Rich but reads Lean

[Newlogics]

Dont want to steal your thread, but I have a dumb question? Where is the knock sensor located?

On the block, firewall side, under the intake manifold.

 

[FORMONTOYA]

Let's put our scientific troubleshooting skills to the test.

On a NB O2 sensor in closed loop operation 0v is lean, 1v is rich. If the ECU saw 0v it will dump fuel in to try to bring it to around .5v, if it saw 1v it will take away fuel to try to get the ~.5v. The ECU will actually overshoot the target in its zeal to try and get it at the desired ~.5v, which is the reason you will see the sensor readings cycle up and down.

In a perfect system with no leaks between the metering device (stock MAS, MAF-T) and the front O2 sensor the O2 reading will cycle in a fairly predictable manner. If there is a leak between the metering device and the O2 sensor the reading will be more erratic, or it might not even cycle at all.

We can use this to our advantage in troubleshooting where to begin to start looking for leaks in the system if we give this some thought.

Exhaust leak:
The metering device has told the ECU how much air has entered the system and the ECU squirts XX amount of fuel for that amount of air. It goes through the combustion process and exits the cylinder and encounters the exhaust "leak" that allows air to be sucked into the system and past the O2 sensor. The sensor will "read" the extra air that is in the system and "think" that it has to add fuel to compensate for the extra air, even though the initial air/fuel ratio was correct. It goes through this process until it reaches the projected ~.5v, but more than likely it dumps in the max amount of fuel that the ECU is calibrated for. This is the reason that the car runs rich.

An exhaust leak will almost always have lower O2 readings (and the ECU adds fuel) under idle, cruise and slight boost conditions because of the extra air seen by the sensor while the car is actually running rich. WOT tuning will also be a pita to get completely under control.


Intake/boost leak:
The metering device has told the ECU how much air has entered the system and the ECU squirts XX amount of fuel for that amount of air. Before the metered air enters the combustion chamber it encounters an intake/boost leak. During idle and cruise the intake track is under vacuum and will suck extra air into the system with the same results as an exhaust leak. The difference is that with different levels of vacuum you will have a different amount of air being sucked into the system to the point where you start to hit boost and then you will be blowing air out of the system. The O2 readings will reflect this change by adding fuel when under vacuum to pulling fuel when under boost.

An intake/boost leak will almost always have lower O2 readings (and the ECU adds fuel) under idle and cruise conditions because of the extra air seen by the sensor while the car is actually running rich. Under slight boost the car will have higher O2 readings (and the ECU pulls fuel) because of the air being blown out and the ECU "thinks" it is running rich while the car is actually running lean. Of course WOT tuning is also a pita.

This is the reason a logger is most important as troubleshooting a problem is very difficult without one. You should NOT tune off O2 readings, but combined with fuel trim readings it can be very easy to diagnose where to start to look for leaks in the system. Also, this is why you should periodically log the operation of your car to spot potential problems before they arise.

Of course one would hope that you could hear a boost leak, but sometimes that is not the case. This is the reason boost leak test are preached about on this site as that is probably the number one cause for people having problems with their fuel trims as intake/boost leaks are probably more common than exhaust leaks.

And having a loggable WB O2 sensor will inform you as to what the real air/fuel ratio is for comparison, plus it is a great tuning tool.


Jim

 

[unclesam099]

Jim, that post should be written up somewhere for everyone to be able to read!!!

From the sound of it, the Boost/Intake leak sounds exactly what I have because the O2s at WOT are ok, but a little bit low... But these sentences describe my problems almost to the T:

"The difference is that with different levels of vacuum you will have a different amount of air being sucked into the system to the point where you start to hit boost and then you will be blowing air out of the system. The O2 readings will reflect this change by adding fuel when under vacuum to pulling fuel when under boost."

Thanks so much for the assistance.

 

[DSM_23]

So in regards to my post, could some of this problem be attributed to that sensor I have had unplugged? or is that sensor even important?

 

[DSM_23]

Well I just did a little searching a found my own answer, it seems that it isnt the temp sensor, its the engine coolant temp switch for the ac. Damn, now back to square one...

 

[Nautica985]

Exhaust leak:
The metering device has told the ECU how much air has entered the system and the ECU squirts XX amount of fuel for that amount of air. It goes through the combustion process and exits the cylinder and encounters the exhaust “leak” that allows air to be sucked into the system and past the O2 sensor. The sensor will “read” the extra air that is in the system and “think” that it has to add fuel to compensate for the extra air, even though the initial air/fuel ratio was correct. It goes through this process until it reaches the projected ~.5v, but more than likely it dumps in the max amount of fuel that the ECU is calibrated for. This is the reason that the car runs rich.

An exhaust leak will almost always have lower O2 readings (and the ECU adds fuel) under idle, cruise and slight boost conditions because of the extra air seen by the sensor while the car is actually running rich. WOT tuning will also be a pita to get completely under control.

My o2 housing has a tube that dumps into the atmosphere when the waste gate opens. When i ran seafoam through my car smoke was comming out of that tube. could air be sucked through the tube and past the o2 sensor? my o2 reads .02 at idle and low load.

 

[RiceKiller_TSi]

There's no vacuum present there.

 

[DSM_23]

Ok I have figured out the idle issue. I had to richen up the Idle from -10 on the afc to around -2, I guess the trims were to out of wack and throwing it into open loop. Now last night I did some tuning, and of all the knock I had (43 counts) I was able to bring it down to around 20 ish, but by richening the afc. It doesnt sounds right to me based on what everyone elses settings are, but my afc high settings are at +5 right now all through the rpm range. Why should I be having to add so much fuel?? I have 660's, a 190 pump, 2g mas, and everything else. I know the 2g mas should bring the compensation down for the 660s but I didnt think I would have to add fuel. Im still getting knock about 20 counts and now about 10-12 degrees timing advance. We could feel the car pull harder at about 5000rpms when timing would advance to 15 degrees but my knock reading was at 16-18 counts. Whats the deal here?

 

[FORMONTOYA]

My o2 housing has a tube that dumps into the atmosphere when the waste gate opens. When i ran seafoam through my car smoke was comming out of that tube. could air be sucked through the tube and past the o2 sensor? my o2 reads .02 at idle and low load.

Yes, air is being sucked into it. That is at least one of the places you have a leak causing the low readings.

Jim

 

[FORMONTOYA]

There's no vacuum present there.

True, there is no "vacuum" there. How the air gets sucked in is behind the pulses. Your exhaust is not a steady stream of air, it actually pulses. The "vacuum" is generated behind the pulses. A leak has rushing exhaust gas that would create an area of low pressure outside the exhaust at the leak. That would pull air in from outside and spike the O2 sensor with oxygen (lean), thereby causing a drop in voltage. It is even made worse under idle and low rpm when the "pulses" are at the greatest distance apart.

Jim

 

[FORMONTOYA]

Ok I have figured out the idle issue. I had to richen up the Idle from -10 on the afc to around -2, I guess the trims were to out of wack and throwing it into open loop. Now last night I did some tuning, and of all the knock I had (43 counts) I was able to bring it down to around 20 ish, but by richening the afc. It doesnt sounds right to me based on what everyone elses settings are, but my afc high settings are at +5 right now all through the rpm range. Why should I be having to add so much fuel?? I have 660's, a 190 pump, 2g mas, and everything else. I know the 2g mas should bring the compensation down for the 660s but I didnt think I would have to add fuel. Im still getting knock about 20 counts and now about 10-12 degrees timing advance. We could feel the car pull harder at about 5000rpms when timing would advance to 15 degrees but my knock reading was at 16-18 counts. Whats the deal here?

The main question you should ask yourself is: "Why do I have to richen this thing up so much?"

Answer: "A leak." More than likely an exhaust leak. Do the seafoam test and then tell us what happened.

Jim

 

[RiceKiller_TSi]

True, there is no "vacuum" there. How the air gets sucked in is behind the pulses. Your exhaust is not a steady stream of air, it actually pulses. The "vacuum" is generated behind the pulses. A leak has rushing exhaust gas that would create an area of low pressure outside the exhaust at the leak. That would pull air in from outside and spike the O2 sensor with oxygen (lean), thereby causing a drop in voltage. It is even made worse under idle and low rpm when the "pulses" are at the greatest distance apart.

Jim

Makes sense. I just thought that applied to N/A cars though. A turbo makes our system more "exit only" I would think. Like a continous flow of exhaust y'know what I'm saying? If you've ever seen the turbine spinning while the car is idling it doesn't seem to pulse at all like its being affected by vacuum.

 

[FORMONTOYA]

When you car is at idle, walk around back and stick your hand a few inches in front of the exhaust tip.

Jim

 

[RiceKiller_TSi]

Ha ha ha! I'll try that when I get it put back together again.

 

[DSM_23]

Can we please stay on topic here?

 

[FORMONTOYA]

Can we please stay on topic here?

We are just waiting for the results of your leak test so you can tell us where it was at. By looking at your posts, I am betting on an exhaust leak.


Jim

 

[DSM_23]

Alrite, well I did a seafoam test finally.. and you were right, there WAS an exhaust leak at my turbo to mani connection. I guess whenever I got the car back from Polk everything was new and needed to be retorqued after a little while. Well anyways we tightened the bolts and later went for some logs. Welll, same as before, its still running "lean" and my afc settings are up to +6 and knock is around 38 counts at 6700rpm. Boost is at 19 psi, Injector duty cycle was at 115% seems kinda high for 660's ?? Im trying to post up a log but I have to d/l Excel first but, Let me know what you guys think should be my next step.

 

[FORMONTOYA]

Did you also do a boost leak test? That would be my next check, make sure there is NO leaks between the metering device (MAS, MAF-T) and the O2. You have the exhaust side done, now it's time for the intake side.

Jim

 

[spyderturbo007]

Welll, same as before, its still running "lean" and my afc settings are up to +6 and knock is around 38 counts at 6700rpm.

How do you know it's running lean? Are you putting all of your trust in a blinking AFR gauge? Do the boost leak test and then start with a normal injector compensation value for your injector size.

You are experiencing rich knock. By bumping up the SAFC-II to 6% you are now adding almost 40% more fuel than you should be.

 

[Stingray90]

Sounds like my symptoms. I have a pre-o2 exhaust leak and I always suspected it could cause some problems. You guys Really confirmed it.... Better get busy.

 

[Stingray90]

Quick question. Where did you guys buy the seafoam? Pepboys or something.
Im just trying to save some gas here.

 

[FORMONTOYA]

I get it at AutoZone. I quess PepBoys has it, but just let your fingers do the walking and give them a call to find out before you drive all over town.

And yes, any leak between the metering device and O2 sensor, especially an exhaust leak pre O2 will screw up the fuel trims.


Jim

 

[DSM_23]

How do you know it's running lean? Are you putting all of your trust in a blinking AFR gauge? Do the boost leak test and then start with a normal injector compensation value for your injector size.

You are experiencing rich knock. By bumping up the SAFC-II to 6% you are now adding almost 40% more fuel than you should be.

Well I know im lean because First I pulled the plugs and they read lean, And my logger voltage is around .87-.85 and keeps dropping, and the only way we could make the car feel anyfaster was by adding fuel. Otherwise it would knock alot and pull all timing possible. My blinking guage is not my tuning device, its just there, doing nothing but flashing. My low fuel trim is around 115. A boost leak should make it rich though am I right?

 

[FORMONTOYA]

A boost leak at idle and cruise should drive your fuel trims up to compensate for the extra air, which would cause the car to run rich. Under WOT/boost the pressure will "push" air out, causing the ECU to pull fuel and run lean. Also with all that fuel your dumping into it, it should practically have gas pouring out the exhaust.

With all that said, I would forget about fuel trims and O2 readings, DO THE BOOST LEAK TEST, DOUBLE CHECK THAT THERE ARE ABSOLUTLY NO LEAKS BETWEEN THE METERING DEVICE AND THE O2 SENSOR, PERIOD.

Get the air side fixed, then worry about the fuel side of the A/F equation.


Jim

 

[sbiggi]

I just read your whole thread.

So you still havent checked base timing with a timing light?

Bust out the timing light, ground out the connector on the firewall per vfaq.com and check the base timing. The logger has absolutley no idea what you base timing is, only what it thinks it is.

Sound like your base timing is way advance and you cant richen it up enough to stop the knock. Black smoke is rich regardless of what the narrow band thinks it is. Not to mention 115% idc.

-Seth