LGT AVCS Tuning Discussion

[LittleBlueGT]

Not really. LBGT meant air-fuel mixture going out. The amount of excess fuel relative to oxygen content is still the same.

 

I fully agree with this Mike, and all my testing agrees with it as well.

[BlackHole]

Not really. LBGT meant air-fuel mixture going out. The amount of excess fuel relative to oxygen content is still the same.

 

I'm curious how this would be the case - if the air-fuel mixture is passing through and not combusting, or there are flow reversion issues due to too much cam overlap creating excessive VE / scavenging, then the AFR would go lean since the amount of excess fuel relative to oxygen content would be very different. I've searched around online and haven't found much helpful in educating me otherwise.

 

From my data so far: at +30 AVCS, stock timing & AFR maps at idle the AFR is 0.75 leaner than at 0 AVCS, same everywhere else. But the calculated VE actually is lower - 50 vs 51%. At +30, the car idles like it has big cams (go figure ).

[mickeyd2005]

If the MAF reads 100 g/sec and the fuel injected is appropriate for 100 g/sec then the percentage of excess fuel in the exhaust is the same.

 

 

It doesn't matter if it was oxidized or not. Where did the excess fuel go?

 

The O2 sensor oxidizes the excess fuel at the tip.

[BlackHole]

It doesn't matter if it was oxidized or not. Where did the excess fuel go?

The O2 sensor oxidizes the excess fuel at the tip.

 

I would disagree with this - the O2 sensor does not oxidize the fuel (generally there isn't appreciable combustion outside the cylinder). The fuel hasn't gone anywhere, it's just that the O2 sensor doesn't measure it. The O2 will read a lean environment since there is a lot of available O2 due to the reduced scavenging (lack of combustion).

 

For example, this is a common problem on large cammed NT motors:

http://www.thirdgen.org/techboard/diy-prom/310130-how-fix-some-those.html

post#4: http://www.hotrodders.com/forum/rich-cruise-lean-wot-142173.html

speed density: http://www.gmhightechperformance.com/tech/0704gm_efi_live/index.html

[mickeyd2005]

I would disagree with this - the O2 sensor does not oxidize the fuel (generally there isn't appreciable combustion outside the cylinder). The fuel hasn't gone anywhere, it's just that the O2 sensor doesn't measure it. The O2 will read a lean environment since there is a lot of available O2 due to the reduced scavenging (lack of combustion).

 

For example, this is a common problem on large cammed NT motors:

http://www.thirdgen.org/techboard/diy-prom/310130-how-fix-some-those.html

post#4: http://www.hotrodders.com/forum/rich-cruise-lean-wot-142173.html

speed density: http://www.gmhightechperformance.com/tech/0704gm_efi_live/index.html

 

How do you think a WBO2 sensor reads AFR lambda < 1?

[mickeyd2005]

Why don't you post your logs and your AFR table.

 

I think the lower VE just moved your cell to the left to a leaner target AFR.

[LittleBlueGT]

Why don't you post your logs and your AFR table.

 

I think the lower VE just moved your cell to the left to a leaner target AFR.

 

Agreed!

[BlackHole]

How do you think a WBO2 sensor reads AFR lambda < 1?

 

The O2 sensor can only read free O2. Even under rich conditions the fuel has been combusted and the inherent oxygen freed and scavenged.

 

From http://wbo2.com/lsu/lsuworks.htm, which will also apply to our 4-wire planar sensors.

 

"When using a 5-wire sensor we make certain assumptions about the environment the sensor is used in, for example we assume the sensor is used to measure the exhaust byproducts of fairly complete combustion. The combustion can be internal, as in a conventional vehicle or external combustion, as in a furnace or other device consuming fuel and oxygen. If these conditions are altered then inappropriate readings may result. For example, if a miss-fire occurs and unburnt droplets of fuel go through the motor, then the sensor will read lean as it will not detect the liquid fuel. When tuning a vehicle, you should not rely solely on the reading from a sensor. Let your common sense, and a little knowledge of how the sensor works, guide you."

[BlackHole]

Why don't you post your logs and your AFR table.

 

I think the lower VE just moved your cell to the left to a leaner target AFR.

 

On the LGT I wish I could specify AFRs this low. As stated before the AFR (target = 14.7) and ignition tables are stock. I'm confident that AVCS +30 is enough overlap at idle that some of the air-fuel mixture is passing right through the combustion chamber w/o being ignited, thus causing the O2 to indicate a lean condition.

 

For AVCS = +30, 800rpm, 0.3 load cell AFR avg = 16.04 (1300 data points)

For AVCS = 0, 800rpm, 0.3 load cell AFR avg = 15.08 (4100 data points)

 

What I'm still not sure about is why my VE calculations aren't biased high. It could be (hopefully is) that the VE drops because of the drop in intake vacuum, which would increase the theoretical amount of air that should be flowing, even though the actual amount of air hasn't changed.

 

I can't post the logs due to their size (3MB each) - PM me privately if you want a copy.

[mickeyd2005]

The O2 sensor can only read free O2. Even under rich conditions the fuel has been combusted and the inherent oxygen freed and scavenged.

 

From http://wbo2.com/lsu/lsuworks.htm, which will also apply to our 4-wire planar sensors.

 

"When using a 5-wire sensor we make certain assumptions about the environment the sensor is used in, for example we assume the sensor is used to measure the exhaust byproducts of fairly complete combustion. The combustion can be internal, as in a conventional vehicle or external combustion, as in a furnace or other device consuming fuel and oxygen. If these conditions are altered then inappropriate readings may result. For example, if a miss-fire occurs and unburnt droplets of fuel go through the motor, then the sensor will read lean as it will not detect the liquid fuel. When tuning a vehicle, you should not rely solely on the reading from a sensor. Let your common sense, and a little knowledge of how the sensor works, guide you."

 

You should think about how a WBO2 measures lambda < 1. It pumps in air until the fuel in the mixture is consumed. The fuel is oxidized on the sensor's catalytic surface. It's clearly spelled out in your link and yet you wrote this.

 

I would disagree with this - the O2 sensor does not oxidize the fuel (generally there isn't appreciable combustion outside the cylinder). The fuel hasn't gone anywhere, it's just that the O2 sensor doesn't measure it. The O2 will read a lean environment since there is a lot of available O2 due to the reduced scavenging (lack of combustion).

[mickeyd2005]

On the LGT I wish I could specify AFRs this low. As stated before the AFR (target = 14.7) and ignition tables are stock. I'm confident that AVCS +30 is enough overlap at idle that some of the air-fuel mixture is passing right through the combustion chamber w/o being ignited, thus causing the O2 to indicate a lean condition.

 

For AVCS = +30, 800rpm, 0.3 load cell AFR avg = 16.04 (1300 data points)

For AVCS = 0, 800rpm, 0.3 load cell AFR avg = 15.08 (4100 data points)

 

What I'm still not sure about is why my VE calculations aren't biased high. It could be (hopefully is) that the VE drops because of the drop in intake vacuum, which would increase the theoretical amount of air that should be flowing, even though the actual amount of air hasn't changed.

 

I can't post the logs due to their size (3MB each) - PM me privately if you want a copy.

 

If you were truly in closed loop mode, how did you hit AFR 16.04. If your front O2 sensor was active, your AFR should have been 14.7 and your AFR corrections should have been + 10%. That would have indicated the problem. If you're hitting AFR 16.07 then something else is wrong.

 

And in any case, I haven't seen any problems with AFR in WOT logs. I run 30 degrees AVCS and when the ecu is reset, it's about 8 degrees constant.

 

BTW, I did not know you were talking about closed loop stoichiometric. For that, it is very easy to check. First change AVCS to all 0 in the cruise range and then drive around until AFR learning is set. Then change AVCS to all 30 in the cruise range and then drive around until AFR learning is set. Compare AFR learning A, B, C, and D using learning view.

 

Lots of people have already unwittingly made this mod. The stock 2005 LGT rom has a "hole" in the AVCS map in the cruise range. The Cobb AP stage 1 map fills in that hole. I don't recall seeing different AFR learning values from people running Cobb maps.

 

I still don't know how you got AFR 16.04 in closed loop.

[BlackHole]

If you were truly in closed loop mode, how did you hit AFR 16.04.

 

Bear with me - for the sake of discussion accept the premise that the O2 sensor does not read raw, uncombusted fuel thus reads leaner than the air-fuel mix really is. And that the valve overlap is long enough that some of the raw air-fuel charge is passing through to the exhaust w/o being ignited.

 

Under this set of conditions, no matter how much fuel the ECM may add the engine will always report a lean AFR since part of the air-fuel mix is still liquid and won't be measured. Or another way of looking at it is that w/o combustion the raw fuel cannot scavenge O2, thus the WBO2 reports that there is more O2 (leaner) in the exhaust stream. I believe that this is what happened. According to RomRaider, the engine was in closed loop (CL/OL status = 8) for all data points.

 

You should think about how a WBO2 measures lambda < 1. It pumps in air until the fuel in the mixture is consumed. The fuel is oxidized on the sensor's catalytic surface. It's clearly spelled out in your link and yet you wrote this.

 

You should re-read the page and think a bit more. As stated on the top of the page and I quoted - the WBO2 catalytic reaction does NOT work for fuel in liquid form. If the air-fuel mixture is not ignited, it is still a vapor - liquid fuel droplets in air. The chemical reactions of combustion converting fuel + O2 into CO and H2O (which are not measured by the WBO2) are not happening, leaving excessive / unreacted O2 for the WBO2 to measure thus a "lean" condition.

 

Sorry for being OT, we now send you back to your regularly scheduled AVCS discussion.

[Infamous1]

I enjoy all types of discussion, plus it is somewhat still on topic....

[Guest]

Question is, if one submerges WBO2 sensor in gasoline what will it read!

[mickeyd2005]

Why are you saying stuff like liquid fuel? We're talking air-fuel mixture.

 

I know exactly how a WBO2 works. It pumps air until all the fuel is consumed.

 

Why don't you tell us how you read AFR 16.04.

[NSFW]

It should be possible to answer (and quantify) the valve overlap question mathematically... does anyone know, or know where to find, at what angles the intake and exhaust valves open and close?

[BlackHole]

Why are you saying stuff like liquid fuel? We're talking air-fuel mixture.

 

I know exactly how a WBO2 works. It pumps air until all the fuel is consumed.

 

Why don't you tell us how you read AFR 16.04.

 

mickeyd - I respect a ton the work you've done and your level of tuning expertise is generally far above mine on Subies. I hope your comments were not meant to be hostile but they could be interpreted that way. I understand how O2 sensors work, but proper operation is dependent on the state of the air-fuel mix being measured. With all respect, let's leave it at we agree to disagree.

 

I am open to other ideas and supported arguments on how my car reads 16.04 AFR under closed loop. I have a ton of data that supports the hypothesis that the raw air-fuel mix is blowing through the cylinder due to excessive cam overlap and causing the O2 sensor to read artificially lean. This effect is repeatable and I have posted evidence to support that this is a known phenomena.

[NSFW]

If you were truly in closed loop mode, how did you hit AFR 16.04. If your front O2 sensor was active, your AFR should have been 14.7 and your AFR corrections should have been + 10%. That would have indicated the problem. If you're hitting AFR 16.07 then something else is wrong.

 

14.7 + 10% = 16.17

 

BlackHole, does this phenomenon occur when you let off the throttle with high AVCS, or at WOT? Have you logged the fueling mode when repeating this phenomenon?

[LittleBlueGT]

mickeyd - I respect a ton the work you've done and your level of tuning expertise is generally far above mine on Subies. I hope your comments were not meant to be hostile but they could be interpreted that way. I understand how O2 sensors work, but proper operation is dependent on the state of the air-fuel mix being measured. With all respect, let's leave it at we agree to disagree.

 

I am open to other ideas and supported arguments on how my car reads 16.04 AFR under closed loop. I have a ton of data that supports the hypothesis that the raw air-fuel mix is blowing through the cylinder due to excessive cam overlap and causing the O2 sensor to read artificially lean. This effect is repeatable and I have posted evidence to support that this is a known phenomena.

 

 

Their is only one right answer, and the majority of us (MickeyD2005) included) want the right answer.

 

Post logs!!!!!!!!!!!!!!!!

 

Only way to start some investigation!

[BlackHole]

Their is only one right answer, and the majority of us (MickeyD2005) included) want the right answer.

 

Post logs!!!!!!!!!!!!!!!!

 

Only way to start some investigation!

 

As stated previously - PM me your email addys and I'll send them. They're too big to post (14MB total).

 

Data update: now have 3 sets of DOE data complete with the only variable being the AVCS timing (stock fuel and ignition) for the 800rpm, 0.3 g/rev load cell:

 

AVCS=0 AFR=15.08

AVCS=15 AFR=15.07

AVCS=30 AFR=15.68 (came back up a little)

[PeterJMC]

Question. So does the LGT only allow for the modifying of Intake AVCS timing and not Exhaust AVCS timing?

 

Reading this article... '08 STis are capable of exhaust AVCS tuning.

http://www.drivingsports.com/site/2008/11/modifying-the-subaru-sti-avcs/

[Guest]

Yes, USDM LGT unlike JDM has intake AVCS only.

[mwiener2]

Now to throw out everything you know....

 

What does my higher compression ratio want me to do with AVCS?

[Infamous1]

Now to throw out everything you know....

 

What does my higher compression ratio want me to do with AVCS?

How much higher compression are you running?

[mwiener2]

1 point higher. 9.2:1