Stumble with cruise control

[fishbone]

Under fuel tables tab:

Open loop primary tab:

Primary open loop fueling: Change all 14.7 values to 14.68.

It is not allowing me to. The minimum one step decrease is 14.59

So I did this, as found here

http://www.romraider.com/forum/viewtopic.php?f=33&t=7897

The smallest increment I can make on the open loop fueling tables is down to 14.59. So I set everything at 14.7 down to 14.59 and set the minimum enrichment to 14.65. I'm not overly concerned about running at 14.59 - last I checked the gas has ethanol added anyhow and stoich is closer to 14.2 =P

[fishbone]

Flashed the ECU and I got a chance to engage cruise for about a mile coming in to work. It seems to be working great, no more stuttering. On my way home tonight I'll jump on the freeway and set it at 80mph and see how it goes.

Overall the car runs just as before but does seem to pull harder at WOT and as a bonus, the torque converter is in lock-up mode at all times. I kinda like running in OL all over the place

 

So if the issue is fixed, where do I go from here? Does this rule out an air leak then?

[NSFW]

You are talking about the uppipe sensor and in that case yes, I have mine completely removed and have the resistor mod. The rear O2 sensor is connected at the end of the downpipe. Stock there were 3: front (still there), mid (uppipe removed), rear (downpipe).

 

That was an EGT sensor in your midpipe.

 

In your first post you asked about AF Correction values... those are basically just noise. Unless it's pegged at +25% or -25% (or whatever the limits are) then you can ignore it. AF Correction is measuring deviations from the target AFR in realtime. Since both the MAF sensor and the O2 sensor are noisy, the deviations are noisy too.

 

Also the measurements are being made at a faster rate than your logger can keep up with, so you're only seeing a fraction of them. Since the logger is sampling a rapidly-fluctuating number, the results look very random.

 

AF Learning is basically the long-term average of the AF Correction values. It is not unusual for AF Learning to move up and down a few percent over time, so I don't think this is necessarily a big deal.

 

It might indicate that your temperature compensations are off a couple percent. It might indicate that a tiny speck of dust has stuck to the MAF wire. It might just be the random accumulation of all the random noise, so it might return to its old value soon. Or maybe the old value was actually slightly off (due to something listed above or something I haven't thought of), and this is the "true" value (as true as a noisy number can be) and you could tweak your MAF scaling to bring it back in line.

 

Notice I said could, not should. AF Learning #1 is the least important of the 4 values, since it only describes what's happening at idle.

 

If you have a vacuum leak, you'll find AF Learning pegged at +15% and AF Correction pegged at +25%. Otherwise you're probably fine.

[fishbone]

Thank you for correcting my idiotic statement regardin three oxygen sensors. I don't even know why I wrote it that way since I know full well what type of sensor I pulled as well as why, thanks to the resistor mod, the EGT is one single value at all times

 

I appreciate your insight and I wanted to come back with additional info.

The stumble is gone and has now become an oscillation, so the issue is not resolved. It is very very subtle, almost to the point that is not felt unless you are paying attention. The revs go up and down by about 300rpms. This is masked when the torque converter is locked up, but is also visible in TPS on the AP. The throttle angle varies by about 5%.

This cruise stumble/oscillation has also decreased in effect since I replaced the front O2 sensor.

The best way I can describe it as it is now is like a novice driver struggling somewhat to keep steady speed, so they feather the throttle for a second, then back off, then again. When the car is on an incline going uphill, the oscillation goes away.

 

AFR is running on the lean side by a factor of, err, one unit or whatever that is called. At cruise it is between 15.5 and 16.0, same as idle.

 

The only thing I can see at this point to do is pressurize the intake to be 100000% there is no leak. I think so far my plan is that, unless anyone else suggests something else or I think of anything else, to wait until pumps here switch to summer gas. Last time I pulled a LV and used cruise was last year in May and my fuel trims looked just about perfect. I have not touched the car between then an now other than an oil change.

 

Part of me wants to replace the rear O2 on the simple basis of it being old and still possibly flaky, but as I understand open loop does not utilize it and if that's the case it ruled out that possibility and may very well be money wasted.

And I cleaned the EFF out of my MAF so I know for a fact it is as clean as a whistle

 

I remember having read some Cobb literature that was saying the expected normal oscillation of AF Correction to be +- 8%, so I was concerned about the spikes in the min and max values possible. The A/F Learning #1 is not even consistent, one day it will be 9%, the following 11%, never out of that range though.

[f1anatic]

I also cleaned my MAF and in my case it was dirty (last cleaned 2-3 years ago). It helped but not a whole lot. I still have this residual oscillation feeling, on cruise control up to about 65 mph. Higher speeds - it's gone. I will change the rear O2 sensor when the weather warms up, along with spark plugs. Maybe I will even change the knock sensor while at it.

[LongHiway]

Well, you've ruled out my (wacky) theory on brakes but, what bothers me in all this, is why does this condition only exist when operating with cruise control on?

[fishbone]

I am completely speculating here, but it could be the rear O2 sensor being wacky which is confusing the ECU with flaky readings and as a result the ECU is trying to hunt for the correct AFR to ensure cat efficiency etc.

 

While I was running with closed loop disabled, the car was running on the lean side, around 16.5AFR at cruise.

I monitored Comm Fuel Final and the ECU was targeting the correct stoich.

 

Comm Fuel Final in the AP is: Final commanded fueling before the injector pulse width is calculated. This includes all corrections to fueling, including short term and long-term fuel trims.

 

So the lean condition was a bit curious.

 

What is the MAF supposed to read at idle?

Since I ran the car in open loop and the stutter/stumble went completely away (somewhat replaced by a feeling of oscillating gas pedal), doesn't that at least narrow it down some more to possibly the rear O2 sensor? What does everyone think? Part of me wants to replace it simply because it's got 100K miles/9 years on it.

[rhino6303]

Does your car have non-oem throttle maps? The throttle maps can cause surging during cruise (especially when going downhill/off acceleration).

[fishbone]

I'm not sure. The tune I am running is a tweaked Cobb Stage 2 map. I would have to look, but I am pretty sure whatever throttle map Cobb has in there is what I have.

However, the issue was never there prior to this year and I've been running this same tune for several years.

[fishbone]

Does this LV look like an air leak, scaling issue, or something else, or can't tell? First one in the sequence dated 5/5

The dates are on the bottom of each pic.

I also included for reference one from 2/27 and one from last year 5/12 when everything looked good.

 

I've got a couple things in mind at this point.

Buy a boost leak tester and pressurize the intake, or replace the rear O2 sensor.

 

So following my disabling closed loop, the stutter went away with cruise control, but my AFR during cruise was 1 unit higher, so it was hovering at around 15.5

Would this indicate a leak or scaling issue?

 

The car was tuned in Nebraska at 1300 feet of elevation and now it's at 150 or so. I take it this isn't an issue?

[Max Capacity]

Have you asked one of the Tuners for help ? may be send www.tunigalliance.com a email with this thread in it and see if he has any thought's.

 

He has my wagon running very smooth with his latest MAP. You've been chasing this for almost 3 month's...might be time to pay the man.

[fishbone]

I'm sure they get a ton of questions thrown their way. I'm looking to get some help here from those that know their way around tuning to provide some direction.

I'm not dismissing the possibility that I might need a retune, but I am looking to eliminate other possibilities.

The LV going hand in hand with a stumble/stutter at cruise and slightly lean AFRs under open lool leads me to believe the issue might not be just a retune. I appreciate the suggestion though.

I'm wanting to see if the LV is pointing in an obvious way to a leak or not.

Looks like the warmer it got, the richer it got across the board.

So I am planning to either buy a rear O2 sensor or a boost leak tester as my next step to at least rule those out.

Given that a leak tester will be around $30 as opposed to $100+ for a sensor, maybe I should just go that route for now.

From my searching and reading other forums/threads, people say that a faulty rear O2 sensor can in fact affect fuel trims.

The fact that running in open loop eliminated the stutter may support this, since the rear O2 oscillates AFR for optimal cat efficiency, right?

[NSFW]

It's hard to say for sure whether this is just normal variation or not, but personally I wouldn't lose sleep over it unless there was some other sign of trouble.

 

I have the same trims visible in real time on one of the screens on my carpc, and I check them randomly, and I see the numbers wander around +/- 5% week after week with no changes to tuning or hardware, and I'm not worried. The sensors in our cars aren't perfect, so there's noise in the measurements they make, and that's why the ECU has these features in the first place.

 

Do you have a WBO2? It would be interesting to know whether the WOT AFR is holding steady (which would indicate that the trims are working perfectly) or moving with or against the trims (which might indicate a leak - but probably a tiny leak, given that the trims are moving so little).

 

My AFR oscillates in cruise, however I don't know if the rear O2 sensor's job is related to driving that oscillation, or if it is just there to monitor the cat. Did you still see those cruise oscillations with closed-loop disabled?

 

And here's a failure mode that was new to me... I had some EGT sensors welded in a few months back, and they broke (bottomed out when tightened, then cracked, and popped out of their fittings), and that caused my A trim (idle) to stay pegged at +15.0%. But my idle and cruise AFRs were normal - it was apparently right on the edge of what the ECU could "correct" for. I have to put "correct" in quotes since the leak only affected the sensor readings. I was probably idling pretty rich during that time. Took me forever to figure it our because I was sure it was a vacuum leak. So if you are sure you have a leak (and I kinda doubt you do) the exhaust system between the block and the sensor should also be checked.

 

Now I need to go back under the car to see if the sensor are still in their fittings... I used an exhaust repair glue to fix them, but it's kind of hard to believe that any sort of glue can actually withstand manifold temperatures. And I still need to hook up the EGT sensor controllers so I can find out what those temperatures really are. Anyway, enough rambling...

[snow05gtRI]

i had the issue in the winter. since then, i had my AEM CAI (full pipe and filter) replaced and my MAF was cleaned during replacement. used cruise control again and it seemed to work fine, though I was on a flat road and only used it momentarily to see if it would freak out again, but I gave it a good 30sec to a minute to see if it would happen.

[fishbone]

Disabling closed loop got rid of the stutter so I am going to replace the rear sensor for now and go from there. I already replaced the front.

Will return with results.

[nicholaschase29]

I got tuned from Mike at Tuning Alliance. I also had an oscillating only while on cruise. He decreased the part throttle wastegate duty cycle. and removed 60-70% of the oscillation. It's still there but it's much more subtle now.

[fishbone]

Replaced rear 02 sensor today. No change in cruise, stumble is still there.

A/F learnings, for now, after a reset, appear to be slightly different but need to drive more to confirm.

At idle, A/F Learning will climb to 7-9%, and as soon as the A/C kicks in, it drops to 0.0. What would this suggest, if anything?

[Aktyler]

did you ever get this sorted out? I'm having the exact same issues as you and am puzzled.

[snow05gtRI]

did you replace a wheel bearing recently? The theory has moved on to non-OEM wheel bearings (except for one vendor, I forget which) since the wheel speed sensor is in there.