Rich but reads Lean

[Newlogics]

Well, I guess it's not technically a phantom knock, but can't call it real knock either, since it's not actual pre-detonation (although carbon deposts do create hot spots sometimes, causeing actual predetonation) I'm glad it worked out for you. (don't you get that warm feeling inside, when everything tunes out exactly as you predicted )
However, getting rid of phantom (or non-phantom) knock isn't always as easy as MCCC. I've gone through that. It tends to start creeping back after a month or so. In my case, only pulling the head and physically cleaning carbon off helped for good.

What you're going through I have battled for three months now.
I replaced the knock sensor, no effect.
I replaced the o2 sensor, no effect.
I replaced the plugs and wires, no effect.
I tuned and retuned till I nearly wore out my maft knobs... no effect.

When I got rid of my 43 count "phanton" knock, (which happened to be REAL knock) all my other problems vanished with it. What I did was simply clean out the head and pistons with 2 cans of MCCC (Mopar Combustion Chamber Cleaner).
It was a very simple solution to a nightmare problem. I havent had phantom knock since doing this.

Grab yourself a few cans and use one full can on the head (Let the engine vacuum suck this stuff up). And on the pistons I evenly divided a can directly into the spark plug wells and let it sit up overnight then changed the oil. Problem solved.

 

[92GSXBaltimore]

Running some combustion cleaner through your engine couldn't hurt the situation, as RiceKiller said. Although, I believe you problem is a pre-o2 housing leak. I had 3 large cracks in my exhaust manifold and experienced alot of tuning problems. My low trim was 148 and the mid was around 119 (mods in profile). I could tune it to run good but had to dump alot of fuel, but shortly after the tune it would get all screwed up again. After I welded them up, my LFT and MFT went back down to normal.

 

[DSM_23]

Well no place had o2 sensors today in stock so instead we just cleaned it with carb cleaner, put it back in and it functioned fairly normal, however after a while of driving the LED guage would stay locked on rich, but the logger read consistant cycling of voltage... Why would I have to run the cleaner through the engine so soon after it was just dissassmbled and ported, hot tanked etc a few months ago? Overall the car runs great. Just when it comes to making the power it should be able to make.... thats a different story. So what should my list of first priorities be in tackling this issue now? You all have had great imput keep it coming.

justin

 

[Newlogics]

First off, if loger shows O2 cycling, and gauge doesn't, double0check gauge wiring, better yet, rip it out and use as christmas decoration.
Secondly, carbon deposites can develop VERY quickly if you were running very rich for a while. The difference is that they're not as 'baked-in' as long-term deposites, which is probably why your MCCC has done such a good job. So if your fuel trims are normal now, and O2 behaves like it should (disregarding the gauge), then you're all set up for some power tuning. Which issue are you refering to in your question?

Well no place had o2 sensors today in stock so instead we just cleaned it with carb cleaner, put it back in and it functioned fairly normal, however after a while of driving the LED guage would stay locked on rich, but the logger read consistant cycling of voltage... Why would I have to run the cleaner through the engine so soon after it was just dissassmbled and ported, hot tanked etc a few months ago? Overall the car runs great. Just when it comes to making the power it should be able to make.... thats a different story. So what should my list of first priorities be in tackling this issue now? You all have had great imput keep it coming.

justin

 

[need2boostpsi]

Well i'm the other friend helping Justin with his tuning problem. couple of points that i'm seeing; while at idle, mmcd reads around 15 degree's of timing advance, we didn't do a base line timing check. Another thing is the "light show" was fine while the engine was cool but after a couple of spirited runs it was pegged on rich while mmcd showed a cycling of voltage. We get up to 43 counts of knock (obviously some type of phantom knock) under any type of boost. O2 voltage was at 0.89 max, which is a lean condition. Under WOT were getting a rich condition (unburned fuel exiting the exhaust). Some things I have thought of are pulling the plugs after a WoT run and checking. Were going to try to use sea foam to check for any exhaust leaks which may be present. We'll keep you posted as we use trial and error.

 

[Thomas91169]

your a/f gauge is worthless other than a pretty light show. the only time a a/f gauge is accurate is when its connected to a wideband o2 sensing unit. stop using it to determine the status of your vehicle.

if your engine is pretty new, i doubt its carbon deposits. your o2 sensor looks just like mine did at 60k. with dsmlink i was getting very erratic o2 volts at idle and cruise, and at WOT the o2 volts hung in there till around 5.5k and then it died. replaced the o2 sensor and alot of driveability issues(mainly horrible mileage and reliable o2 voltage levels) were resolved.

it could also be the knock sensor. most knock sensors pick up alot of road noise, suspension noise, and valvetrain noise and think thats knock. a good sensor will know the difference. a overactive sensor could be on its way out. i was getting quite a bit of knock retard(read via dsmlink) and i just got rid of the ecu's ability to read knock under 3* of retard, which basically wiped out the knock sensor from making any sort of phantom knock an issue, till i was WOT then the ecu saw it all like normal.

my thoughts. check exhaust for possible leaks before the cat while youre down there with your new o2 sensor, which should be replaced anyways to mantain good mileage. while that WILL fix your o2 voltage problems, you could still get horrible retard from P-Knock. check the knock sensor out, and possibly replace it. if that still doesnt help, run a can of seafoam through the motor via sucking it through the bov line(or any line thats a vacuum that leads to the intake manny), and keeping the revs around 3k(to achieve good atomization).

if that all fails, get dsmlink and rid yourself of phantom knocking woes forever, while you bask over its ability to tune your car in ways you never though possible.

 

[titogs95]

If I were you I would get an WIDEBAND. and make sure that your knock is not messing with the timing because that will create the smoke at WOT. RICH can create Knock.
Also don't even bother to tune with narrowband, my car sees .87 v when my AFR is at 10.5.
A wideband should tell you what spot you are running rich.....

Wow, your car must be my car's twin brother. I had some of those issues.....

 

[DSM_23]

well i just ordered at new o2 sensor im hoping that will fix the problem, cus alot of times it reads 0.0 volts and idles rough, then it will start working a little and occasionally go to 0.0 volts again. Does that sound like good reason to buy another? And could that be a possilbe solution for my knock problems whenever boost hits?

 

[Thomas91169]

possibly, if its making the car run so rich as to get a condition called rich knock, a condition i experienced personally in my beginning dsmlink days when my o2 sensor was bugging out as well. while the o2 sensor was not 100% at fault, id did make a difference, though the rest was tuned out by changing my airflow tables after i wasnt fully satisfied with the o2 sensor fix and went for more advice which led to it running right. like i said, at 5500rpm, my o2 sensor died, went to 0v, and ran super rich, at a point in a run where my boost and rpm's were coming up pretty fast.

 

[andymoraitis]

Justin,

Sure it can. It looks lik we're back to the original fix and it seems like it'll make the greatest difference in correcting the issue. I would still run a can of Seafoam through the motor to see if you can find any pre-O2 leaks that would cause you to run excessively rich while getting any deposits off the piston tops. Don't forget to pull the heatshields so you can see everything clearly and change your oil when you're done.

Let us know how it goes,

Andy

 

[RiceKiller_TSi]

Secondly, carbon deposites can develop VERY quickly if you were running very rich for a while. The difference is that they're not as 'baked-in' as long-term deposites, which is probably why your MCCC has done such a good job.

Exactly.

You're obviously running too rich so make sure you dont have a leaky fuel injector that is stuck open. Do the test using mmcd with your logger and see if you notice inconsistency while testing each injector. If you find that the car runs differently while testing one particular injector pull your plugs and see if that one is extremely black. If so, that is an indication of a rich condition possibly caused by a stuck injector. Your car will run hella rich and you wont be able to lean it out if this is the case.

The faulty injector will dump fuel causing severe carbon deposits which will cause hot spots that preignite fuel causing knock.

Just another thought that doest require spending any money or replacing any parts.

 

[DSM_23]

Alrite thoughs are some good ideas...Im going to wait till I get my sensor and see how the results of that play out, based on that is what my next move will be. However I do think I'll try and get around to doing the seafoam test if I get some extra time.

justin

 

[Newlogics]

Another possible solution to phantom knock.

Sometimes phantom knock just can't be eliminated. A while back, my rattling tranny was tripping my knock sensor. There was a thread about a very interesting experiment (her , on tuners). A guy, who's car has the worst case of PK, decided to make 1G knock sensor to perform more like 2G (and we know that 2Gs don't have nearly as much of a problem with PK). He basically welded the pin-hole in pick-up side of the knock sensor (2G knock sensor doesn't have it in the first place). He got VERY promising results. His phantom knock went away COMPLETELY, but real knock wasn't 'muffled'. He was still getting same counts of real knock at full boost (aroun 10counts), so apparently that mod didn't affect sensor's ability to pickup real knock, but completely eliminated phantom knock. I'm yet to find a customer with bad case of PK to try it out on, but somehow I have confidense in it.

As another thing, as was siad before, just because O2 reads .89V, doesn't mean it's lean. My O2 reads .81-.82 with 11:1 AFR. Narrow-badn voltage reading is only good for comparison purposes (comparing one run to another on the same car that is)

 

[DSM_23]

Update!
My new o2 sensor came in today and I put it in and it cycles the whole time, EXCEPT, whenever I turn on the car. It will set at 0.0volts until I begin to start driving I found that to be odd. The logger still says its lean with the fuel trims at 138-139ish. Knock occurs whenever there is basically ANY load while accelerating My base timing is set at 5 degrees on the logger. Mid fuel trim is 101. Ok I hope somebody can help me solve this, it doesnt seem to make since with my fuel setup. Its driving me crazy

justin

 

[FORMONTOYA]

It will set at 0.0volts until I begin to start driving I found that to be odd.

That is exactly what mine was doing when I had a leak between the turbo and o2 hsg except that at idle it would read .02-.04 max but would not start acting properly until I started driving.

I will bet money that you have an exhaust leak pre o2. As stated, use the seafoam to find it.

Jim

 

[unclesam099]

The O2 sensor won't start cycling until the car is warm.

All of these conditions sound exactly like mine. Mine is that my WOT runs are fine, but the idle has to be pushed up to +16 or +20 on the SAFC at the 1000rpm point in order to pull the Low Trim back from 139. Anything less and it just stays up there at 139.

 

[FORMONTOYA]

OK, let me clairify...after the car had warmed up and was running 190-200 temps after a run sitting in the garage at an idle the o2 would read .02-.04 because IT HAD AN EXHAUST LEAK PRE O2. The car "thinks" it is running lean and dumping a ton of fuel to compensate for the exhaust leak causing rich knock. Grab a can of seafoam and find the leak.

Jim

 

[Nautica985]

I am having the same problem with my 02 reading .02-.04v at idle. I have a leak between my compressor housing and exhaust side of the turbo. My MAS is right before my TB so the leak would not effect my airflow reading. Could I be having the same problem that you mentioned above? the leak is on the intake side of the turbo so i dont think it would but i just want to make sure.

 

[FORMONTOYA]

I am having the same problem with my 02 reading .02-.04v at idle. I have a leak between my compressor housing and exhaust side of the turbo. My MAS is right before my TB so the leak would not effect my airflow reading. Could I be having the same problem that you mentioned above? the leak is on the intake side of the turbo so i dont think it would but i just want to make sure.

Part of what drives your closed loop fuel trims is the front O2 sensor. That said, any pre O2 leaks will make the car "think" it is running leaner than it actually is and it will dump extra fuel to try and bring it in line. The low O2 reading you have (if the car is up to operating temp) would suggest an exhaust leak pre O2.

The easy way to find exhaust leaks is like andy said...seafoam. It is cheap and even if you do not have any leaks you will at least have cleaned out the combustion chamber.


Jim

 

[unclesam099]

But during driving it was reading just fine??

That's really odd, but it does make sense. I think that I might have a crack in my manifold somewhere, and have been wanting to get a 2g/evo3/tubular manifold... I think I will try the Seafoam deal...

 

[FORMONTOYA]

But during driving it was reading just fine??

That's really odd, but it does make sense. I think that I might have a crack in my manifold somewhere, and have been wanting to get a 2g/evo3/tubular manifold... I think I will try the Seafoam deal...

In my case it was NOT reading "just fine". It was just cycleing as it should while driving, but never going above about .75-.8. But then again, I had a pretty good leak.

Seafoam is great for finding those pesky exhaust leaks that end up screwing your fuel trims.

Jim

 

[DSM_23]

Well I just realized something that MAY be at some fault. 3 yrs ago I put in an aftermarket water temp gauge and used the top sensor unit. I simply unplugged the sensor and set it aside. Well now that this issue has been occuring I checked into it but the haynes manuel is kindof unclear as to what that sensor is. After all the reading I have done in the last week im hoping this could be my coolant temp sensor and if so it would be telling the ecu to basically flood the engine. But on my pocketlogger it shows the temp as normal, my guage inside the car works, and my aftermarket guage works. What sensor is this thing? Hope this info solves something.

justin

 

[turbohcar]

I do have a manifold stud with no bolt on it, so if this doesnt fix it I'll look into that.
justin

This could very well be the source of your pre-O2 exhaust leak. It's a $0.50 nut, or even cheaper if you can find a spare one from another car. Fix it!

On that note, I've got similar problems with knock. I've got a brand new O2 sensor, a new knock sensor to put on, new exhaust manifold gaskets, and I've got to buy a new 02 housing gasket. Mine may actually be more lean condition related because I'm seeing 104% injector duty cycle at WOT but 0 counts. I'm seeing 30 counts though at cruising speeds. The entire fuel system is getting upgraded for this season along with a wideband 02, eprom ecu, and a GM MAF setup.

 

[unclesam099]

I'm almost thinking that I could have a vacuum leak after the MAF (even tho it's running blow-thru). My head was just redone and the manifold was pulled off, gasket replaced. I just did a replacement on the Manifold - Turbo and the Turbo - O2 gaskets. So I kinda doubt that I have an exhaust leak. My O2 cycles up to .7v - .8v when cruising.

 

[screwdgie]

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