Strange AFR at idle

[Cybnext]

I tuned my first LS2 tonight, a 2006 GTO m6 w/ long tubes, off road mid, and a torquer v2 cam. After getting the VE and MAF as close as I could for tonight (rain came in), I was checking everything over and noticed that the car would almost oscillate the AFR. It starts at 14.7 and would slowly makes its way up to ~19:1 and then slowly start to makes its way back to 14.7. I put the car back to CLMAF and it did the same thing, but didn't go nearly as lean, probably only went to ~16:1. Idle stayed pretty much solid at 850.

Sadly, I don't have a log file of this happening. I thought I had saved it, but apparently not.

I'll attach the tune though.

[BBA]

You did not fix your spark tables to a static value at low RPMs.
You have to have stable spark advance to get stable idle RPM, and you have to have that before your AFR at idle will work right.

[Cybnext]

Spark is one of the things I haven't really figured out how to tune it. I've searched and searched but can't find anything that really helps me understand what I'm supposed to do in the spark tables.

I'm sure I could explain my problem better, but basically I just haven't really figured out how to tune spark.

Are the spark tables "divided" into sections sort of the like VE table? Obviously this would be more of a virtual thing, just so I could see where the "idle, cruise, WOT, decel, etc" parts are.

[BBA]

There are three spark tables to worry about: High Octane, Low Octane and base timing.

The tables cover timing in RPM bands and each band is scaled by air flow as sensed by the MAF. The stock tables have negative timing at high airflows and low RPMs. I have no idea why other than some engineer figured you would never have low RPM high airflow so it wouldn't matter...but what it does is kills WOT take off power because you actually hit the high airflow tables on a trasnsition from idle to WOT due to the air surge into the large volume intake manifold when the throttle blades open.

The car idles with base table, but if the high table has a lower value, it will take the lower value. If knock is sensed, it will proportionally reduce timing to a value between the high and low octane tables until KR goes away, this way it can handle different grades of gas without major problems.

In the case of a cam with a lope to it, the idle airflow goes all over the place really freaking out the timing tables and causing engine timing to go all over the place resulting in an extremely unstable idle. This causes MAP pressure to be unstable and really freaks out everything the PCM is trying to control.

[01 MidMet WS6]

Cybnext, I just finished a tune on an identical setup, the only difference being mine is an A4. You do need to change your timing tables a lot, copy over my high, low and both idle spark timing tables. Yours all appear to be stock and this will help a lot with idle surging and overall power output. Also copy over my iat spark advance correction table. Drive the car, see how well it works out for you and please post up a log after your test drive.

[Cybnext]

There are three spark tables to worry about: High Octane, Low Octane and base timing.

The tables cover timing in RPM bands and each band is scaled by air flow as sensed by the MAF. The stock tables have negative timing at high airflows and low RPMs. I have no idea why other than some engineer figured you would never have low RPM high airflow so it wouldn't matter...but what it does is kills WOT take off power because you actually hit the high airflow tables on a trasnsition from idle to WOT due to the air surge into the large volume intake manifold when the throttle blades open.

The car idles with base table, but if the high table has a lower value, it will take the lower value. If knock is sensed, it will proportionally reduce timing to a value between the high and low octane tables until KR goes away, this way it can handle different grades of gas without major problems.

In the case of a cam with a lope to it, the idle airflow goes all over the place really freaking out the timing tables and causing engine timing to go all over the place resulting in an extremely unstable idle. This causes MAP pressure to be unstable and really freaks out everything the PCM is trying to control.

So should I basically use the VCM Controls and use the "Plus" part of the spark control and add until the map reaches its lowest value, and then add that to the entire idle base map? and to the HO map in the idle areas?

Is there a way to "zone" out spark tables to visually see where the car idles, part throttle, accel, WOT, etc?

[ShoddyHog]

Did it do this shortly after startup, or after idling for a while? It took me forever to find it, but I did find an example. I have an 06 M6 GTO with mods in my sig, and have seen the same behavior. I've only seen it do that right after startup, and have no idea what caused it.

Gerry

[ShoddyHog]

Is there a way to "zone" out spark tables to visually see where the car idles, part throttle, accel, WOT, etc?

When I wanted to see where the timing tables were hit at idle, cruising, or WOT, I'd load a log, scroll to the place on my chart display where I could see the WOT pull (for instance), and with hystogram #1 open (Main Spark - Timing Advance) I would just drag my cursor across the WOT area, and see where the hystogram was being hit.

Don't know if there's a better way or not. I'm a newb tuner.

Gerry

[Cybnext]

That looks fairly similar. Looks like you started driving right after. We sat there for a bit and let it go back and forth from ~stoich to lean and back again a few times.

[ShoddyHog]

I've seen that a couple times...I assumed it was something to to with my wideband, but what do I know? I should add that I've found that assumption to be wrong, otherwise I should be able to reproduce the problem at will. I can't.

Here's an example of how I map out my WOT runs. I just right click on the hystogram and choose "clear all data" then drag my cursor across where I want to see what timing cells are hit.

Gerry

[Cybnext]

So should I basically use the VCM Controls and use the "Plus" part of the spark control and add until the map reaches its lowest value, and then add that to the entire idle base map? and to the HO map in the idle areas?

Is there a way to "zone" out spark tables to visually see where the car idles, part throttle, accel, WOT, etc?

Anyone verify this?

[5_Liter_Eater]

For you guys seeing lean AFR on cold starts do you still have the AIR pump hooked up? The air it pumps into the exuahst to reduce emissions creates an artificial lean reading. You can turn it off in VCM Controls.

[5_Liter_Eater]

Anyone verify this?

This method works in theory but you will find that the MAP does not vary much with large timing changes. For WOT tuning it's common practice to add timing until KR starts, then back off a little. For idle timing it's best to find the ideal timing and then back off by ~25% to allow for idle underspeed timing to be added. Mid-range timing tuning...no idea.

[ShoddyHog]

For you guys seeing lean AFR on cold starts do you still have the AIR pump hooked up? The air it pumps into the exuahst to reduce emissions creates an artificial lean reading. You can turn it off in VCM Controls.

I'm assuming I do unless it was removed during my header or cam installation by the person who did the install. Here's a snapshot of my cold start this morning...it was what I consider to be a perfect start for my cammed goat.

WHY did GM make it so that a few thousand tables are used to control startging?

Gerry

[5_Liter_Eater]

Are there smallish tubes going to the top of your headers? If so you still have the air pump.

Well next time you're doing a cold start, if it goes lean, go into VCM Controls and turn off the Air Pump and see if it comes back to where it should be.

[ShoddyHog]

Are there smallish tubes going to the top of your headers? If so you still have the air pump.

Well next time you're doing a cold start, if it goes lean, go into VCM Controls and turn off the Air Pump and see if it comes back to where it should be.

Nope...they are standard Kooks long tubes. I'll have to look at my exhaust manifolds too. I don't remember seeing any tubes going to those either. I'll check when I get home and will let you know.

One "problem" I have on startup is that no matter if it is cold, warm, or hot start...the exhaust is pig rich by smell, yet my wideband reports stoich or damn close to it. After running for a bit though, the idle fumes smell like stoich.

Gerry

[ShoddyHog]

No fittings or anything going to the stock manifolds....of course, once I thought about it, I read ls2gto.com tech forums all the time, and I would have known if there was an air pump involved with header installation. I guess the GTO just doesn't have one.

Gerry

[BBA]

I don't think any LS2 has an air pump.

I did my tables like this and it works very well (Oh, the reason my base table has a lower values in the mid flow region is to allow P/N timing adders to control idle, large cam motors don't need the lower values in the low rpm mid flow regions on the base timing table):

[BBA]

You might think it looks funny, but every value is targetted to work via experimentation. The reason I go up in timing at low rpm's and high flows is to keep timing from dying off when I stomp it from idle...and this works very well.

As I said before, setting the base timing high at low RPM's is not desirable at normal idle air flow values, because it will cause idle instability.

Large cam motors do respond well with even more timing in the low RPM cells, one LS2 with a 228/230 110 cam I tuned even needed all the low RPM cells set over 30º, anything less and it wouldn't idle for nothing and wouldn't hot start worth a darn either.

[ShoddyHog]

What are those timing tables for? Your Base idle table doesn't look like mine. Mine only have 400-1600 rpm.

Gerry