able to monitor the timing? would an emanage be able to monitor it so i know how much i have to change? i havent seen anything that says it can monitor timing. timing seems to be the only thing i would have trouble with for the 1.8 ECU.

since we would be using the 1.8 ECU on 2.2t injectors, the car would run really rich at idle correct? the ecu thinks it has the same injectors in it, and keeps the same duty cycle, except that these injectors are bigger, causing a rich condition. when boost would hit (say 8-10 PSI) is there effectively a different A/F ratio than the ECU normally sees and it doesnt dump enough fuel up top? i have no idea why the timing would change at this point. that is what i am stuck at so far...

[vrg3]

A scan tool for your ECU will tell you what timing the ECU has chosen. As long as you're not using an S-ITC or anything, this will be correct.

If you're on a dyno, you can also use a timing light to read ignition timing directly.

I've never used the e-Manage, so I don't know what its software shows you.

If you're tuning with a piggyback, you may not necessarily need to know the actual ignition timing as long as you know how you're changing it and you're not changing it too drastically.

[my3awds]

2/ The case where you want to alter mixtures without involving the ECU. As I said above, the ECU knows (or thinks it knows) how much the injectors flow and pulses them accordingly. If you increase fuel pressure you increase fuel flow. Aside from the effects of feedback in closed-loop operation, IDCs do not change and the ECU has no idea anything is happening differently. You just get more fuel flow. An RRFPR is a pretty neat, if crude, way to do this -- it enriches the mixture as boost increases.

Does that make sense?

Ah, it makes perfect sense. So, in a sense, one could boost a 1.8, and run low boost (3-5psi) on 91+octane and fight the effects of high timing in Openloop with just an rrfpr, theoretically. Also, it would make sense then that an rrfpr has the same effects of solely getting larger injectors, based on your explination (unless I don't understand it correctly). The ecu would tell the injector to hold the valve open for the same X amount of time at open loop, but you would get more fuel since you've increased the size of the valve. In both cases, you would have to adjust your closed loop operation accordingly, but you'd get the same effect right?

A scan tool for your ECU will tell you what timing the ECU has chosen. As long as you're not using an S-ITC or anything, this will be correct.

The only bad thing is they don't have freeware scan tools for us 1.8 impreza users like you've created for the 2.2t legacy ecu. I think the only hope is that vwrx software, but the cable might not even work properly on our cars. I was planning on using an S-ITC to pull timing at 10+ boost psi, since I already have an S-AFC, but it sounds like you don't recommend using this to retard timing. Is there a particular reason why? I was planning on simply pulling back about 10 degrees at the top of the RPM band, and just tuning out the rest with the safc & rrfpr and monitor it via wideband/knocklite. I know sounds crude, but I'm not sure what else to go by when you can't see how much timing you have.

[vrg3]

Ah, it makes perfect sense. So, in a sense, one could boost a 1.8, and run low boost (3-5psi) on 91+octane and fight the effects of high timing in Openloop with just an rrfpr, theoretically.

Yes, pretty much. Except -- again, be careful with terminology. In the context of engine management, "open loop" and "closed loop" usually refer to fuel. They don't have anything to do with spark. The loop in question is the feedback loop completed by the oxygen sensor.

Spark timing is actually controlled by another feedback loop, with the feedback provided by the knock sensor. This feedback loop is active pretty much all of the time as far as I know.

And yes, a low-boost turbo setup will often do just fine with an appropriate RRFPR and without extra timing controls as long as you use the highest-octane fuel you have available and you have a capable enough fuel pump.

Also, it would make sense then that an rrfpr has the same effects of solely getting larger injectors, based on your explination (unless I don't understand it correctly). The ecu would tell the injector to hold the valve open for the same X amount of time at open loop, but you would get more fuel since you've increased the size of the valve. In both cases, you would have to adjust your closed loop operation accordingly, but you'd get the same effect right?

Not quite. There's a little more magic to an RRFPR. An RRFPR increases the pressure across the differentials in proportion to manifold pressure. Furthermore, the more sophisticated RRFPRs don't even act until manifold pressure goes above atmospheric. That means that at low loads the RRFPR has minimal (for the simpler ones) or zero (for the better ones) effect at all. So off-boost fueling remains exactly the same as it was stock!

Neat, eh?

The only bad thing is they don't have freeware scan tools for us 1.8 impreza users like you've created for the 2.2t legacy ecu. I think the only hope is that vwrx software, but the cable might not even work properly on our cars.

The cable should. You'd just need to figure out all the parameters, their addresses, and their transfer functions.

Maybe one day I'll modify my scantool for your ECU. I just owe the XT6 guys an attempt at theirs first.

I was planning on using an S-ITC to pull timing at 10+ boost psi, since I already have an S-AFC, but it sounds like you don't recommend using this to retard timing. Is there a particular reason why? I was planning on simply pulling back about 10 degrees at the top of the RPM band, and just tuning out the rest with the safc & rrfpr and monitor it via wideband/knocklite. I know sounds crude, but I'm not sure what else to go by when you can't see how much timing you have.

I just don't recommend it because it's so crude as you say, and because it seems like the bang for the buck isn't very good.

I'll repeat that I'm no expert, so don't give my opinion more weight than it's really worth.

But an S-AFC plus an S-ITC are gonna run you somewhere in the vicinity of 500 dollars, right? That's so much money! With that kind of cash you could buy a J&S SafeGuard which can retard timing as boost increases while also greatly improving upon the stock ECU's knock correction (wait -- does your ECU even have knock correction?). A SafeGuard plus an RRFPR would do great!

Or you could probably pay someone less than $500 to build you a harness to make an EJ22T ECU practically plug right into your car, and you'd have the right ECU for your engine. OEM reliability and robustness with no need for tuning.

[Project_Legacy]

ok kool. i read a "howstuffworks" article about timing and i think i know what i need to know about it.

so i could go with: (prices are rough)
-Apexi S-ITC $300 new (about $150-$200 used?)
-Apexi S-AFC $300 new (about $150-$200 used?)
-Wideband o2 sensor and monitor $300 new

or i could go with:
-Greddy Emanage with timing/fuel harnesses and software $500-$600 new
-wideband o2 sensor with monitor $300 new
-borrow a laptop $FREE$

should i need anything else for this set-up? Just thinking out loud here.

on another note, how would you wire in the wideband o2 sensor? it doesnt go into the ECU correct? basically, would the wiring be o2 sensor goes to o2 monitor, and thats it? because, all i would really need it to do is just to monitor it right? would i be able to use one of those $50 dollar A/F gauges with a Wideband o2 sensor?

EDIT: wait a minute! the hell is the J&S safeguard?? would that actually be better for us than using all this money???? let me know wassup with that too! thanks! haha.
EDIT #2: although, if i wanted to do this, i would still need an AFC to control the fuel then... as i would want to run more PSI right? hmm... i guess it wouldnt be for me then, since the price with the AFC would be like $800. it sounds like a great machine though.

[my3awds]

Here you go Project : http://www.jandssafeguard.com/safeguard.html

From my understanding, it looks like a way to control ignition AND it acts as a guard against detonation.

Maybe one day I'll modify my scantool for your ECU. I just owe the XT6 guys an attempt at theirs first.

If this ever happens, I would gladly paypal you for your services. People don't realize how important and valuable this scantool is...and you offered it free. You should have a little link on your site for paypal contributions, at the very least imo.

would i be able to use one of those $50 dollar A/F gauges with a Wideband o2 sensor?

I think the a/f gauges you are thinking of are those cheap autometer gauges that are narrowband only. They are only good for monitoring your o2v signal from your ecu. I think you can find the complete wideband kits, gauge and sensor, for around 299+ new. I'm hoping I can source a cheap Uego, since the gauge is clear as day at what your air:fuel ratio is.

[Project_Legacy]

yea i checked that out earlier. doesnt look like it would be for me, since i would still have to address fueling issues. i think ill stick with the emanage.

EDIT: yea i was talking about those cheaper autometer ones. so id still have to get the expensive ones then huh? so what about the install of the wideband o2? how would that go? that doesnt connect to the ECU at all does it?

[my3awds]

Yeah it is independent of the ECU since it gets it's own reading from it's own sensor. The sensor just needs to be tapped somewhere early on the downpipe. The wrx downpipe I have, I tapped a bung over by the stock location of the 1.8 o2 sensor, which leaves the stock wrx rear o2 sensor bung free. If it's not too far to get a good reading, I'll probably use that for the wideband.

[my3awds]

Here's my projected spending, maybe this might help your decission, Proj.

I'll be getting the following next paycheck:
Vortech FMU $90
Knock Lite $120
Walbro 255 $100
Boost Controller $20

Total $330 to allow me to safely run low boost (3-5psi). I already have the safc, so I can adjust a bit as well, but probably isn't needed.

Then on the last check:
UEGO wideband $300
Super ITC $100-200

Total $4-500 and install the larger stock 370cc injectors (instead of the 1.8s in there now), retune for 9-11psi and I'll be set! I'll have pretty good control over air, fuel, and timing by this point (not the best, but good enough) and I'll be able to monitor my knock. And if vrg3 makes the scan tool compatible with the 1.8ecu, we might be able to get our fuel trims close to 100%