So I'm attempting to diagnose my oil consumption problem and I am thinking I may have a ringland or something going on with one of the stock pistons. This morning I attempted to do my first leakdown test. I have a Snap-On cylinder leakage tester. Tested each cylinder by ensuring piston was at TDC, valves closed, and hooking up the shop air. All cylinders tested between 15-20% leakdown. That is less than half of what the gauge says is low or normal. The gauge reads (%) 0-40(green) 40-70(yellow) and 70-100 (red). What is normal or acceptable leakdown %? Any input is appreciated. -Aaron
warm the car up and test it again its good to get the reading both hot and cold. 20% leakdown is not good
The car was warm. I did it right when I got to work this morning. It was at normal temp. Should I do it when it's cold and compare the #'s? What about the test procedures? Is it necessary to turn the crank and try to compress the air in the cylinder?
Is there gonna be a way to tell what is going short of tearing it down? I looked with a bore scope and don't really see any damage around the crown or on the walls. It also doesn't have any drivability issues, no smoke, no misfire...
What do the plugs look like? If all the cylinders are the same then its a really hard call to make. Not to say that you did it wrong, but you might want to have it professionally done before you tear into the motor That wouldn't be good. Take a look at the plugs and tell me what you see. Also tell me how old the plugs are. One more thing, how much oil are you consuming? Does the car smoke at all in any conditions?
The plugs all look normal. There is no sign of oil on the electrode's, just a normal light brown. Plugs have about 25K on them. The car has 65,XXX. The oil consumption has been progressively getting worse over the past year I'd say. I change the oil every 2K at the most. Problem is I probably add 2qt's between oil changes! The oil consumption is dependent on how hard I drive the car obviously and how much boost it sees. Just for reference, I changed the oil and drove to work and back for a week running the car at 25psi(what it is tuned for). Not wide open all the time but on occasion. In 100 miles I was 1qt low!!! And it doesn't smoke!! I have yet to pull the downpipe to make sure the turbo isn't leaking, but wouldn't you think it would smoke if the turbo was going out? If i drive like Morgan Freeman it does much better, but it's not as fun. I know something is wrong, just would like to pinpoint the problem. Oh, I am a tech for a Honda dealership, not to say that qualifies me to perform leakdown tests on Subaru's, but I got detailed instructions from 2 different Subaru techs before I started.
Did you run the pressure gauge to 100psi? The % is based on 100psi going into the cylinder. If you had 20% leak down the engine would not run, period. Also, 1qt per 1k miles is rather normal for most STi's. What are the compression #'s?
Oh and change your PCV valve assembly or install an AOS since it sounds like you are running a larger than stock turbo.
What turbo are you running, and what type of crank case ventilation do you have? Subaru motors with stock ventilation for the crank case don't like 25psi very well. Crawford AOS might just do the trick if you don't have one. Is the bottom end stock? Or does it have forged internals?
Yes, I ran the gauge up to 100%, then attached the air to the cylinder. It gave a reading of 15-20%. I disconnected the fitting and the gauge read 100% again. I have been running an IXIZ AOS for over a year now as I thought that would be an easy solution. I haven't replaced the pcv, but I inpected it to make sure it wasn't stuck or clogged. I can agree that 1 qt per 1000 miles is acceptable with my set up, but 100 miles? Also, the oil gets BLACK, BLACK-BLACK after about 500 miles. Running 10W30 Castrol Syntec.
: That's what I was thinking. It's seems high, but not really too excessive. 1QT in 100 miles is WAY EXCESSIVE.:eek3:
Take it out of the system. An AOS should not be used with the PCV valve hooked up as if stock. Strange. 100 miles is way too much. super black after 500 miles is also bad. It could be that all 4 pistons are broken. What are the #'s for compression?
Last compression test was probably 6mths ago all cylinders were 130ish. All were the same, not positive on the exact # but all were above 130. So the PCV should be taken out completely? I followed Ming's instructions when I installed it, even sent him pic to make sure everything was straight. If I remove the PCV, what do I do with the vacuum line from the intake? Just cap it off?
Cap it. I believe Ming has revised his instructions. But yes, remove the PCV valve. Basically you are probably leaking boost into the crank case through the stock PCV (not designed for 25 psi). That is causing extra blow by. Also possible you have an issue, but that would be the first change I would make.
No worries, just make sure and plug the appropriate hole on the intake manifold. And tighten it down good so it does not pop off under boost.
Like the others have said, remove the PCV valve from the system and cap both ends off according to the AOS's revised instructions. PM Ming, I'm sure he can give you the updated instructions. If your compression is 130 or more all around, it is very unlikely that you have any sort of piston/ring/ringland issues. Most STis will compression 110~150 within 10psi of each other and they are golden. Typically a broken ringland is 30psi or more difference from the rest. Recheck the compression numbers, block off the PCV valve, drive it and report back.
My car with the cracked ringland was : 140, 140, 70, 140. Thanks to WJM and Liquidforce for fixing it! BTW: I can't PM you WJM!
When running, no. During star up, if the starter hits the down cylinder it will for sure sound 'off', almost like no compression compared to the others. I've had several vehicles come in that I diagnosed a cracked piston w/out doing a compression test....the starter test is pretty obvious when you have that large of a difference, especially on one cylinder.
Yeah, I never have any weird noises during start up either. I removed the pcv and capped off the vacuum line last night before I left work. I beat on it a littlerofl on the ride home, as much as I could with road conditions. Maybe I'm just feeling hopeful, but I think I noticed a difference. There didn't seem to be as much exhaust visible when I got on it. You know how you can see when someone's headlights are in your rear view. It pulled really nice, but maybe that was due to the 19 degree temp. Gonna give it a few more miles before I can say for sure though. Thx guys.
Sorry for the threadjack. So what exactly does it sound like/do when you start it on the cracked piston? I ask because pretty often when I start mine it bogs pretty bad and RPMs drop pretty low, then evens back out. But other times, it starts just fine and does not do this. Could this be a case of a cracked piston? I have tried to find out why it does this but I can't find anything that would cause this until I read what you said.
Sounds like a startup tuning issue. The sound is hard to describe, but if you've ever heard an engine try to start w/out compression, then you'd know it.
Ok guys, I thought it was doing better. I changed the oil on 1-9 and drove it hard and often for a week with the boost turned down.(1 BAR) I checked the oil every day. It seemed to be fine, the oil level stayed at the full mark for the whole week. Then I decided to turn the boost up (25psi). I drove it for another week. Checked it Monday and it was almost 1qt low! Almost to the first dot on the dipstick and the oil is black, suggesting blow-by. The oil consumption definitely seems to be boost dependent. I just got done doing the compression test and all cylinders read between 120-130. Is it possible for the turbo to be going out, and not smoke? There is no smoke on start up or idle. Maybe the rings are worn, and not applying enough pressure against the cylinder walls under higher boost levels? How am I gonna diagnose this?
Are you running 25psi on pump? without knock? ballsy... what turbo are you running? I'm assuming a manual boost controller? Sorry if these have already been asked...
The higher the boost, the more blowby you will get. I recently read a comment by the guys at Peterson Fluid Systems (make dry sump, AOS, etc) who said that as much as 2% of boost goes by the rings. So on a 1000 cfm turbo that would be around 20 cfm. This air will pressurize the crank case pretty good. There are a couple different things to try depending on what you are comfortable with. Could it be the turbo? Yes it could. Could it be an ineffective AOS? Yup, that too. I am not sure the air slows down enough in those things to really allow the oil to collect. It could also be that it is not able to flow enough air out and back into the intake to relieve all the pressure. There are lots of possibilities but no firm answers that I have seen. We have a couple things we are working on now to test this all out. But big boost in Subarus with such a small crank case seems to lead to these problems.
Pretty much. Aaron: I think its time someone invented some magical catchcan/AOS to combat this bullshit. Your compression numbers are good. At low boost you use nearly nothing for oil. If your turbo was bad you would be fogging up the road. Your theory on high boost and weak rings was the next thing I was going to visit: Yes its possible. How many miles are on the engine? How many miles with the current setup? (didnt you buy/trade this car from SOG ~3 years ago?)
Well this sucks. So, it's possible/probable that the ixizAOS is insufficient in high boost applications? I tried to get a hold of Ming, but he's in Germany until this weekend. I'd be interested in seeing the couple of things I could try. I just hate to guess at the problem. I DO notice a bit of oil in the turbo outlet pipe, but the IC isn't full of oil or anything...?
The engine currently has 65,XXX miles. It has been running the same setup for close to 50K miles. The Green that I bought the car with failed a couple years ago, I never determined why. I'm thinking it might have something to do with running it hard while low on oil. I don't know. I bought the AOS shortly after as I thought it would stop or at least slow down the oil consumption. I replaced the turbo with a used Green that Bob just happened to have. Seemed a little sketchy at the time, I knew nothing about where it came from, but I needed a turbo and he had it so.... I also have been wondering if the meth could have any effect on the rings or cylinder walls. Isn't methanol highly corrosive. Ring or or cylinder wall damage should show in a compression or leak down test though, correct?