ITC theory..

[Phoenyx]

All:

Yesterday I did a lot of research on timing and knock. I've purchased an Apexi Super ITC for the purpose of retarding my timing. Many people think that the 1G can control timing very well so that you don't need any kind of timing device. That is generally correct, but what I've found out is that it is assumed that this is the case for any mods. The fact is, there is another issue that could make timing *too* advanced and there are only a couple of ways to retard it down.

The issue that I'm having that appears to have advanced my timing is the installation of the Super AFC. Evidently if you are richening the fuel mixture (by lowering the air intake value to the ECU), your ECU will in turn, advance the timing, thinking that its safe to do so. This can lead to too much advance at the upper rpms and cause knock. If you lean out the mixture, you face the possibility of not having enough fuel.

The fixes I see (and please list more if there are more) are:

1) Get the Super ITC and compensate for the advanced condition.
2) Get bigger injectors so that you can lean out the mixture from the AFC which will then retard timing more by the ECU.

I'm not sure why my car is flowing so much air at 16psi on the stock turbo to the point that the Super AFC's max fuel enrichment is -18%, but that is enough to advance the timing a little too far. Which is probably the cause of my knock conditions. Especially in cold weather..

Any comments?

-M

 

[Blk_99gst]

Good question(s); but some of the fundamentals of some of the stuff you said are wrong. I'll go threw some quotes and try to clarify things. First off I would sugguest returning the Super ITC; don't know how much you spent on it (I'm sure it's expensive) but you don't need it. There are lots of fast people out there that don't need additional stuff like that. O.K, here we go (also I'm not an experted on 1g's, but I do know 2g's very well):

Originally posted by Phoenyx
The fact is, there is another issue that could make timing *too* advanced and there are only a couple of ways to retard it down.

There are some pretty extreme mods that one can do to get crazy timing advance. These are not bolt on mods though, this is like adjusting the base timing to be higher, tricking the ecu into things it's at a different altitude pressure; crazy stuff like that. Under normal circumstance and mods, you should never manually have to retard timing.

Originally posted by Phoenyx

The issue that I'm having that appears to have advanced my timing is the installation of the Super AFC.

Just an installation of a safc and leaving the correction values at "0" should NOT do anything to your timing. In fact, a SAFC with values of "0" acts like it's not even in the system. Now if you installed it and played around with some of the settings (hi, low) and/or you installed fuel injectors at the same time, then you can change your timing around. Do you have stock fuel injectors or upgraded ones?

Originally posted by Phoenyx

Evidently if you are richening the fuel mixture (by lowering the air intake value to the ECU), your ECU will in turn, advance the timing, thinking that its safe to do so.

You were on the right track, but one part is wrong. The SAFC does change the air flow signal to the ecu by either raising or lowering it. Usually people tell the ecu that there is less air coming in, because the ecu sees less air, it provides less fuel. So when you use the SAFC to say there is less air coming in, you are LEANING out the mixture, not riching it up.

The reason the ecu will advance timing because it thinks it is safe is because it chooses what timing to award you based on a lot of facters, air flow being one of them. When you lie to the ecu and say less air is coming in, it will use a more aggressive fuel map and provide you with more timing advance because of the lower air count. If you provided the ecu with the actual air flow that was coming it, it would be more stingy with timing advance.

Originally posted by Phoenyx

This can lead to too much advance at the upper rpms and cause knock. If you lean out the mixture, you face the possibility of not having enough fuel.

Knock is your friend in tuning (especially in a 1g car because you can actually data log knock count). Yes you do use the SAFC to lean things out all over the rpm band and at the upper rpm's. You really can not lean it out too much to the point where you get too much timing advance. The thing is if you lean it out too much with the SAFC, you will not have enough fuel and you will get knock. At this point if you're logging knock, you know that you've leaned things out too much and you need to richen it back up a tad. So the way you tune is you keep leaning things out with the SAFC so that you always get timing advance (you never want to see timing advance, drop, and then advance again) and you watch the knock values to make sure you don't get any or just a few.

Originally posted by Phoenyx

The fixes I see (and please list more if there are more) are:

1) Get the Super ITC and compensate for the advanced condition.
2) Get bigger injectors so that you can lean out the mixture from the AFC which will then retard timing more by the ECU.

Like I mentioned above, you shouldn't have to manually "fix" anything. Usually people use a SAFC to control (ie lean out) bigger injectors they are installing. It works like this. If you upgrade injectors say from 450cc to 660cc; your ecu doesn't know this. If it cycles the 660cc injectors like it would cycle the 450cc injectors, you will get way too much fuel and will probably flood the engine. You use the SAFC to lie to the ecu that less air is coming in, therefore it doesn't cycle the injectors as much, BUT because they are bigger injectors, you are getting more fuel when the ecu does cycle them to compensate for the more air coming in that you lied about. The added bonus is that you get more aggressive timing advance because the ecu is seeing a lower air count number coming in.

Originally posted by Phoenyx

I'm not sure why my car is flowing so much air at 16psi on the stock turbo to the point that the Super AFC's max fuel enrichment is -18%, but that is enough to advance the timing a little too far. Which is probably the cause of my knock conditions. Especially in cold weather..
Any comments?

Some of this doesn't make sense to me. Why do you think your turbo is flowing too much air at 16psi? Have you logged how much air it's pushing, either karman HZ from the SAFC or in lbs/min off of a data logger? You say max fuel enrichment is -18%; what this means is that you are saying 18% less air is coming in. This will cause a leaner condiction, not richer. Do you have upgraded injectors at all. If not, I'm pretty sure you should be nowhere near -18%. What is your timing advance and why do you think it's too much. And are you getting knock, have you logged knock count to see what the actual values are.

If you have stock injectors and -18% on the SAFC, I would probably expect you to get lots of knock and no timing advance, probably would get timing pulled.

/End of Book!!!

 

[kpt4321]

You're totally back-asswards.

LEANING out the SAFC by LOWERING the correction factors will decrese the airflow signal given to the ECU, will decrease the amount of fuel delivered, and will increase the amount of timing advance.

RICHENING the mixture by RAISING the values on the SAFC will increase the airflow signal sent to the ECU, will increase the fuel delivered, and will DECREASE the amount of timing advance.

Tuning is not that difficult. If it is knocking at a certain RPM, then RICHEN up that rpm's correction value on the hi throttle map in the SAFC. This means INCREASE the correction factors.

When you richen up the mixture, you will lower the amount of knock present by cooling the mixture with more fuel, and you will also decrease the amount of timing advance.

A SAFC is not a toy, people. You HAVE to know what you are doing with one of these before you buy it, or you can do a LOT of damage to your expensive motor.

1g's do not need ITC's. The SAFC is usually fine, because you can't really have "too much" timing advance, you just tune to get rid of knock. However, with REALLY huge injectors and very negative correction factors, you are still able to retard the timing using the CAS. No reason to buy an ITC.

 

[Phoenyx]

Originally posted by Blk_99gst
Just an installation of a safc and leaving the correction values at "0" should NOT do anything to your timing. In fact, a SAFC with values of "0" acts like it's not even in the system. Now if you installed it and played around with some of the settings (hi, low) and/or you installed fuel injectors at the same time, then you can change your timing around. Do you have stock fuel injectors or upgraded ones?

Sorry if I didn't word my explanation properly, but you misunderstood a lot of what I said.

I didn't just install the SAFC. I installed and took it to a tuning shop where they tuned the SAFC to get the proper fuel mixture.

You were on the right track, but one part is wrong. The SAFC does change the air flow signal to the ecu by either raising or lowering it. Usually people tell the ecu that there is less air coming in, because the ecu sees less air, it provides less fuel. So when you use the SAFC to say there is less air coming in, you are LEANING out the mixture, not riching it up.

The reason the ecu will advance timing because it thinks it is safe is because it chooses what timing to award you based on a lot of facters, air flow being one of them. When you lie to the ecu and say less air is coming in, it will use a more aggressive fuel map and provide you with more timing advance because of the lower air count. If you provided the ecu with the actual air flow that was coming it, it would be more stingy with timing advance.

We are saying the same thing. Semantics. If you tell the ECU that there is less air, you are richening the mixture not leaning it out. To lean it out means to add more air. Your statement is counterintuitive. Since the ECU sees less air, it advances timing.

Knock is your friend in tuning (especially in a 1g car because you can actually data log knock count). Yes you do use the SAFC to lean things out all over the rpm band and at the upper rpm's. You really can not lean it out too much to the point where you get too much timing advance.

I disagree. You can add enough fuel to where the computer is advancing significantly. Look at my a/f graph:

You must be logged in to view this image or video.

If you notice, there is flat spot below 10 a/fr where the tuner had to richen it that much due to severe knock at that rpm range. If the mixture was leaned out, it would knock even more. This is a no win situation here since with him richening the fuel, he is in essence advancing the timing (which would cause preignition). A Super ITC would fix this.

Some of this doesn't make sense to me. Why do you think your turbo is flowing too much air at 16psi? Have you logged how much air it's pushing, either karman HZ from the SAFC or in lbs/min off of a data logger? You say max fuel enrichment is -18%; what this means is that you are saying 18% less air is coming in. This will cause a leaner condiction, not richer. Do you have upgraded injectors at all. If not, I'm pretty sure you should be nowhere near -18%. What is your timing advance and why do you think it's too much. And are you getting knock, have you logged knock count to see what the actual values are.

If you have stock injectors and -18% on the SAFC, I would probably expect you to get lots of knock and no timing advance, probably would get timing pulled.

I have seen Karmen up to 2017Hz with teh Super AFC-II on stock injectors, k&n cone, and boost @16psi with 3" turboback. Those are all the mods I have. For some reason, the car is pushing a lot of air.

-M

 

[Blk_99gst]

I'm sorry but you still have the wrong idea on some of these topics. First off, what injectors are you running. Second, wow, yes 2000Hz on a 14b at 16psi is a lot. Are you sure it's a 14b turbo?

Originally posted by Phoenyx

We are saying the same thing. Semantics. If you tell the ECU that there is less air, you are richening the mixture not leaning it out. To lean it out means to add more air. Your statement is counterintuitive. Since the ECU sees less air, it advances timing.

I'm sorry but your still wrong here, and I think it's because you might not understand what the SAFC is doing. Here's what it's doing; (example just using fake numbers). Let's say you have no SAFC, and for every 1pound of air coming in the ecu provides 1pound of fuel for things to be tuned. Now you put a SAFC in, you adjust it to have a correction value of -50%; this is chaning the signal to the ecu so instead of a 1pound signal, the ecu will receive a .5pound incoming air signal. It sees that, knows that it has to provide .5pounds of gas to make things equal. But the problem is that you don't have .5pounds air coming in, you still have 1pound of true air coming in and the ecu only provide .5pounds of gas. Hence a LEAN condiction. So when you go into the negative correction values on the SAFC, you are LEANING out the mixture.

Also in the above situation, since the mixture will be so lean (1pound air to .5pounds gas) you will run super lean, and get knock, and the ecu will pull timing because of the knock; it will not advance timing.

Originally posted by Phoenyx

If you notice, there is flat spot below 10 a/fr where the tuner had to richen it that much due to severe knock at that rpm range. If the mixture was leaned out, it would knock even more. This is a no win situation here since with him richening the fuel, he is in essence advancing the timing (which would cause preignition). A Super ITC would fix this.
M

Yes, at that data point, you are running too lean and causing knock. By adding more fuel on the SAFC (adding more means going from a higher correction -40% to a lower correction -35%) you will get out of that lean condiction and the knock will go way with the more fuel. You don't have a graph of it, but if you logged timing advance and knock count, you would see timing get pulled at the 10 a/fr mark and knock count would go up at the 10 a/fr mark.

 

[Phoenyx]

I see your point. Question though. If I'm running lean, why is it that I'm way below 10 a/fr??

At this point, I'm very confused. My SAFC-II's graph is all in the negatives starting at a little above 1000rpms. So it looks like this:

X----------------------X
--X-------------------X
---X-----------------X
-----XXXXXXXXXX

This is how the picture looks for the whole rpm band. Doesn't this mean that I have too much fuel?? I have a 190 Walbro with STOCK injectors!

Ignore the -- lines. They are used as spacers.

-M

 

[Blk_99gst]

Wow, something is just not right. I don't know much about exact air fuel ratios so I looked it up:
http://www.sficc.net/features/feature4.html
". Typically cars make best power at about 12.5:1 air/fuel ratio"
See attached graph from that page for rich/lean condictions.

O.K your right, according to the graph you posted, if you are down in the 10a/f ratio, then you are running rich. To correct that problem you would have to lean out the safc more, i.e numbers from -20% to -25% (just used the twenty as an example). But there is NO WAY you can be running that rich and have the safc have negative correction values with STOCK injectors. The only explantion I can come up with is that you have to have bigger injectors in there like 550cc.

Did you buy the car completely stock or was it modded already when you bought it. If you, maybe you think you have stock injectors when you don't. And a walbro 190 pump doesn't come into play at all. It has to be fuel delivery through the injectors. I also question whether it's the 14b turbo in there.

The graph you drew of your safc means that you are leaning things out in the middle of the rpm band, and richening them back up never 6000 and 7000. Your right though that your safc graph resembles your a/f radio graph, except it's inverted.

One other way out there idea; who installed your SAFC? Maybe there's a slim chance that something got wired up wrong and inverted. So when the SAFC is at -18%, when that should be leaning things out, in some freak wrong wiring way that's actually telling the ecu that 18% MORE (instead off less) air is coming in, therefore it is providing 18% MORE fuel, giving you the horrible 10a/f radio of being extremely rich.

Here's my advice. If you belive you have stock injectors, then I would take the safc out of the equation. Either set all the correction values at zero which will act like it's not there, or disconnect power to it. This will let the ecu control fuel with an un-altered air signal and things should operate normal. You could verify this by doing another a/f radio dyno run or something. Is this making some sense? I've encountered a lot of situation before, but this is very very puzzeling. I'm heading off to sleep soon in half an hour or so... but there's always tomorrow or PM's.

 

[DougEFresh]

hey man you dont happen to have like a vortech sfmu on the car or something like a rising rate fuel pressure regulator cause as soon as you hit boost the a/f ratio hits the floor but if your correction values are all negative that makes no sense especially since the a/f ratio looked good up until the turbo spooled cant think of any other reason that the fuel flow would go through the roof like that under boost

 

[Phoenyx]

Originally posted by Blk_99gst
O.K your right, according to the graph you posted, if you are down in the 10a/f ratio, then you are running rich. To correct that problem you would have to lean out the safc more, i.e numbers from -20% to -25% (just used the twenty as an example). But there is NO WAY you can be running that rich and have the safc have negative correction values with STOCK injectors. The only explantion I can come up with is that you have to have bigger injectors in there like 550cc.

This car came from a dealer and was bone stock. The engine looks like new. I doubt it that the injectors are 550cc.

And a walbro 190 pump doesn't come into play at all. It has to be fuel delivery through the injectors. I also question whether it's the 14b turbo in there.

Yeap, its a 14b, but I disagree about the fuel pump. It will deliver more fuel to the fuel rail and the stock regulator may be choking where the extra pressure is coming out of the injectors.

The graph you drew of your safc means that you are leaning things out in the middle of the rpm band, and richening them back up never 6000 and 7000. Your right though that your safc graph resembles your a/f radio graph, except it's inverted.

Well at least we agree on something!

One other way out there idea; who installed your SAFC?

Nah, I checked the wiring diagram and there are only 2 wires that connect to the air flow sensor, the input and the output so they can't be mixed up (otherwise, it wouldn't work at all).

Here's my advice. If you belive you have stock injectors, then I would take the safc out of the equation. Either set all the correction values at zero which will act like it's not there, or disconnect power to it. This will let the ecu control fuel with an un-altered air signal and things should operate normal. You could verify this by doing another a/f radio dyno run or something. Is this making some sense? I've encountered a lot of situation before, but this is very very puzzeling. I'm heading off to sleep soon in half an hour or so... but there's always tomorrow or PM's.

Well, I went out today and check the AFC and I started doing some tests. I changed the TH-points so that they were further apart (20% 70%). FYI the low settings are all zero'd out. So I did a couple of punches and noticed a lot more lag and a lot of hesitation throughout the rpm band. I look at the my knock reading on SAFC-II and it was pretty high. I set the TH back and tried turning up the boost to 17psi. I got a LOT of detonation (car would have some backfires) and the Knock was reading like 217 (not knock counts)! EGTs were going up there too (1550). I turned the boost all the way down to stock and punched it.. Get this, I get creep and it goes all the way to 15psi (3" turboback exhaust). If I hold the throttle pedal to maintain 11psi, I get very little knock (47).

With the SAFC calibrated to be lean out the mixture, and the extra air coming, I have no choice but to believe that the car's timing is too advanced. I will put that ITC on and see if things change. I'll be getting a datalogger this Friday so I'll get some numbers to show soon...

Thanks for the help and if you have any other suggestions - shoot away!

-M

 

[HiFi TSi]

Dude, hate to burst your bubble, but a knock count of 47 is NOT low and will cause damage to your motor in no time running it like that!! I can't imagine you have any timing advance with a count that high so why would you want retard it when there's none left to lower??? You should ideally tune for zero knock, but up to 6-7 is tollerable. Not to be rude, but I think you should get some help from a local dsmer, or go back to that shop.


On a side note for the question about AFR. 14.7:1 is the stochiometric point where 100% of the fuel mixes with the oxygen available. A lower ratio obviously means you're running rich and vice versa means you're running lean. On pump gas it's safe to tune for 11.5:1 and on race gas from 12:1 - 12.5:1. This can't be accomplished on your daily driver though unless you have a wideband o2 measuring your AFR's.

 

[DougEFresh]

hey man he specifically said that he wasnt refering to knock in counts im assuming he was refering to the safc's knock reading im not too familiar with the safc II so i dont know what units it measures in but yeah obviously if it was the knock count his motor would be toast and the whole backfire thing sound like the motor is way to rich to me just out of curiousity what is the setup under car and sensor setup,#of cylinders, sensor type, throttle angle. oh and how are the wires connected to the ecm wires are they sodered, split and twisted or are they hooked up with scotch locks

 

[HiFi TSi]

I didn't realize he was using the safcII. I'm no expert on that device, but I'm assuming it would read knock right from the ecu. It wouldn't be of any use if it measured knock in some fancy fashion that you have to decipher to know your knock count, but then again, who knows! My suggestion would be to go back to the settings from the shop, I hope you wrote them down, and leave them until you get a real datalogger. Assuming the shop is reputable, the setup had to be fairly decent, so you should be good until you can datalog and see what the car is doing.

 

[Phoenyx]

Originally posted by HiFi TSi
Dude, hate to burst your bubble, but a knock count of 47 is NOT low and will cause damage to your motor in no time running it like that!! I can't imagine you have any timing advance with a count that high so why would you want retard it when there's none left to lower??? You should ideally tune for zero knock, but up to 6-7 is tollerable. Not to be rude, but I think you should get some help from a local dsmer, or go back to that shop.

Knock is not measured by counts on the SAFC-II but by percentages. I don't have a datalogger yet.

-M

 

[chinchin]

HAVE YOU CHECKED YOUR BASE TIMING?

 

[Phoenyx]

Originally posted by chinchin
HAVE YOU CHECKED YOUR BASE TIMING?

Actually No. I'm doing that this weekend!

Thanks,

-M

 

[kpt4321]

Originally posted by HiFi TSi
I didn't realize he was using the safcII. I'm no expert on that device, but I'm assuming it would read knock right from the ecu. It wouldn't be of any use if it measured knock in some fancy fashion that you have to decipher to know your knock count, but then again, who knows! My suggestion would be to go back to the settings from the shop, I hope you wrote them down, and leave them until you get a real datalogger. Assuming the shop is reputable, the setup had to be fairly decent, so you should be good until you can datalog and see what the car is doing.

If you don't know anything about the setup, and are just going to make assumptions, then perhaps you should just remain silient.

The SAFCII just intercepts the signal from the knock sensor (you tap the wire) and does it's own math to come up with an arbitrary number. What each SAFCII value correlates to in terms of knock sum, I do not know.

 

[kpt4321]

Now, something is SERIOUSLY out of whack here. First of all, I think you NEED to consider pressure testing your intake. The fact that you are pulling 2100 Hz of air at 15 psi, and the fact that you are running hella rich while in boost with the SAFC in the negatives on stock injectors would indicate to me that you have a LARGE boost leak.

Have you realized how a SAFC works at this point?

Negatives = less airflow seen by ECU = less fuel = leaner = higher A/F ratio = more likely to knock = more timing advance

Positives = more airflow seen by ECU = more fuel = richer = lower A/F ratio = less likely to knock = less timing advance.

 

[HiFi TSi]

Originally posted by kpt4321


If you don't know anything about the setup, and are just going to make assumptions, then perhaps you should just remain silient.

Perhaps I made the assumption because I didn't realize he was using the II in the first place. You're right, maybe next time I'll keep my mouth shut instead of offering advice, since you seem to know it all. If you're only going to post to rip on someone which isn't relative to the post, leave it to yourself.

 

[Phoenyx]

Originally posted by kpt4321
Now, something is SERIOUSLY out of whack here. First of all, I think you NEED to consider pressure testing your intake. The fact that you are pulling 2100 Hz of air at 15 psi, and the fact that you are running hella rich while in boost with the SAFC in the negatives on stock injectors would indicate to me that you have a LARGE boost leak.

Have you realized how a SAFC works at this point?

Negatives = less airflow seen by ECU = less fuel = leaner = higher A/F ratio = more likely to knock = more timing advance

Positives = more airflow seen by ECU = more fuel = richer = lower A/F ratio = less likely to knock = less timing advance.

My car creeps to 17psi and I've seen at best 2073Hz. Since my car peaks at 18-19psi sometimes before falling back, I'm not sure that number is abnormal.

No boost leaks though, If I close the boost controller all the way - I'll get the proposed 11psi until higher rpms where the car will start to creep. A boost leak (especially in the intake) would register right away and I wouldn't see much boost (since my gauge is hooked into the manifold).

Next theory?

-M

 

[kutschca]

a boost leak will not always show up as a reduction in your manifold pressure, the turbo could just be pushing harder to overcome it. This would give you hotter air, and you'd be more prone to knock. Here is a site for making an intake leak tester http://www.vfaq.com/mods/ICtester.html that very well might not be your problem, but it couldn't hurt to narrow it out for sure.

 

[kpt4321]

"kutschca" is right. Say a turbo can flow like 500 cfm of air, and you engine requires 300 cfm. Even if you have an ENORMOUS boost leak of 150 cfm (which is bigger than you would believe) then you will STILL be able to get full boost in the manifold.

The boost controller would just require that the turbo work harder in order to generate the flow necessary to both pressurize the manifold to the set pressure, AND to flow enough air to satisfy the leak.

A boost leak would also explain your SAFC settings, your rich dyno A/F ratio, and you knock problem. The leak would cause the ECU to give you way more fuel then you really need, causing the rich condition. This would necessitate negative corection factors on the SAFC, like you have. This would also increase the temperature of the air, increasing the propensity for knock.

Pressure test it!

 

[Phoenyx]

Ok, Update!

I got the ITC install and it's working great! I've also got my pocketlogger in the mail and have it hooked up! The mechanic who installed my ITC said my base timing was off by 3 degrees (it was around 8 degrees). He put that back. I did a quick datalog of the car and:

RPM: 4031
Knock: 14
O2: 0.92
Inj Duty Cycle: 55%
Air Volume: 5
Air Temp: 109
TPS: 93%
Fuel Trim High: 100%
Fuel Trim Mid: 100%
Fuel Trim Low: 110%

I'll take it to redline tonight and see what I get. Should've logged timing...Darn