2G Oil burning exhaust NOT the turbo

[kjlinwi]

Have a 97 eclipse GSX that we've been completely redoing/updating.

Almost all done, but there is smoke out of the exhaust, rpm related, worse when driving but can do it in neutral as well.

I have the ECM link system and I have turned the turbo controls off while I'm dealing with this. Have checked the turbo inlet and outlet for oil (none). Pulled the BOV and looked into the intake tubing with a boroscope. It is bone dry not a hint of oil in it anywhere. I then did a cylinder leak down test on the engine. All cylinders 95% or better. Could not positively identify where the less than 5% was going, but I don't think that is my problem.

Ordered a new valve cover and made sure that the baffles were removed and cleaned out on it before powder coating, ordered new PCV valve and a special one way check valve that is plumbed in between the intake vacuum source and PCV. No difference.

Took it out for a three mile test drive and made sure to bag it enough that I was smoking the poor cars out behind me to the point that I couldn't see them. Brought it back in and pulled the plugs. All are perfectly clean, no sign of oil burning at all.

The most critical piece of information (I think), this engine never smoked until the head was swapped and the bigger cams were put in. The head was sent out to a machine shop and all new felpro seals were supposed to be installed along with new cams, springs and retainers. Now it smokes horrible. My guess is exhaust valve seals.

I can pull the 02 sensor and verify oil fouling but that still won't tell me the source.

Does anyone see anything that I am missing?

Thanks,

Matt

 

[91talonts1]

I have the ECM link system and I have turned the turbo controls off while I'm dealing with this.

I'm not really sure what you mean by this. You say it's not the turbo, how have you ruled it out?
In my experience leaking valve stem seals smoke less as rpm increases not more. Generally they only make it smoke upon start up unless they're really bad. Pulling the manifold and seeing if there's oil in the exhaust ports before the turbo, or only oil after the turbo is an easy thing to check. But if your plugs don't show signs of oil burning, that would mean the oil isn't in the cylinder (ie coming from the turbo).

 

[livedsm4g63]

leaking valve stem seals

Do a leak down test. This is what I would suggest looking into.

 

[PlanZero]

No offense, but did you take the head to a reputable shop? A buddy of mine took his head to a supposedly good shop for a valve job and some of the seals weren't seated properly, among other things. Since the seals are relatively cheap, I'd try replacing them, even though it can be tricky with the head on the car, before I pulled the head. You've eliminated any other possible cause, and since the problems started with the refreshed head there's only so much you can do at this point.
If the shop is reputable, and has any kind of guarantee (written or implied) there's no reason why you couldn't bring it back, but that's up to you if you want to go through all the labor. I'd attempt the seals first. That whole situation sucks.

 

[91talonts1]

The op says he did a leak down test, less than 5%... Come on man.

 

[livedsm4g63]

The op says he did a leak down test, less than 5%... Come on man.

This is what happens when you are in a meeting trying to pass the time... Good call.

To the op, what about your coolant level?

 

[gsxeclipse102]

Remove o2 housing and check for oil/wetness. If you find it wet remove manifold and check valves for wetness. If the valves are dry but turbine wheel oiley then the turbo is bad.

 

[kjlinwi]

Little more background.

I'm an ASE certified master Mechanic with 26 years experience.

Head was sent out to the same machine shop that I have sent all our shop heads out to including a cummins diesel head. I have never had a problem with their work before.

We purchased a Felpro head gasket kit and sent all the seals with the cams and the head and the valves, springs and retainers. First head I sent over there, they rejected as too thin. So i purchased another 97 head and this is the one that they rebuilt and that I installed.

The turbo was replaced before the engine was torn down and driven on it for a few months with no smoking. After the rebuild is when the smoke started. I checked the inlet for oil (none) outlet for oil (none) and then I pulled the blow off valve off the the intake tubing right before the throttle body. The tube is completely dry. If the turbo were the cause, then it would have to be feeding oil into the throttle body. So regardless how good the turbo looked at the inlet and outlets, for me the proof is in that the tube before the throttle body has not a spec of oil in it. If I'm wrong, I'd be willing to hear how.

Turbo has also been turned off thru ECM link and I can get it to smoke in neutral with no boost on at all.

Typically, I agree that valve stem seals leak mainly on start up. However, I had a car come in the shop a few years ago that was leaking oil out the exhaust (non turbo buick 3.8 rwd) and the customer wanted me to replace the exhaust manifold gaskets to seal the oil leak. Pulled the valve covers and installed new valve seals and it stopped leaking for awhile.

Coolant level in the radiator and the overflow can has been steady. Have not had to add any at all and it has idled for hours with no problems. Before the rebuild, there was some slow coolant usage.

The leak down test was done with the timing belt on and off. Doing so tests the sealing of the pistons and TDC and BDC. Results were the same. I couldn't identify where the loss was coming from and I honestly rarely do leak down tests so I did a leakdown on a 2016 corolla to compare to (easy access) and it was running 90 -95% on the cylinders so I decided to not chase that mouse down a gopher hole. Specs say that anything over 90% is good.

I seriously wondered about the PCV system, but again, I can find no oil in the vent line from the valve cover to the tube behind the air filter that feeds into the turbo inlet.

Or is it possible that only the exhaust turbine of the turbo seal is leaking and this is the oil leak into the exhaust? Going to have to pull the O2 sensor and boroscope that next I think.

Thanks,

Matt

 

[91talonts1]

The turbo was replaced before the engine was torn down and driven on it for a few months with no smoking. After the rebuild is when the smoke started. I checked the inlet for oil (none) outlet for oil (none) and then I pulled the blow off valve off the the intake tubing right before the throttle body. The tube is completely dry. If the turbo were the cause, then it would have to be feeding oil into the throttle body. So regardless how good the turbo looked at the inlet and outlets, for me the proof is in that the tube before the throttle body has not a spec of oil in it. If I'm wrong, I'd be willing to hear how.

Turbo has also been turned off thru ECM link and I can get it to smoke in neutral with no boost on at all.

A turbo cannot be turned off, in link, or any other way. Even if you could turn off the boost (only way to do this is have all the exhaust bypass the turbine), oil would still flow into the turbo. I would think a 26 year veteran mechanic would know that? I'm not sure what you've turned off in link, but it definitely didn't turn your turbo off, or prevent it from leaking oil.

Even with no oil in the intake tract, oil could still be leaking into the exhaust side. Which would cause oil to burn in the exhaust, but not the cylinder.

 

[kjlinwi]

You confuse turn off boost with spinning. Boost can indeed be turned off in link. It has the aftermarket waste gate solenoid that is duty cycled to allow it to create boost (or not). If the waste gate stays open, there is no boost. This happens all the time. Rev your car up in neutral as high as you want and tell me what your boost numbers are. Turbo won't boost without VSS input on even a stock car.

Note no smarty icons....

 

[91talonts1]

Turning that bcs off will just put you back to wastegate spring pressure. The wastegate is not staying open with it off. Just because it's not building boost on no load, does not mean the turbo is off. A turbo will build all the boost the wastegate spring will hold on a stock car, no vss needed.

The icon was intended to convey joking. Do you want help, or to be a smart ass?

 

[kjlinwi]

This is diverging off of the topic. Put a pressure sensor on the intake and measure boost directly with revving in neutral. You won't build any boost. Turbo will spin at a lower rpm, yes, but it won't boost.

The sentence in front of your icon does not convey to me what you say it does to you.

I already asked for input and help. And I answered all questions without any icons.

And I had already asked before your answer, "Or is it possible that only the exhaust turbine of the turbo seal is leaking and this is the oil leak into the exhaust? Going to have to pull the O2 sensor and boroscope that next I think."

I have not seen this happen before, but I think it could be a possibility. When I have time I will go thru with a boroscope and see what I can see.

Thanks,

Matt

 

[91talonts1]

I can build more than 20 psi in neutral man... The boost made free revving and in gear are only different because of load on the engine. It has nothing to do with boost control. I think you need to educate yourself a little more on the subject. That's not a jab. And I apologize if you took anything else I said that way, this is the internet after all.

I answered your other question you just skipped that part. Yes oil can leak into the exhaust side of the turbo and not the intake.

 

[kjlinwi]

Oh boy, start your own post, and post a video of your car building 20 psi in neutral.

Then I'll post my video of my car revving with no boost on the boost gauge and I'll then video the boost gauge feed line to the intake so you can see its hooked up and I'll even video the settings for you. Then I'll drive it on the video and show you no boost.

Then I'll change the settings on the video for you and re rev it showing no boost and drive it showing boost. I have over 20 hours playing with ecm link on this car just playing with settings and learning what I can and can't control.

Start your post and pm me the link.

Done wasting time here on it.

I did see your comment about turbos leaking only on the exhaust turbine side, which I have not seen before. I took that as reinforcement that it does happen. Which is good to know that it is possible. I will still have to check and verify though as it appears we are down to two options. Either the turbo just leaking on the exhaust turbine or the exhaust valves leaking. The history makes it more confusing, so I need to physically verify which it is.

I have replaced a bunch of saab, volvo, vw, Ford and Chevy truck, as well as a couple dsm turbos over the years and I have never seen one leak on just the exhaust turbine side before. I'm more than willing to admit that it never occurred to me that just the exhaust side could leak. Which is why in all my testing and inspecting, I checked the intake side hard for a leak and took the lack of oil as a sign that the turbo is good.

I will be checking it later today and posting what I find on that.

Thanks,

Matt

 

[livedsm4g63]

This may be why I never take my vehicle in to a "dealer" .

Try checking your turbo feed/drain. If the drain is kinked or blocked it will cause oil to burn.

 

[kjlinwi]

Tube is straight and clean. It was replaced when the turbo was and re-verified good during the rebuild while the motor was out and the head was off.

 

[91talonts1]

Okay man, I'll leave you to it then. I'm going to be waiting for that video of 0 psi though. If you wanna see one building boost sitting still simply Google dsm launch control.

Won't take much reading in this thread for you to realize how wrong you are. http://www.dsmtuners.com/threads/th...stegates-and-how-to-turn-up-the-boost.382016/

 

[mitsubishikid]

I used to have a burning oil issue after I rebuilt my 14b, but heres the crazy thing it all went away on it's own, I had to run the car a few times after the rebuild, but eventually all the issues went away, I used to see a cast of bluish smoke, went whitish towards the last few times and then all but vanished, I had to spool the turbo a few times with the 2 step engaged in link, because I did not want to get stranded on the open road, it was a few days of, starting the car, letting it reach normal operating temps, and then running it to 4500 rpms, for about 10 seconds, then letting off, and watching the smoke screen for a second, then doing it again after letting the car idle for a minute or so.

Another thing that I thought might be my problem after chasing all the obvious culprits was to search some of the rare problems, if your center section is cracked in the right spot oil can get into the exhaust.

but it could be that you need to drive it around a bit to work oil into the turbo and the seals etc. It took a few days for my smoking issue to dissolve away after my 14b rebuild I also had a new head installed on my car too.

 

[livedsm4g63]

That is very true, no one mentioned how many miles have been placed on the rebuild. Granted Ive done multiple with zero issues after.

 

[kjlinwi]

I'd agree, but it didn't smoke before the rebuild. Course I may just get frustrated enough to try it and see if it "heals itself"

I just did a classic example of what sounds good is very different than what works.

I pulled the O2 sensor housing as suggested by several to inspect and see if I can find the leak. Did the labscope first and saw a slighlty fouled O2 sensor when I pulled it (not bad aat alll) and got a great view of the turbine wheel. Not great enough so I pulled the housing. No puddles, slight oil spots on the housing, No drip or endplay movement from the turbine. Pulled the housing, can see some wetness behind the turbine on the outside edge of the cartridge inside of the turbine wheel. Nothing on the center shaft.

So i pulled it for a better view and nada. Still learned nothing more.

Sounded like a great idea, didn't work so well in real life.

I think next is pull the exhaust header off the head and see if its wet on the back of the valves.


Matt

 

[jimsgsx]

So you saw wetness behind the turbine wheel. Well there ya go! Only takes a drop of oil to burn and make a cloud. Put in some new seals and or rebuild the turbo. If it was valve seals it wouldnt smoke while driving it, mainly smokes at idle and dissipates away when driving. Just my 2 cents.

 

[kjlinwi]

Shouldn't I see wetness at the center shaft of the turbine wheel? And since the exhaust valves feed the exhaust through the turbine wheel, how do I know that the oil isn't from the valve seals and the spinning of the turbine isn't throwing it to the outside edge of the cartridge/housing?

 

[gsxeclipse102]

Yes your next step is to remove manifold and check valves. If theyre dry then the turbine wheel seal( like a piston ring) is worn out. The other cause of turbine leaks is a partially clogged drain. When you reinstalled the drain did you use a new gasket or rtv sealant? Sometimes the rtv blocks off the drain hole.

 

[kjlinwi]

I used a gasket at the bottom of the drain with the Right Stuff on the bolt threads to seal off bleed through from inside the pan.

 

[Tyeler18]

At this point you need to pull the manifold. You've found traces of oil in the exhaust tract, pulling the manifold will prove if it's head/engine related, or turbo depending on where the oil is coming from.

As far as turning off the turbo in link, he's right. It doesn't work like that. You are limited to the minimum amount of pressure your wastegate spring is rated at. An MHI wastegate spring is 7-8PSI for a 14b/16g setup. You can turn the duty cycle to zero in link, you can loop the wastegate line to the wastegate itself bypassing the controller completely, it will still boost 7-8psi. You can't "shut it off" completely, even with the wastegate tied fully open it'll still have flow past the turbine wheel allowing boost to build to an extent. If you're turning anything off in link and driving around with 0PSI of boost you have a mechanical failure, boost leak, bad turbo etc. Just want to clear that up for someone who may find this thread later on.