Fixed! Off idle lean spot

[flea]

After LTH and CAI install dialing in WOT and part throttle AFR was relatively fast and painless (finished this months ago). Since the mods I have had a strange 1-3 sec lean spot (15.2-16 AFR) on part throttle off-idle starts. After chasing my tail with VE, MAF, transient, etc. (countless hours) the fix appears to be the cylinder charge temp filter table. Apparently it was not moving to the 16 g/sec bias temp quickly enough. Averaging the 16 and 32 g/sec values to arrive at a new 16 g/sec value ended up being the "sweet spot".

Frustrating that seemingly minor tweaks can sometimes solve big problems.

Gotta love the E40 LS2's

[Blades]

Would this also play into cars running MAF only (dyn airflow disabled)?

[flea]

Based on my experience, yes. I should have provided some additional details. I was having this problem in CL or OL MAF (with or without dyn air disabled). After many hours of data logging I'm also not convinced that dyn air can be entirely disabled with the E40 LS2s. My perception is you can dramatically minimize VE input but it is still referenced to some extent. This is with the enable/disable set below the idle rpm.

[Blades]

Yeah, probably because transient fuel is still active, and that refs VE. I'm having a problem right now with my air/fuel just being "tweaky".. I know its not a wideband issue since the readings are not tweaked out when I shut the car off and its still reading the gases inside the exhaust. .Perhaps its an exhaust leak, but I have a pretty tight band clamp where it was leaking before. Could be a misfire as I have my plugs gapped at .055.. or rather .054-.053. The right hand side (end) of the 'tip' of the spark plug is lined up with .055. Its hard to tell the difference between false and true misfires with a cam and stall.

How can I test for this problem? Would it change with IAT when in park or ECT? Probably ECT, correct? If I sit and let the IAT 'bake' and then coast off and start driving, the readings should be rich because the bias table is now looking at a very hot (and untrue) IAT.

[flea]

heat soaked IATs = lean

We have GM to thank for this. Maybe the rational was "pull timing and fuel" to safeguard against detonation with high charge/IAT temps? This is valid up to a point however GM seems to be far to conservative.

[flea]

Too quick with the submit button. AFR stability is the basis for the bias table post. The OEM GM bias table calibrations result in AFRs drifting all over the place (high or low) in OL. I don't have a problem with AFR going a little lean as IATs increase however drifting from 14.7 to 15.3 when the IAT increases 10 degrees is crazy.

[Blades]

Thats what mine does. However, wouldn't it make the car richer? Because its adding fuel and pulling timing? And lower IATs would be..... Are you saying they were under-careful. That'd be a first for a stock gm tune.

[EC_Tune]

Flea, can you post your before & after tweak to make the lean spot go away? Just the table before & after would be enough.

[flea]

I'm still scratching my head over the dramatic difference. I have hours of data before and after at a wide range of IAT temps (90 - 130) and it's still working like a charm. If it works it works ...........

Data log (before change) - afr was ~stoich before spike. high point of spike afr was ~16. It required almost 3 seconds for afr to drop below 15.

Data log (after change) - afr was ~stoich before spike. high point of spike afr was ~15.4. Time to drop below 14.8 was less than one second.

[flea]

Thats what mine does. However, wouldn't it make the car richer? Because its adding fuel and pulling timing? And lower IATs would be..... Are you saying they were under-careful. That'd be a first for a stock gm tune.

Blades,

I'm not sure what to make of any of this right now. All I can tell you is my car will begin to lean out as the IATs increase (with the OEM bias/filter settings) and will maintain a more lean afr (15 -15.3) at steady state or part throttle until the temps come back down.

[0rion]

subscribing

[Blades]

I changed my VE table by failing the MAF and adjusting with my wideband. The stock table w/ my cam was so god damn rich it would barely run.. So then I decided to see if the changes would show up in my MAF only tune.. And yes.. they did.. went from lean idle to very lean idle.. Odd how the car will do 12:1 AFR at idle maf is failed and 16.2 AFR when maf is not.. Scary lean. No happy medium for me.. heh.

However, this tells me that VE is still in play.

Wouldn't your lean spike actually show up as an odd looking change in the inj pulse width?

My lean idle issues are just stupid.. Could it be airflow final min set wrong? I need some better sleep if I am going to tackle this stuff new bias table stuff. Good luck man.

[BBA]

My tables were totally different value wise, but I did move the 8, 16 and 24 g/sec up somewhere mid between the initial and the next and it did make the hot running start from idle dead pedal problem go away.

Man...I wish I had a wideband now! (even though HPT support for VE tables sucks on 2007 TBSS).

[redhardsupra]

this is some good finds guys!

does anyone know what the numbers in the filter table mean, or at least what units they're in? i cant find anything about how exactly it's taken into establishing which bias table value to use when.

[BBA]

Heck, I'm still trying to figure out exactly why this makes a difference. I know it has to do with changes in charge temperature on a throttle opening event, maybe the cold rush of air from opening throttle whacks out the charge calculations from airflow and this makes up for it?

[flea]

this is some good finds guys!

does anyone know what the numbers in the filter table mean, or at least what units they're in? i cant find anything about how exactly it's taken into establishing which bias table value to use when.

Just speculation on my part based on lots of data logging and trying to identify patterns. The filter coefficients appear to be nothing more than the rate of change (in seconds) from one bias setting to another. With that said we don't currently have much to work with in terms of trying to narrow this down. Could be this simple or much more convoluted .............

After trying your initial suggestions (not knowing at the time they were based on readings you took from a GTO) I have gone even more in the direction of tweaking the filter coefficients while leaving the OEM bias settings alone. So far the results seem to be much more dramatic and immediate. At this point it's still almost entirely based on trial and error to determine the effects. I just kind of stumbled on to this but it really does seem to work if someone is willing to spend some time with it.

The only thing that comes to mind with this so far is

"Not everything that counts can be counted and not everything that can be counted counts" -Einstein

[flea]

I meant to add the following.

Increasing the rate appears to minimize undesirable transient behavior (e.g. 3-4 second part throttle lean spikes). Although I wouldn't consider 3-4 seconds a "spike".

[Cyclone Chris]

I'm gonna have to give this a try ....

Since we have had 110* days, my car has felt sluggish. Basically from a dead stop I start to give it gas, and the car barely wants to move. So I have to reset and give it more RPMs (gas) before I let out the clutch. Kinda fustrating when you're in an intersection.

[flea]

Chris,

I know what you mean in terms of sluggish. Had to drive to Chicago over the weekend in 95 plus weather and recorded some of highest IAT's I've ever seen on my car (~150 F). I have already tweaked my IAT base spark a number of times but this has motivated me to spend some more time with this. Particularly the rpm based multipliers.

Side note: what's up a 60 year old man (with wife in the car) in a C5 literally going out of his way not to let a GOAT merge into interstate traffic. You could tell his wife was absolutely embarrassed. If the traffic had not been bumper to bumper it would have been fun to stomp him. I'm not sure what's wrong with people sometimes.

[Blades]

this is some good finds guys!

does anyone know what the numbers in the filter table mean, or at least what units they're in? i cant find anything about how exactly it's taken into establishing which bias table value to use when.

Maybe it might make more sense when looking at a TBSS bias table, as the Y value is in mph.. The filter could be the values for how interpolates. The description refers to it as a rate. It has to be some sort of ratio or the result of a ratio. I have no clue, its melting my mind.

There is also a map filter in transient fueling.. heh, same units. We need filtered MAP and filtered airmass PIDs.