Full Lean in Closed Loop Operation?

[WithHonor037]

Hey everyone,

Ive got a bit of a head scratcher on my hands and am hoping someone might have some new suggestions or ideas for me. Ive got a '91 hatch with a basically stock ka with an emissions delete done and just the other day apparently started running full lean when the engine is completely warm. I can let it sit and idle with no problems even when the engines fully warmed up but when its taken for a drive (engine under load) it will peak and stay full lean after a maybe a mile of driving. Ive checked almost everything I can think of with no evident issues. Ill simplify by just simply listing everything ive tested and checked out ok:
1. TPS - reads spot on at closed and full throttle
2. Coolant temp sensor - .29 on a warm engine (swapped a new one in just
to be sure, no change)
3. MAF - voltage signal tested normal
4. AAC/FICD - taken apart and inspected (not dirty at all, Im pretty good
about keeping it clean), No signs of an issue. Replaced with new gasket.
5. Vacuum Leaks - ran engine at idle and sprayed all intake components with
carb cleaner, no change in idle.
6. Fuel Pressure - fuel pressure stayed at a steady 39psi.
7. Valves - Pulled valve cover and inspected for sticking valves, no obvious
damage
8. Spark Plugs - Pulled all 4 and all were white tipped so all 4 cylinders are
running lean, replaced with brand new plugs and checked gapping
prior to install.
9. Distributor/Cap/Rotor - All were replaced with brand new units within the
last 6 months. Double checked with timing light just to be sure and is
firing right where it always has.
10. O2 Sensor - Have a narrowband hooked to a relatively new Bosch sensor
for monitoring the engines operation (don't tell me theyre terrible
and shouldn't be used for tuning. I know. Its merely a diagnostic tool)
AFRs are normal until engine gets fully warm after driving and then
runs a steady 25.5 (max of the monitor) on accel, decel, idle, etc.
until the engine cools down.
11. Cooling system - Havent checked the pressure or anything but its full and
shows no sign of overheating at any point
12. Oil - No coolant in oil, its dirty, but that's about it

Im sure Im missing a few things I checked but Ive been at this thing for 2 days now and cant seem to solve this. I even desperately tried what appears to be a homemade consult cable to an adapter board. Got it to connect to the ecu using ECUtalk software but it wouldn't display any realtime data. Any suggestions are welcome at this point. Hopefully some of you are smarter than I am! Thanks in advance!

[brndck]

Hey everyone,

6. Fuel Pressure - fuel pressure stayed at a steady 39psi.

8. Spark Plugs - Pulled all 4 and all were white tipped so all 4 cylinders are
running lean, replaced with brand new plugs and checked gapping
prior to install.

10. O2 Sensor - Have a narrowband hooked to a relatively new Bosch sensor
for monitoring the engines operation (don't tell me theyre terrible
and shouldn't be used for tuning. I know. Its merely a diagnostic tool)
AFRs are normal until engine gets fully warm after driving and then
runs a steady 25.5 (max of the monitor) on accel, decel, idle, etc.
until the engine cools down.
6. are you checking fuel pressure just at idle? it should increase with engine load.

8. how long have the old set of plugs been in for? have you removed the new set to see how they look after some miles on em?

10. you can't accurately tell afr from a narrowband. do i need to repeat that?


is the car running rough, shitty, poor mpg, stalling, hesitating, etc, or are you only saying its running lean due to the tips of the plugs and the reading of your narrowband?

[WithHonor037]

6. No I haven't checked it as load increases but the fact that it runs ok in open loop makes me think that it's fine.

8. I haven't pulled the plugs again yet as the symptoms are still present so I'm pretty confident the issue is the same.

I realize that you can't accurately read afr from a narrowband but the stock o2 operates in 3 positions. Rich, stoics, and lean to tell the evil which way to adjust. I can read mv output on the o2 to verify proper operation of the o2 with it.

When the car runs lean it's more than just the narrowband telling me so. The car misses occasionally, backfires so to speak, and has almost no power. But only under cruising conditions with minimal throttle change. If the throttle body is opened into a heavy accel position it will seem to catch and run smooth but only while accelerating but the o2 reading never changes from full lean. But on the other hand if driven from a cold state the car runs normal until it gets fully warm which leads me to the conclusion of an issue only during closed loop since it's designed to run rich in open loop it makes sense.


Sent from my iPhone using Tapatalk (http://tapatalk.com/m?id=1)

[mxexux]

I am also curious as to how you are measuring AFR. Also, did you check codes? I know you said you have emissions deleted, but the ECU can still store additional codes.

[WithHonor037]

I'm basing my afr off of the stock o2 sensor. Since the engine is stock that should be more than sufficient for what I need. All I need to know is if my mixture is rich, stoich, or lean which are the 3 positions a narrowband (stock) o2 sensors reads in through outputting millivolt signal to the ecu. I have a narrowband monitor tapped directly into that signal wire.

I have not pulled codes from the ecu because in the past it has proved to be unreliable by presenting odd code readings that I could find no corresponding error with even through contacting the dealer. This was a large reason for attempting the consult cable to monitor the information the ecu was receiving but that was unsuccessful. At his point in curious if I have a bad ecu but I would rather know it's bad before just buying one to try and finding out it's not the issue at all.


Sent from my iPhone using Tapatalk (http://tapatalk.com/m?id=1)

[Kingtal0n]

What voltage is the O2 putting out. Does it start at 1V when the engine is cold. Does it go to 1V when you go wide open throttle. Does it ever go 1V? It could just be a bad O2 sensor.

The plugs should be fairly white if the engine is running right. That old school tan colour is more for non-fuel infected engines.

O2 sensor should put out either 1V or 0V. you need to start there. If you go wide open throttle and the car RUNS right, but the O2 sensor says 0Volts, then either the O2 is bad, or the fuel pump is going bad, or the maf is going bad (but the maf is almost never bad) it could be an air leak after the maf also. there are plenty of other things of course (injectors regulator ecu etc...) but those are also very unlikely. You need first to make sure the O2 sensor is actually working.


The only thing the ECU actually uses the narrowband for is cruise situation. When you cruise, it flips from 0V->1V->0V constantly. If it isnt doing that while you drive, then it's bad, or the computer is not kicking into closed loop for some other reason. Another way of saying this is, when you go wide open throttle, the O2 sensor is just a monitor for YOUR eyes, not the computer. When you go WOT it should just sit at 1V. If it does not, then it is either bad, or the engine is not getting enough fuel because (fuel pump / maf / etc)

[WithHonor037]

The o2 sensor was just replaced in August so it's not that old. I don't have exact numbers on voltage for it but the voltage does fluctuate until the engine reaches full operating temp.

I checked the fuel pressure at idle. I'm going to revisit that tomorrow and check it at WOT. Unfortunately I can't monitor it while driving. But it held a steady 39psi at idle with no deviation. Even after opening the purge valve on the gauge a few times it still returned to 39psi. So the fuel pump seems to be fine.

I tested the maf and the voltages were within factory spec so I'm confident that's not the issue.

I've checked for vacuum leaks several times with no signs showing. If I can't come up with anything by this weekend I'll be pulling the manifold off and replacing all the gaskets and hoses again. It was done once a year ago when the engine was rebuilt.


Sent from my iPhone using Tapatalk (http://tapatalk.com/m?id=1)

[WithHonor037]

No updates yet as all 3 of my kids, wife, and myself have been sick. Hoping to get back at this thing bright an early tomorrow. If anyone has any other suggestions on top of what's already been offered it would be greatly appreciated.

My first steps are going to be double checking my fuel pressure, pulling the pump and checking the strainer, and I saw another guy with an SR found his issue to be his alternator so I'll check that out too although my voltage gauge reads just above 14v running. Maybe the amperage is too low though. Thanks again to everyone for they're .02


Sent from my iPhone using Tapatalk (http://tapatalk.com/m?id=1)

[Kingtal0n]

No updates yet as all 3 of my kids, wife, and myself have been sick. Hoping to get back at this thing bright an early tomorrow. If anyone has any other suggestions on top of what's already been offered it would be greatly appreciated.

My first steps are going to be double checking my fuel pressure, pulling the pump and checking the strainer, and I saw another guy with an SR found his issue to be his alternator so I'll check that out too although my voltage gauge reads just above 14v running. Maybe the amperage is too low though. Thanks again to everyone for they're .02




Sick you say? Now thats more my area than cars. What sort of sick? Sounds viral with the way it spread so easy. Hopefully self limiting and I wish you all a speedy recovery. Wipe everything you touch down and wash or replace everything you can. Even in the presence of antibodies against viruses the chances of infection are still there, and go up as you encounter more of them in your environment.

before you go pulling things apart I suggest you simply verify the O2 sensor is doing it's job. If you have to, buy a cheapy A/F gauge from autozone and wire it up to watch closed loop in action. Even with a bum fuel pump it should still kick into closed loop operation and flip the gauge back and forth. Even with a low voltage (no alternator). 99% of the time, a lean closed loop operation is due to a bad O2 sensor. Especially if the car runs fine in open loop.

[WithHonor037]

Yea pretty sure it's viral and something one of my kids picked up at daycare. Fortunately seems to be a 24 hour bug as my oldest and wife are already much better. Thanks for the good wishes though.

I'm not sure I understand what u mean by wiring the a/f meter to monitor closed loop? When the ecu operates in open loop it simply ignores the signal from the o2 sensor until it's warm enough to trigger closed loop operation. Or at least that was my understanding of it. Right now I have a Jet Air/Fuel meter tapped directly into the single wire from the o2 sensor. I can see the change in voltage readings from the sensor and it converts it to a rough afr. The sensor appears to be working fine. I'm not sure if that's what kind of setup your describing?


Sent from my iPhone using Tapatalk (http://tapatalk.com/m?id=1)

[Kingtal0n]

If the sensor is working fine, then it is flipping from 0-1-0 voltage constantly.

And if it is doing that, then you are not lean during closed loop! (thread topic)


In other words... ( I love saying that ) The only way we have of knowing a car is actually in closed loop (besides having some kind of OBD monitor device) is to watch the o2 voltage flip back and forth. If it's flipping back and forth... your O2 is working and your car is in closed loop doing what it should be doing.

[WithHonor037]

The reason I suspect it's a closed loop issue is because the symptoms only appear once the engine is fully warm. Which as we all know is when the ecu switches to closed loop. Open loop is just for warm up. Hence why it runs rich.

Now the o2 sensor should be constantly changing if the car is running right. In a situation where your ecu is constantly adjusting fuel trim causing it to read that way. But I there is an issue causing the engine to run lean or run rich it will not flip back and forth regardless. It will simply display information from what it sees.


Sent from my iPhone using Tapatalk (http://tapatalk.com/m?id=1)

[Kingtal0n]

The reason I suspect it's a closed loop issue is because the symptoms only appear once the engine is fully warm. Which as we all know is when the ecu switches to closed loop. Open loop is just for warm up. Hence why it runs rich.

Now the o2 sensor should be constantly changing if the car is running right. In a situation where your ecu is constantly adjusting fuel trim causing it to read that way. But I there is an issue causing the engine to run lean or run rich it will not flip back and forth regardless. It will simply display information from what it sees.


Sent from my iPhone using Tapatalk (http://tapatalk.com/m?id=1)

I dont understand quite what you mean. If the O2 sensor is not flipping back and forth, you are not in closed loop. Just because the engine is warm does not automatically mean closed loop. True closed loop depends on O2 sensor flipping 0-1-0 voltage. You may have an issue preventing you from going into closed loop....


Which is why I am so interested in your oxygen sensor. First we need to determine what that O2 sensor is doing, precisely. Is it flipping voltage or not? Autozone sells an air fuel gauge that wires to the O2 signal and will go red->green->red all day long during closed loop.

[stprasinz]

^ well put, and o2 has 2 readings rich lean stoich is the switch over, rich lean rich lean. there is no way to know how rich or lean, only rich or lean. While cruising it actually should be lean, but closed loop like king said. Also the rich and lean would not be actually rich and lean, but instead under over the afr of what the ecu actually wants to see in closed loop.....

All that being said, why do you think it is "going full lean"? Does it shut off? Does it stumble? Does it pop under accel?

What makes you think it is not running correctly, other than your gauge hooked up to your narrow band sensor?

[Kingtal0n]

But I there is an issue causing the engine to run lean or run rich it will not flip back and forth regardless. It will simply display information from what it sees.


Sent from my iPhone using Tapatalk (http://tapatalk.com/m?id=1)

this ^^ you are telling me that you are not in closed loop, but rather, the O2 sensor is reporting 0Volts constantly when the engine warms up. Is that right?

If you were in closed loop, and the O2 sensor was pegged at 0volts, then the ECU would keep adding fuel to it's maximum adjustment in an attempt to bring that sensor voltage to 1.01volts. You would be telling me this: "My O2 sensor is pegged at 0volts, but my car is running super rich and blowing black smoke" which would mean bad o2 sensor for sure.

But instead you are saying this: "my car warms up and the O2 voltage pegs at 0volts but the engine does not get rich in an attempt to fix the lean situation" which tells me that the ecu is not attempting a correction which tells me that you are not in closed loop (or the computer would be trying to fix the 0voltage problem). On the other hand, it may be attempting a correction in closed loop and you could be in closed loop, but the fuel requirements are not being met because of an issue with your: (fuel pump, regulator, fuel filter, blockage, fuel level, fuel sock, injectors, etc). We could verify this by driving the car. If it cannot maintain acceleration above a cruise (falling on it's face because of a lean situation during driving) then we stop looking at the O2 sensor and start diagnosing a fuel system issue.

[WithHonor037]

There seems to be confusion as to what im saying about the O2 sensor so Im going to try and restate my issue.

The O2 sensor appears to be working fine. Im measuring it in mV not V. The readings from the O2 sensor are normal on cold startup until the engine gets completely warm (when closed loop should begin). Then the O2 readings drop almost to 0 (about 145 mV or .145V) Indicating a lean condition. The car will stay here until shut off to cool down.

While its running warm and full lean it will pop, backfire, and miss occasionally. Virtually undriveable. Under heavy acceleration the car will accelerate and sound normal but under cruising or slight acceleration it will barely move itself and shows all the symptoms previously stated.

I hope that's a little more clear.

[WithHonor037]

I do suspect a fuel system issue but I have found no evidence suggesting one. Idle psi was 39 and Im about to test under accel now. Ill post back with results shortly.

[WithHonor037]

Under accel with no load im getting 42 psi. That seems reasonable to me?

[Kingtal0n]

1. When the car is cold, does it drive normal? You tried wide open throttle etc... and it runs fine?

Q If so, try unplugging the O2 sensor and driving the car warm to see if the poor performance disappears.

Q If it still runs lean (pops backfire missing), plug the O2 sensor back in, and unplug the maf sensor while watching the O2 voltage. If the O2 voltage suddenly jumps rich when you unplug the Maf, you may have a dirty/bad maf.


A A fuel system issue will not "disappear" when the engine is cold, usually. So establishing proper cold performance is essential to our diagnosis.

Q Another thing I need you to test is wide open throttle performance while WARM with everything plugged in, while the car is displaying it's issue. You said "AFRs are normal until engine gets fully warm after driving and then runs a steady 25.5 (max of the monitor) on accel, decel, idle, etc." But I need you to verify that includes WOT.

A If the cold performance including wide open throttle is "good" (car seems to run normal), and the engine runs lean at wide open throttle while fully warmed up, it is starting to sound like a MAF / ECU related issue. Could be a vacuum leak after the maf as well.

[WithHonor037]

Cold the engine runs completely normal. Accel, decel, and WOT. No issues at all.

I did run it without the maf plugged in. All it did was amplify the issues already present. The car ran far worse plus the whole 2500rpm rev limiter thing.

Unfortunately the car stalled and will not restart now after my pressure test so im on the trail of that before I can test it without the O2 sensor plugged in.

[Kingtal0n]

1. Before you unplug the O2, disconnect the battery and cycle the ignition to clear the block learn multiplier from the ECU.


2. disconnect your monitor device if possible. make the wiring as factory as possible. Check that the ECU ground by the kick panel is in place. check the ground on the intake manifold that connects to the battery tray. verify the battery terminals are tight and corrosion free. Check the plug in the ECU to make sure it is tight.

3. then unplug the O2, start and warm it up. If it still runs bad, be sure to try wide the open throttle performance before shutting off and reporting back.

[WithHonor037]

Aright so things just got more interesting. After sitting for a little the car will now start and stall at random times. I haven't found any notable pattern yet as to why. And on the last test drive the rpm would jump to almost 2000 between shifts. Pulling back into my driveway it stalled and will not restart again. I had the fuel pressure gauge on it the whole time and it never dropped below 39psi so Im pretty confident it is not a fuel issue now.

[WithHonor037]

Checked ECU codes. Code 55. Apparently nothings wrong...

[Kingtal0n]

Did you check the grounds and the ECU plug? At this point I would recommend swapping the ECU, and checking for leaks between the maf and throttle body. If unplugging the maf seems to make the problem worse, then it is probably not maf/leak related, but just to be sure.

Any pictures of the engine? Being that it was rebuilt have you checked compression?
and what do you mean jump. its automatic? So when it goes to shift, the rpm climbs up 2000rpm, then drops suddenly?

Sounds more and more electronic, wiring, grounds, computer. But the trans ecu is separate from the main ecu, so if the transmission is suddenly acting up AND the engine as well, clearly electrical / grounding needs to be checked. Did you ever check the alternator is putting out 14volts?

More and more I wonder about the mechanical aspect of your rebuilt engine. please get back about the compression results asap. Actually should have done that first, I thought you did but I dont see it on your check list.

[WithHonor037]

No its manual. I mean when the clutch is pushed in between shifts the rpm would jump. The engine was rebuilt a year ago so its by no means fresh. it runs too smooth when cold to be a compression issue.

I went back to basics first and have the manifold torn apart. Ive had a lot of vacuum leaks in the past so im doing my best to absolutely eliminate it as a possibility.

I checked all my grounds and couldn't find anything loose. I didn't go as far as to check resistance through wires yet.

[Kingtal0n]

No its manual. I mean when the clutch is pushed in between shifts the rpm would jump. The engine was rebuilt a year ago so its by no means fresh. it runs too smooth when cold to be a compression issue.

I went back to basics first and have the manifold torn apart. Ive had a lot of vacuum leaks in the past so im doing my best to absolutely eliminate it as a possibility.

I checked all my grounds and couldn't find anything loose. I didn't go as far as to check resistance through wires yet.

Just boost leak check the engine for potential air leaks, dont bother pulling it apart. If there is a leak after the maf, unplugging the maf would help it run better, and the idle would be higher in both situations (maf / no maf).
did you try the ecu reset / unplugged O2 sensor yet? those order of operations are diagnostics to make conclusions.

Its usually the simplest thing. Do you have an air filter on the maf? try to upload a picture of the engine.

A vacuum leak makes sense but it would need to be very large. When cold the enrichment provided allows the engine to run normally, but the idle would have to be very high. Once warm the air leaks cause a lean condition, but the O2 would compensate and the idle would shoot up. however, there is a max compensation, but it is still enough to cause a very high idle. the rpm shooting up 2,000 between shifts indicates a vacuum leak of course. engine speed is controlled by air, not fuel, for the most part.