Weird smoke Issues

[96gsdsm]

This problem has been plagueing me for over a month now and I'm but here goes...

I bought my tsi (profile is up to date) about 3 months ago. It was rebuilt with all OEM parts from the top down 6k miles ago by the previous owner. I have reciepts and everything. Well... the s16g that the car came with looked like it sucked in some debris, because I could see dents in the fins. I figured I would just drive it untill it went and get myself a nice new Evo3. So one day I let the car warm up, leave the house, hit boost ... and literally 1/2 ways down my block I glance behind me and see the dreaded white plume (synthetic white) of a blown turbo. Before this there was absolutely no smoking at all. period. Ok, no biggie... I bought that evo3 I always wanted... Installed it... started it up... more white smoke. I figured it was burning off so I let it idle for about 1/2 hour untill It got darker and I could open it up down the block. Same thing... smoke EVERYWHERE. So (to shorten this story up some) that evo3 had a 3/4" sliver of metal jammed in one of the turbo bearings, I shipped it back to Forced performance, and they rebuilt it under warrenty and I recieved the rebuilt one today.

I installed the turbo again.... same thing. It's still eating my oil up. So I actually removed the turbo, took it apart, and checked the bearings for more crud. ... nothing... absolutely perfectly clean. So I reinstall it and it still smokes. All the time. Idle, under boost, cruising ... everything.

Keep in mind the car ran PERFECT before all of this.

I removed the elbow that connects to the compressor housing and I can see a small pool of oil at the bottom inside the compressor housing. ... So I hooked the MAF up directly to the throttlebody elbow... bypassing the turbo's intake system... and it STILL smoked. I can see oil being forced through the crack in my crappy 1g exhaust manifold ... and it also drips slowly out if the 1/8" wide hole in my flexpipe. (1 drip every 3 seconds at idle)

There is no oil in the antifreeze or vice versa.

Compression is (#1-2-3-4)
dry 140- 90 - 90 -100
wet 175-100-110-150

I'm at my wits end. Any suggestions/tips/hints would be great.

 

[fourreGsixty3]

With the huge difference in compression dry test to wet test my bet is on your rings. Looks like oil is getting pushed past them, and thats why its smoking. Time to replace those suckers. Also on an added note service limit of a 2.0L 4G63t 2G is 133psi. The piston rings in cyls. #1 & 4, are worn, if the wet test improves compression by more then 30psi they are worn and need replacing, the rings in cyls. #2 & 3 are totally shot because they are at 90psi (43psi under service limit), and since the oil didn't help seal them that much they are worn so bad that the oil just passes straight past them for the most part. So with that being said I will say this one more time, "Your rings are gone, so you need new ones."

Dustin

 

[96gsdsm]

I'll try to get my hands on an air compressor and do a leakdown test asap.

Hopefully it ends up being the valvestem seals
I'd rather not rip the block apart to change rings and have it end up being the valve stem seals, or vice versa.

Although piston rings do look to be the culprit, I don't know that I'm ready to accept that.

What could have caused something like this on a practically brand new motor???

 

[96gsdsm]

Also I should note that oil is still leaking past the compresser housing (turbine housing unknown) I can see a small pool of oil with the compressor outletpipe detached.

This is on a brand new evo3 16g, removed twice and disassembled once, to be sure that there is no oil blockage causing it to squeze past the turbo seals.

Right now I'm using 10w 30 Mobil 1 and I'm about 1/2 quart low (to bring down oil pressure) ... and it is still doing this.

 

[fourreGsixty3]

What is your oil pressure at? All arrows are pointing to it being rings. It is quite odd that this happened all of a studden when the S16G let go. There is a thread somewhere on the forums where a kid had problems with brand new rings working fine, and the next day, they were totally shot. Let me try and find it for you to perhaps give you some more idea's.

Dustin

 

[96gsdsm]

It reads very low, just a hair above the needle at idle. ... But this is my stock gauge and I'm not 100% sure how accurate it honestly is.

Before this at idle it would read 1/2 way in between 0 and the first marker line, at cruise the needle would usually sit right blow the first marker line ... and when boosting hard it would be right around 75% between the 2 normal operating pressure lines.

 

[96gsdsm]

Man I keep forgetting things...

I'm actually not sure if the gauge is working at all right now.

 

[96gsdsm]

What kind of pressure should I use for this leak down test?

... I don't have the special gauge thing, but I do have a nice air compressor with adjustable output pressure. I'm planning on just filling the cyclinder with air and listening for leaks at the intake manifold, out of the 02 housing, and in the valve cover by means of the oil cap to try to pinpoint this thing.

 

[groundPork]

You need to test at at about 100 psi. It'll give you a some what realistic pressure situation in the cylinder (30-50 is nice if sealing, but is a little too sensitive). But under the cercumstances, I would agree with the other guy... your rings are not sealing. And I also agree the occurance along with tthe turbo failure is odd and it makes me think they are related. You need to realize that if the engine runs lean because of not enough fuel it can potentially detonate, breaking the second land which can stop the two lower rings from controling oil propperly. You oil pressure sounds like it is very low, and the comp test #'s sound real bad. The oil in the compressor housing, might warrant a check up by FOrced to make sure it's not that. The two situations sound related yet, the comp test numbers would not likely suck because of a leaky turbo. I know I'm belting out a bunch of scenarios, but these are things you should seriously concider.
I would say all else being ok, a hone, and set of rings at least. maybe new pistons and an overbore even.

 

[96gsdsm]

You need to test at at about 100 psi. It'll give you a some what realistic pressure situation in the cylinder (30-50 is nice if sealing, but is a little too sensitive). But under the cercumstances, I would agree with the other guy... your rings are not sealing. And I also agree the occurance along with tthe turbo failure is odd and it makes me think they are related. You need to realize that if the engine runs lean because of not enough fuel it can potentially detonate, breaking the second land which can stop the two lower rings from controling oil propperly. You oil pressure sounds like it is very low, and the comp test #'s sound real bad. The oil in the compressor housing, might warrant a check up by FOrced to make sure it's not that. The two situations sound related yet, the comp test numbers would not likely suck because of a leaky turbo. I know I'm belting out a bunch of scenarios, but these are things you should seriously concider.
I would say all else being ok, a hone, and set of rings at least. maybe new pistons and an overbore even.

I'm not even going to bother with a leakdown test. If I unscrew my oil cap mad smoke comes out, there's also alot of fast moving air going through my catch can as I can actually hear it whistle.

My assumption is that the pressure building up inside the crankcase is back-flowing the oil return tube and causing oil to be forced through the seals. This sounds reasonable enough to me, let me know what you guys think.

Regardless, valve stem seals are being replaced, my 1g rods are going to be machined to accept 2g 8.5:1 pistons, and while I'm at it I might as well pull the balance shafts out.

The turbo is going to be the least of my worries untill my compression is withen specs.

Thx for all the help guys

 

[96gsdsm]

Well... I pulled the head today....

lets just say some of the valves are sealing better then others... none hit pistons... they just don't seat right

Looks like all I have to do is re-build the head.

The cyclinders are glass smooth, no scoreing, nothing. Pistons look A+ ... everything checks out but valves.