The Top DSM Community on the Web

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. Log in to remove most ads.

  • Join the Community!

    DSMtuners is a massive archive of DSM information - but more importantly, it's a COMMUNITY! Join in and participate with other DSMers, and invite all of your DSM friends to make this place their home. Chat with others, create a build thread, post questions and answers. Get involved! Logging in will also remove many of the advertisements, along with this notice. ;) It will also allow you to view images in threads.

Please Support Morrison Fabrications
Please Support ExtremePSI

Leaking oil from exhaust manifold

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
I decided to chase down an oil leak my car has that I never could figure out. It was dripping from the coolant large water pipe and dripping off that onto everything making it difficult. I pulled the exhaust manifold off and noticed three out of the four exhaust ports on the head had an oil like substance inside of them and on the mounting surface where the gasket mates. I did some reasearch on here and found a few threads where people stated they had similar issues. Some older threads didn't get resolved, one I believe was resolved by being the pistons rings, another I believe was too much oil pressure or something like that. People also suggested valve seals/guides.

This engine has about 15,000 miles on max since it was rebuilt. I bought it from someone on facebook believe his name was Cody Murray. He rebuilt it putting in wisco pistons, BC springs and retainers. I am not sure if he did new valve seals or any of that though. I may try and reach out to him to see if he remembers what all was done to give me an idea if it could be valve seals/guides or what.

I was getting a little smoke at idle sometimes and I was going through wideband sensors quickly. There is also a little oil on the air filter side of my turbo so I guess it could be the turbo leaking through the intercooler and in to the intake possibly or do you think the oil would be burnt during combustion?

My research also shows that one of the exhaust bolts goes into the coolant passage and that it sometimes can leak there. I don't think thats the case here because it looks like oil to me and it is actually inside the exhaust ports themselves.

Forgot to mention, I don't see any signs of the valve cover leaking down onto the manifold/gasket and getting in that way so pretty sure that is ruled out. I posted a picture of the manifold gasket and most of the oil on it is from around the bottom area with top being dry.

So I guess my next steps are to do a compression and a leak down test. I will post my results here and let you know what I see.

The oil is in the manifold ports 1-3 with the 4th port nearest the timing belt being dry.

I did forget to mention some of the exhaust manifold nuts were a little loose. I cleaned all the oil up from the engine and in the ports torqued the manifold down and let it run for about an hour. I didn't see any more leaks so I don't think its coming from the head gasket and somehow making it up there unless it only happens under boost as I let the car idle only and did not drive it.

When I go to do my compression and leak down tests should I do them with engine warm?

You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.
 

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
Also adding, I did check the PCV valve. It was replaced when engine was swapped a few years ago. PCV valve rattles, I can blow air through it from the valve cover end. If I try and blow air from the intake manifold end it doesnt flow.
 

shellhouse

Supporting Member
100
34
Nov 8, 2021
Atlanta, Georgia
Leakdown test should help to troubleshoot.

Oil only on the air filter side of the turbo? How are the intercooler pipes and TB/IM? Can only guess that the turbo inlet oil may be from BOV recirc, however it's curious.

Edit: from some of the pics it looks like there's a lot more oil from the valve stem seal area, and toward the combustion chamber is dry
 
Last edited:

TK's9d2TSi

Supporting Member
5,767
2,885
Sep 11, 2017
Cincinnati, Ohio
Also adding, I did check the PCV valve. It was replaced when engine was swapped a few years ago. PCV valve rattles, I can blow air through it from the valve cover end. If I try and blow air from the intake manifold end it doesnt flow.
All pcv valves leak even oem. Go with a one way check valve from US plastics that’s oil resistant. Like mentioned about, it’s most likely either valve guides or valve stem seals.
 

steve

DSM Wiseman
14,074
1,112
Feb 3, 2002
Boulder, Colorado
I did check the PCV valve. It was replaced when engine was swapped a few years ago. PCV valve rattles, I can blow air through it from the valve cover end. If I try and blow air from the intake manifold end it doesnt flow.

There is no way you're going to make anywhere near the pressure the PCV sees from the intake manifold under boost with stock human lungs. You have to use a compressor.
 

1990TSIAWDTALON

Moderator
8,563
4,463
Nov 14, 2013
Independence, Kansas
If you decide to check the valve stem seals, let me say that if the stem wiggles in the guide, seals will last about 2 weeks to a month and be bad again. We have done that, and that is how long they lasted before the car started smoking again.
This is what our car did when it had bad guides with new seals after less than a month.
You must be logged in to view this image or video.
 

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
Alright guys, I did some tests for everyone and got the results here. I performed both a compression and cylinder leak down test. I performed the compression test because I would like to know the health of the pistons and rings. If the bottom end also needed work I would maybe just consider trying to find a rebuilt engine somewhere other then trying to tear into this one.

So anyways, lets get on to the tests. I warmed the engine up to normal operating temp letting it run for about 20 minutes. I pulled the plugs and took some pictures that I have included in this thread. I didn't really see any oil on them. I then performed a compression test twice for each cylinder. The engine has Wisco pistons and stock rods with a 6 bolt bottom end with 2g head. Compression test as follows. Keep in mind, for all my tests when I say cylinder #1 I mean the far left cylinder closest to the battery and when I say #4 I mean the cylinder closet to the timing belt going left to right.

Cylinder #1: 180 tested again at 180
Cylinder #2 180 tested again 180
Cylinder #3 170 tested again at 170
Cylinder #4 185 tested again at 185

I then dumped a small amount of oil in each of the cylinders and ran the test again

Cylinder #1 196
Cylinder #2 205
Cylinder #3 187
Cylinder #4 203

So from what I understand my compression seems to be ok, however cylinder #3 is slightly lower than the others. All are within 10% of each other and none of them increased a large amount by adding oil. When I added the oil i didn't measure it or anything scientific, just dumped some in each cylinder so I would imagine some cylinders had more oil in them then others.



I went on to do a cylinder leak down test as follows. I rotated the crankshaft until the little notch lined up with the line that says T or whatever it shows on the timing cover (the last marking). The cams lined up with their timing marks also. Please look at my photos I provided to make sure everything looks right. I then went on to test cylinder #4 as I call it LOL or the one closets to the timing cover. From my reasearch on here from a prior thread I saw that cylinders #1 & #4 are supposed to be at TDC at the same time and when cylinder #2 is at TDC so is #3 at TDC. I am not sure if this is true or not, but I performed my tests as if it is.

Cylinder #4 set the gauge so it was pushing 100 psi into the engine, the gauge to the right read 97 so loosing 3 psi
I then went on to check Cylinder #1 since it should also be at TDC without changing anything according to that prior post I read.
Cylinder #1 100 psi in held 94 psi so 6 psi loss.

According to the service manual I have cylinder #3TDC is approximate 80 degrees from cylinder #1 TDC. I turned over the crank while I stuck an extension in cylinder #3 spark plug hole. Once the crank was approximately 80 degrees from the original TDC, the extension in the spark plug hole appeared to be at the highest so i tested #2 & #3.

Cylinder #2 100 psi in and it held 96 psi so only 4 psi loss. Since cylinders #2 & #3 are supposed to be at TDC at same time, I went on to test cylinder #3 without touching the crank.
Cylinder #3 100 psi in and held 23psi.

I rotated the engine around again so the timing marks on the crank and cams lined up in the photos I included in the thread and performed cylinders #1 & #4 tests again. They tested similarly only loosing 5-6 psi. I rotated the engine over so #2 & #3 were at TDC and they both tested the same again with cylinder #2 only loosing 4 psi but cylinder #3 loosing 77psi.

I am assuming since cylinder #2 matched what the others lost did means that I didn't mess up the TDC for #3. If thats true that #2 & #3 are at TDC at the same time, cylinder #2 would be loosing just as much as #3 if I wasn't completely at TDC for #2?

So now on to what I heard:

Basically all cylinders I could hear a slight hiss from the oil dipstick and inside the oil fill cap. Nothing extreme, just a small hiss. I could put my finger over oil dipstick hole and here a change. I didn't see any bubbles from under the radiator cap for any of the cylinders. Sometimes I would hear a slight pop from inside the spark plug hole and then air would start hissing out from inside the cylinder hole being tested. Normally once I got up above 70 psi it would pop and start hissing. I thought at 1st it was the o ring on the cylinder leak tester but I was putting oil on it to lubricate it and it is not torn or anything and no matter how tight I turned the tester into the cylinder it didn't make a difference in the amount leaking. I believe it is is actually coming from inside where the spark plug gasket that goes around the spark plug hole seals the valve cover gasket from the spark plug hole is. It did this for 2 or 3 cylinders.

With the cylinder #3 that was loosing so much pressure, it hissed from all the above mentioned but also hissed from the the BOV vacuum line when i pulled it off the intake manifold. It mostly sounded like it was coming from cylinder #1 spark plug hole. I threw a spark plug in #1 and it definitely quieted down the air leaking noise.

So that is where we are. From what I understand the compression test was ok and that the pistons and rings are probably fine. Cylinders #1 #2 & #4 appear to only be loosing 5-6 ps of pressure from what I am putting in. Cylinder #3 is loosing a crap ton for some reason. It is also reflected in the compression test I guess because it was 170 psi instead of the 180-185 the others showed. If this is a valve seal issue for #3 why would that cause cylinders #1 through #3 to have oil coating inside like seen in the pictures? Is there a way to test valve guides without pulling a lot of stuff apart? If I know this is just a top end issue, i guess my next step would be to pull the head and send it to a machine shop to have the seals/guides replaced and be machined. I guess this explains why I was going through wideband sensors left and right and why it would smoke a little sometimes and also why I had a boost leak when I was boost leak testing it that I could never find.

You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.
 
Last edited:

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
the car sometimes would smoke a little yes. It seemed to go off and on sometimes but yes I did see smoke
 

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
yes I had the cams on the engine when i did the leak down test. i have a friend who is a mechanic coming over to make sure i did everything right today. I guess we could pull valve cover and check and make sure that cams are not pressing down on the valves for #3. c like iIO an anyone verify that the two center cylinders reach TDC at same time like i read? The car has stock cams I believe so nopt sure if one cylinder starts to open its intake valves before the other or not
 

TK's9d2TSi

Supporting Member
5,767
2,885
Sep 11, 2017
Cincinnati, Ohio
Although two cylinder (1&4 2&3) pistons will be at the top of its stroke at the same time, only one will be at tdc WITH valves closed which is when you’d want to perform leak down test for that cylinder.

I find it odd that you have air leaking into another cylinder that’s not right next to one another.
 

DSMPT

DSM Wiseman
1,747
1,307
Jun 12, 2014
Japan / Mexico, Arizona
yes I had the cams on the engine when i did the leak down test. i have a friend who is a mechanic coming over to make sure i did everything right today. I guess we could pull valve cover and check and make sure that cams are not pressing down on the valves for #3. c like iIO an anyone verify that the two center cylinders reach TDC at same time like i read? The car has stock cams I believe so nopt sure if one cylinder starts to open its intake valves before the other or not
You are correct that 1/4 and 2/3 have TDC at the same moment but that's only about "piston position". The way you think would be fine if you do the test without cams/timing belt installed. But if you do the test with cams/timing belt installed, you need to do the test at TDC in compression stroke. Keep in mind that the firing order is 1-3-4-2, so when you set the piston #2 at TDC with both valves closed which means it's in compression stroke, #3 piston is at TDC at the same moment but it is in exhaust stroke, which means valves are open.
 
Last edited:

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
So from here on out I am going to correctly reference cyl #1 as the one closest to the timing end and #4 being closest to the battery. Thanks for bringing that to my attention about me counting them wrong.So ignore what I said in my initial test about counting them left to right.

So I was definitely wrong on trying to test it the way I was. I pulled the valve cover and reran the test starting at cylinder #1 closest to the timing belt at TDC and then rotating the crank 180 to get to the next or cyl 3, then rotate another 180 back to the timing mark on crank to get to the next cylinder or #4 and finally another 180 to get the last cylidner #2. So #2 was the one that was leaking horribly when I was testing it wrong. Testing each cylinder putting 100 PSI in I got the following:

Cylinder #1 said Set or less than 20% loss
Cylinder #2 said 20%
Cylinder #3 said set or less than 20% loss
Cylinder #4 said set or less than 20% loss.

That was using my friends Snap On leak down tester.

Using my harbor freight one I reran it again putting 100 psi in and got:

Cylinder #1 96 or only 4 psi loss
Cylinder #2 90 or 10 psi loss
Cylinder #3 97 or 3 psi loss
Cylinder #4 94 or 6 psi loss

So definitely seems like cylinder #2 is loosing more than others but it still is within the green 20% range on the snap on cylinder leak tester. It almost sounded like I heard air in the intake manifold flowing through but not leaking out anywhere. I pulled the BOV vac line off intake and didn't hear any air coming out

I pulled the exhaust manifold again and noticed oil in all four exhaust ports now instead of 2, 3, & 4. Now keep in mind I did dump some oil in each of the cylinders when I was doing the compression test so it could very well still be from that but I did idle the car for 15 20 minutes after doing the compression test to warm engine back up before doing the leak down test so not sure if that would be enough to burn that oil in the exhaust valves out or not.

So do you think I am still probably still looking at valve seals/guides being bad? From what I understand sometimes they don't always show up in a leak down/compression test.

When I first pulled the exhaust manifold before doing any of these tests or putting some oil in each of the cyldiners, cylinder #1 exhaust port appeared to be the most dry with little to no oil inside the exhaust port and it had the highest compression at 185. The other 3 cylinders that had oil in the exhaust ports from the beginning had slightly lower compression at 180/180 with cylinder #2 being the lowest at 170.
 
Last edited:

TK's9d2TSi

Supporting Member
5,767
2,885
Sep 11, 2017
Cincinnati, Ohio
Like previously mentioned, I think your #2 cylinder valves aren’t fully seating. You’d have to pull the head to take a closer look and inspect the valves in the guides to see if any are loose. If you don’t want to take that route, you can just replace the valve seals with the head still on.
 

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
Like previously mentioned, I think your #2 cylinder valves aren’t fully seating. You’d have to pull the head to take a closer look and inspect the valves in the guides to see if any are loose. If you don’t want to take that route, you can just replace the valve seals with the head still on.
Ok that's what I was afraid of but it does explain why I was going through so many wideband sensors, why I couldn't get the engine to hold boost when I boost leak tested it. I messaged the guy that built the engine trying to ask him what all he did to the head. I know he said he did BC springs and retainers but don't know if he did valve guides or anything and he isn't replying to me so who knows.

I am just gonna pull the head and send it to a machine shop to have it pressure tested, decked whatever else I guess. No sense to go through all the trouble to replace valve seals only to take the chance a valve guide is bad. Might as well look for some upgraded 272 cams to have put in it as well. Thanks guys for the help. I will probably start a new thread detailing the tear down process as a way of keeping track of everything as I never did this before. I see a couple helpful videos on youtube that show how to pull the head so that will be no big issue.
 

DSMPT

DSM Wiseman
1,747
1,307
Jun 12, 2014
Japan / Mexico, Arizona
So I was definitely wrong on trying to test it the way I was. I pulled the valve cover and reran the test starting at cylinder #1 closest to the timing belt at TDC and then rotating the crank 180 to get to the next or cyl 3, then rotate another 180 back to the timing mark on crank to get to the next cylinder or #4 and finally another 180 to get the last cylidner #2.
You got it. You can know which cylinder is in compression stroke by seeing the cam dowel pin position on cam gear like the pic below. 180 degree at crank is 90 degree on cam.

So #2 was the one that was leaking horribly when I was testing it wrong.
Probably valves were open. It's the same thing as I mentioned above, if you tested without rotating the crank when you moved from #3 to #2, #2 was in exhaust stroke.

So do you think I am still probably still looking at valve seals/guides being bad?
Could be or could be oil control rings or worn/scoring on cylinder wall etc. I would suspect cylinder wall condition. A bore scope would be useful in that case.

You must be logged in to view this image or video.
 

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
You got it. You can know which cylinder is in compression stroke by seeing the cam dowel pin position on cam gear like the pic below. 180 degree at crank is 90 degree on cam.


Probably valves were open. It's the same thing as I mentioned above, if you tested without rotating the crank when you moved from #3 to #2, #2 was in exhaust stroke.


Could be or could be oil control rings or worn/scoring on cylinder wall etc. I would suspect cylinder wall condition. A bore scope would be useful in that case.

You must be logged in to view this image or video.
my friends a mechanic and I may borrow his borrow scope and post what I find. When I pulled the plugs originally they didn't look wet with oil though. The pictures I posted earlier in the thread where how they looked when I first pulled them.
 

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
So I just some photos with the boroscope. All cylinders had a small pool of oil on top where the valve indentations are on the top of the piston. Keep in mind when I did my compression test I did dump some oil down in the cylinders to see if it changed anything. Although I did idle the engine about 2 or 3 times after doing that so not sure it would burn all the oil up by idling it or not. It wasn't driven or reved or anything, just idled. Had a really hard time getting the borocope to aim at the cylinder walls so I got a few photos showing a random cylinder walls. I also stuck it inside the exhaust manifold to try and get a better look at where the oil was coming in.

The best photo of the valve area is the one where the valve is open. You can see the line of oil where it appears to start where the valve is normally closed. Since the valve is open, you can see some clean area that then as it gets deeper into the head appears to show oil. Also from the valve seal end of the valve there is what appears to be a drop of oil that was thicker there.

I cleaned all the oil out of the manifold ports with a rag. I left the spark plugs out, disabled spark and fuel and cranked the engine several times, repeatedly giving the starter time to cool. According to my aftermarket oil gauge that is reading pressure at the oil filter housing, I was building about 70 psi of pressure while cranking the engine. Oil was definitly making it up to the cams and around the lobes. After cranking for several times I didn't notice any oil in the exhaust manifold ports except for maybe the above mention possible drip or wet spot near the valve seal area. (not 100%) sure that wasn't there before doing this test though. Also the level of oil on top of the pistons did not appear to increase from before I did the test.

You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.


You must be logged in to view this image or video.
 

TurboSpider

Proven Member
159
10
Nov 2, 2015
York, Pennsylvania
Just to update everyone, I pulled the head and dropped it off at machine shop. Some of the valve guides are out of spec but the intake valve shafts are all oil coated. The machine shop wants me to pull the pistons to check rings.
 
Last edited:
Support Vendors who Support the DSM Community
Boosted Fabrication ECM Tuning ExtremePSI Fuel Injector Clinic Jacks Transmissions JNZ Tuning Kiggly Racing Morrison Fabrications MyMitsubishiStore.com Raven Fabrication RixRacing RockAuto RTM Racing STM Tuned VR Speed Factory

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds

Top