lakeffect2

Looking at the EBL What's Up Display, at idle in closed loop, I'm watching the O2 millivolts go through wild variation from 12-804-6-752.

I understand that I should see a fliipping from high/low as it crosses it's boundary, but can cause such erratic shifts> It seems unusual.

Dave

luvofjah

iTrader: (2)

I had a similar problem...
Some of the bolts on the tpi were lose, vacuum leak...
After that, purred like a kitten

Rafael

ULTM8Z

If it idles nicely otherwise, I wouldn't fret over it too much. Mine was doing similar things with the O2 voltage. I ended up going to open loop on idle recently because the improvement it gave in the off idle transition.

ULTM8Z

Btw... with your 406 and that cam, what is your idle vacuum?

lakeffect2

The idle vacuum is usually at 12, sometimes 11

ULTM8Z

Is that in gear or in neutral?

Street Lethal

iTrader: (16)

Dave, the first thing you want to make sure of is that your O2 signal wire to the ECM isn't routed near any ignition wires. Second, the O2 measures the amount of oxygen in the exhaust so if the swing seems wider than normal then you need to continue dialing in your VE. At 12" idle vacuum, this tells us you are running a lot of duration, so the overlap plays its' part as well. The VE Learns will get it close, but with cams leading more towards the radical side, you may need to go into the VE and adjust the idle area in the surrounding cells...

- Rob

Street Lethal

iTrader: (16)

Also, the difference between 12 & 6 as well as 804 & 752, if this is what you are questioning, is because no two intake ports flow the exact same amount of cfm, and no two injectors flow the exact same amount of fuel, despite being the same size. You will see a minor variation such as that, it is normal. The idea is to get the sweep as close to 450 on both ends as possible, doesn't matter what numbers they read so long as your air/fuel is where it needs to be...

- Rob

lakeffect2

Rob, I guess my point is seeing a swing from 400 to 500 would tell me I'm in good shape for VE table. Even 350-550 would be nice to see. But seeing it swing from single digits to 750-800 like a pendulum back and forth tells me It has issues, and I am no where near where I should be. Could be there is a secondary problem I should be hunting down in play here,. like a leaking exhaust header on the O2 sensor side.

(along with the fact that I discovered a dead battery cell the other day - see prior posts. first step is to replace it ).

Thank you to all for reading and commenting, your time and ideas are greatly appreciated.


Dave

ULTM8Z

I think you're probably hear a leak in the header if it was significant enough to disrupt the O2 reading.

Ok, so let's assume for a minute that the observations are in fact abnormal...

Do you have a datalog? I can take a look at it. ultm8z@yahoo.com.

Preferably with the SAUJP ADX file... That ADX gives a crazy amount of fidelity and visibility on what's going on.

Also, what are your idle fueling parameters set at?

Fast O2 R/L threshold
Fast O2 R/L Threshold hysteresis
Upper zero error ref slow O2 R/L
Lower zero error ref slwo O2 R/L

One thing to look at too is if you're triggering delta MAP events during idle... Large cams may be prone to this. Assuming your 12 is in neutral (mine is ~14), so then your cam is proportionally a little more radical then mine is for my 350.

But, at 0% TPS, a factory 8D will enable delta MAP fueling if it detects a change in MAP of 3.13 kPa, and it'll add in AE fueling which will easily peg the O2 rich... In other words, even if you don't have your foot on the gas, the ECM can still deliver AE fueling based on sufficiently large delta MAP events.

I can then envision... if you don't have the switch enabled to "Reset Integrator When AE in Progress", the ECM will then react to that extra AE fuel while it's occurring and start trying to reduce the BPW to compensate. But when the AE delta MAP event ends and that extra fuel is gone, the BPW (already being low) will cause a momentary lean condition, at which point the ECM responds by adding fuel.

That kind of tail-chasing DROVE ME FREAKING NUTS when I was tuning my AE (until I discovered that switch)... but I digress

Any rate, if you datalog with the SAUJP adx file, this kind of thing should be pretty easy to detect.

In the end, my guess if you want that kind of O2 stability, you'll need to go open loop and fix the AFR via the open loop tables and/or the idle added-AFR in the constants. When I went open loop, my O2 voltage stays pretty steady right around 800 mV (I idle at between 13.2 and 13.5 as reported on the WB).

My approach to idle overall is to have the ECM do as little as possible (which, with the miniram, the better the idle quality, the better the transition to off-idle).

Air flow => increase throttle blade openings to miminize IAC counts
Fuel => Went open loop and fixed the AFR
Spark => zeroed out the spark adder/subtractor for compensating on idle speed variation as well as the spark correction bias vs coolant temp.

It always seemed to me like the ECM was chasing and trying to balance too many things for such a simple operating mode as idling... granted, with larger cams and lower idle vacuum, we make it more difficult. But it seems like the less I have it trying to do, the better the idle is on my modified engine.

Street Lethal

iTrader: (16)

If we were able to have a narrowband swing at such a thin margin there would honestly be no need for a wideband. The overlap in the cam is more than likely having an effect as they can be a bear to tune, so the VE table needs attention around the surrounding cells wherever your fueling is hovering. Pull a plug or two and take a look to see how it's burning if you get a chance. An exhaust leak would cause an O2 issue, absolutely, but you would hear a very prevalent and noticeable ticking at idle. Just make sure the ECM line from the O2 sensor itself isn't running parallel with a spark plug wire, coil wire, or ECM ignition wires. That is one of the reasons why GM routed it the way that they did. Dead battery cell and or alternator issues will effect injector voltage, so see if the problem goes away with a fresh battery swap...

- Rob

lakeffect2

1) I am hearing a tickin, but on thepassenger side where there is no sensor. Both WB and NB are on the drivers side and 90* apart. New gaskets will be going in tomorrow with new battery to eliminate both issues.

2) I am using the EBL and therefore have the EBL's XDF file in use. I assume this is no longer $8D and not all descriptions of terms are the same.

3) requested current fuel parameters as follows:

int-O2 window for fast R/L: 158.18
O2 error for INT correction: 35.36

INT- idle O2 window terms: Min R/L 609.96mv rich 609.96 lean 579.02
INT rich O2 (upper) 654mv@30kpa, 583@ 60Kpa (from 50-100kpa, idles at 65)
INT lean O2 (lower)565mv@30kpa, 495@60 (from 50-100)

I will need to do data logging
I will check path of O2 sensor harness path separation in reference to ignition wires. Recently did some zip ties at firewall harness entry point.

Can'tsy that I know how to attach a file within a post as some have done, but will forward via Email given.
Thanks to all, will report back.


Dave

ULTM8Z

In that case maybe do an excel export of the log...

Also your idle O2 thresholds are a little lean for your cam IMO. Before I went open loop, it seemed that my car liked about .750 for the R/L threshold...

lakeffect2

Do I understand correctly that a reading of .450 mv approximates an AFR of 14.7? Maybe I've lost memory of one of the basic relationships.
perhaps I was under the misguided impression that higher voltages meant richer, and lower were lean, where as I am all over the place from single digits to 800. No consistency on an instantaneous basis.

If my lean terms are at .560 and I am seeing 6-12 that would help explain 19-20 afr, but still needto find why they are so low to begin with.

Dave

Street Lethal

iTrader: (16)

Yes, .450mv is approximately 14.7 A/F, or a 128-BLM. But again though Dave, it is important to tune the surrounding cells in the graph, don't just let the VE Learn self adjust that one cell and think you are done. You have to get the surrounding cells tuned because they "pull" on the cell in which you are currently in. This is why you are seeing an erratic air/fuel. It's just like tuning your SA timing, if you have one cell targeting 24* at idle, with the surrounding cells targeting 19*, 15*, 25*, etc, they will play tug of war with each other and you won't see a stable spark advance at idle. RBob explains how to allow for the EBL to tune for such areas in which are hard to reach with the VE Learn. If all sensor data is on target, and your wiring is okay, then this is the only possible explanation especially if the NB and WB are reporting the same...

- Rob

lakeffect2

I've always been a believer that "idle is an area, not a signal point. I'll cross check VE and spark for large shifts in the table.

By the way, How to I insert a .bin file into this thread? I see" image" and "link" above.

Street Lethal

iTrader: (16)

The moment we settle on throttle body, intake and cam size, sadly it becomes a signal point.

Underneath Submit Reply, in the Additional Options section, click on the Manage Attachments hyperlink, then upload the file you want to be attached. It will provide you a listing of what is accepted...

- Rob

lakeffect2

Post flattening idle area, and raising low kpa/rpm area.
New exhaust gaskets, New battery.

The beast is back!
Little things to pick on, but at least it idles without surging at projected idle of 900rpm. Went from open to closed without hiccups. Did not go lean when coming to lights. (I'll live with a bit rich for now)

Once I replace the tranny speed sensor. I should be able to do some learns. Not getting any info at present.An issue no doubt.

Dave

lakeffect2

Having gotten into the habit of always used "quick reply" The added instructions were never visible.
When mentioning idle, I mis-spelled "single" as "signal; My belief is that idle is not a "Single" point, but an area.

As a point of reference, the way I would open the graph and twist the graph of the "Lower VE Table" to make points easier to pick off, made things appear much flatter han they really were. Observing the real shape of the graph when it first opened allowed me to realize it was not flat around my idle point, because I was skewing the way I was looking at it. In truth what seemed flat (but skewed) was either dropping off, or shooting up, depending on which way I was headed. A big mistake in understanding what I am actually viewing by the angle it was being viewed at.

ULTM8Z

Are you still getting the fluctuating voltage? How much variation in VE values did you have around the idle cells?

lakeffect2

I noticed a few things after recalibrating,
Checked fuel pressure (46lbs)
Check TPS voltage, 0.54 v
Minimum throttle position to hold idle
Base timin at 14*

I HAD 2-3% difference from cell to cell in any direction. I made it pretty much flat from 800-1000 rpm and 50-65kpa, my target idle areas. AS mentioned , my skewed angle of looking at the VE charts did not help the situation in the least. Should have copied and pasted those areas.

I had the Heads Up Dislay visible while cruising around the block, but have not yet done datalogging until my VSS sensor for the tranny comes in. Notable I did not see the wild voltage swings, nor feel/see wild surging at idle like before. Additionally, at least for the 10 minutes that I went around the block, It was going systematically going leaner to 18-20 AFR and stalling out., it was around 13 at idle to 15.0 on the road.
Overall vast improvement. When I stomped on it once, the vintage 65 year old Butt-O-Meter was very happy. (OK I lied, I stomped on it twice)

Street Lethal

iTrader: (16)

Glad to hear she's back and you're having fun with her Dave, that is what it is all about. The O2mv issue is always just temporary for everyone until you dial her in, with or without a healthy charging system. This was a few months back when I decided to resurrect my turbo GTA project, some of the data is a little off in this video to my liking because not only did she sit, but I made some changes without touching the bin, but I got it where it needs to be immediately after. Time and patience, never let the initial data readings upset you until you know for sure that the charging system is solid, and that the sensors are reading correctly. Sorry for the volume cutting out, my i-phone gets effected by the open exhaust...