Turbocharger Oil Seal Leakage Explained
 

Even though I don't have a single turbo car at the moment, I understand the frustration that turbo oil seal leakage can cause. Given such, I wanted to take some time to explain how the seals in a turbocharger works and why they sometimes leak oil. The picture below is the cross section of a ball bearing center housing. I labeled the main parts pertinent to oil sealing.

http://i7.photobucket.com/albums/y269/h0mesauce/spd.jpg

The seals in almost all turbos today are piston ring seals. They look like piston rings that are compressed when they are installed to leave minimal gap between the ends of the rings and around the rings. This forms a torturous path for the oil to prevent leakage into the air passage. Most performance turbos use only a single piston ring on the compressor and turbine sides because they do drag the rotation of the turbo which reduces mechanical efficiency. In applications sensitive to oil leakage for emissions, etc. it is not uncommon for 2 piston rings to be used on both the compressor and turbine sides.

On the compressor side, generally there is some kind of oil splitter which throws oil away from the seal as it is spinning. Because the piston rings and the splitter are dynamic seals, when the turbo is spinning faster, the better the seal operates.

The oil leak is driven by pressure, like any other fluid flow. When the oil pressure inside the center housing of the turbo is higher than the air pressure outside of the seal, the oil wants to travel outside to the air passage. The air that is holding the oil back on the compressor side is the compressor outlet pressure and for the turbine side, it is the turbine inlet pressure. That is because it is the air pressure that is behind the wheel (depends on wheel geometry and air pumping by the wheel too). This is why turbos leak more at idle/low load conditions and leak less when under full load/boost. The picture below is a depiction of the oil leak path.

http://i7.photobucket.com/albums/y26...sauce/spd2.jpg

Common failures of piston rings and sources of turbo oil leakage:
1. Seal collapsing- When the seal collapses from oil coking and heat and no longer provides a positive pressure on the sealing surface.
2. Seal wear- Even though the seals are made of very hard steel, they can wear thin axially and the leak path is now larger.
3. Contact surface wear- If the center housing bore or compressor backplate bore diameter is worn from seal contact and is no longer smooth or round, an oil path is formed.
4. Incorrect piston seating- It is possible to damage the piston ring, sealing surface or just not get a good seat when the turbocharger is assembled.
5. Out of tolerance parts- If the seal bore is oversized or if the piston ring is out of specifications it can create a large leak path.
6. Pressurized crank case- This can cause higher center housing oil pressures.
7. Oil drain kinked or too horizontal- As a rule, the oil drain should not be any more than 15º from vertical so there is no problem with oil drain that can back up the oil similar to #6.

Why oil seal leak is so evident in single, big powered cars:
1. Large turbos on small displacement engines means that turbo speeds are lower than properly matched turbos, therefore, the dynamic seals are not as effective. Furthermore, the air pressures at idle are likely lower on the turbine side with a large turbo.
2. Turbo oil inlet pressures and flow are not closely monitored nor properly restricted. (Should be around 20-30psiG for a ball bearing, a little higher for journal bearing)
3. Insufficient oil return line size- If oil is backing up and not properly draining from too small of a drain line, then the oil pressures become much higher in the center housing which drives more leakage.


I hope this can clear up questions for all of the large single turbo guys with oil seal leak problems.

Kevin

Thanks Kevin :icon_tup:

You might want to bring up carbon (compressor) seals, since stock FC's and (I think) stock FD twins use them instead of dynamic seals?


-Ted

Good post. :)

Funny timing you mention this. I recently tore down a supposed " fresh garrett rebuilt turbo" from ebay. Turbine seal land out of spec, and an out of spec chra, causing a turbine seal leak. Just another reason to stay away from ebay..

Definitely stay away from ebay for turbos. Anybody can take apart a turbo and put it back together but that doesn't mean the turbo is good again (as you pointed out).

I've never worked with carbon face compressor seals but instead of using a torturous path with a piston ring, they are made of carbon for low friction and create a positive seal (makes physical contact to seal). This can be used for applications when compressor pressures are really low (like if you have to mount your compressor stage post throttle body). Unfortunately these seals aren't the best for performance since they have more drag.

how do u fix a leaking seal on a turbo? do you have to replace the entire turbo or can you just replace the seals. im not very familiar on this stuff but my speed3 has a seal leak that mazda doesnt wanna cover under warranty so im trying to figure out what im gonna do

thanks

In order to fix a leaking seal on a turbo, the problem first needs to be identified. If can be as easy and unkinking a return line, but if it is the actual turbo, then the turbo needs to be rebuilt. Also, in order to say if you can just replace the seals depends on whether the seal is collapsed and stuck on the shaft wheel or if there is damage to the seal bore. If neither is the case, the seal can just be replaced. I can't comment much on why Mazda wouldn't fix your turbo if it is under warranty.

Mazda should cover this and it is a very common problem with the speed 3 and 6 there is something wrong with the design. My dealer has been replacing alot of them but i think they are just going to leak again so you are better off going with an aftermarket turbo kit if they will not cover it to prevent that problem and get some more hp which you will probably want anyway.

"The oil leak is driven by pressure, like any other fluid flow. When the oil pressure inside the center housing of the turbo is higher than the air pressure outside of the seal, the oil wants to travel outside to the air passage. The air that is holding the oil back on the compressor side is the compressor outlet pressure and for the turbine side, it is the turbine inlet pressure. That is because it is the air pressure that is behind the wheel (depends on wheel geometry and air pumping by the wheel too). This is why turbos leak more at idle/low load conditions and leak less when under full load/boost. The picture below is a depiction of the oil leak path."

I wanted to highlight on this statement above as to have a better understanding on how oil seals work, and also to identify on something I have noticed.

Could I be right in stating that you could create an oil leakage through something as simple as a compressor pipe popping off and continuing to run it in this state for a period of time. As the compressor side of the turbo would lose its capability of pressurizing the turbo from keeping the oil out of the surge path. And in fact would create an almost vacuum effect sucking oil into the compressor path As you have stated this would create higher pressures inside of the center housing than the compressor pressure.

The reason I bring this up is I notice when I pop off a compressor pipe and drive it for a period of time say I'm late for work or something. I notice more oil residue then normal, could this pressure difference be the problem.

If your compressor outlet pipe was removed and the turbo was run in this condition, it is possible to leak oil a bit more because you are not building as much boost. This problem would go away after you reinstalled the pipe though, and should leave any long term damage to the oil seal.

The main concern with driving like this is that you are no longer loading your compressor wheel, so you are free wheeling it and you can overspeed and damage the turbo this way.

I have sometimes, had to add a small pump in the turbo drain pipe, just to make sure there was no oiling probs. it actually puts a vacuum in the drain to suck any possible oil from going past the seals. HEY! it has worked for me on stubborn projects. gear pumps and gerotor seem to work the best.

Thx Ron

Good post !

Scavange pump? I was looking into it but they are pricy for a project I have, I might have to bite the bullet .

I have a bit of a dilemma.

Not my Rx-7 but another project I have.

I installed a single turbo t04e on a mazda Kl03 motor.

In a mission to understand how the oil bearings work, I ran 0 boost temporarily on this project car for a short period of time. The other reason is I haven't got my fuel mods in yet and wanted to test out the install briefly. I tried to do a couple of test runs with one of the compressor pipes off, and it would spray oil out of it but only under load. At idle it will not spray engine oil out the compressor pipe no matter how depressed the throttle is. And it will not smoke from the exhaust in either condition.

Okay, so then I hooked the compressor pipe back inline and removed the wastegate spring off the exhaust path to limit exhaust entering the turbine side which would also lower turbine pressure. The compressor pipe oil spraying was fixed but then the exhaust began to have a slight smoke from oil entering exhaust side.

So from these test what I am coming to understand is you cannot run the turbo without complete pressure from either side (turbine or compressor) otherwise the oil seal will be suseptible to leakage. From what I have read the oil seals require positive pressure from both sides to seal the oil in. Is this right or am I smoking too much green stuff.

I would avoid running without a compressor outlet pipe attached. This makes your turbo free-wheel which can overspeed your turbo and cause damage.

The compressor side oil seal needs to be able to generate boost in order to seal oil.

The turbine side sealing will also be affected by the wg since it lowers the turbine inlet pressure, but I have not heard of that being the cause of an oil seal leak.

Thanks

Hi CarbonR1, I'm the new user here.

Where did the pictures go? Stickies are important... Too bad.

Hi CarbonR1,I'm the new user here.I've found your post on the google.
I have been experiencing kind of smoking out issue for a years.Currently,I have 1995 Audi UrS6,the car smoking out when the car is started up and left it idling.
when I swtiched off and on again,on the idle,the car will begin smoke out again.
I have tried to send PM to you but I couldn't because the board's rules required atleast 3 posts here.

Well,Sometimes it smoked out a bit when It has been sitting on the traffic for a period.Currently ,I have 6 KKK K24s on my hand.I'm sure it's from turbo.because the car will smoke more or less on eachs.I have removed the crankcase ventilations valves,breather hoses for cleaned up last sunday,I have finished the valve stem seals 2 years ago,Motor swapping,also done compression test on it.
Some of my K24s ,I noticed some greasy came out from the hotside housing too,
So that's why It's most likely the smoke is from the the turbo.Maybe I have a bad luck on getting a good turbo. The problem,what is the cause to be shorten the turbo's life? I replaced drain line,old one also no pinched,clogged.Too much oil pressure? how is possible? since motor swapping ,I also used the oil pump
from the new motor.But the car is still smoke out from the boths motor.
Regards,
POP

good post
 

good post

I removed the pictures, but I will redraw something and post up new pictures sometime soon.

popdemonic-
I recommend that you start by checking shaft play on your turbos. If they have low mileage and have no shaft play, the seals are probably fine and it is likely something with your setup. Blocked oil drain, too light of oil weight, crank-case pressures too high, too high of oil inlet pressure, etc.

It can be difficult to differentiate a problem with your engine or your turbo. I know because I rebuilt my cylinder head once thinking that it was the valve stem seals when it was actually the turbo.

Inspect your turbos first, look for signs of oil leakage around the turbine since most oil from engine problems would likely be burned off already.

Thanks CarbonR1,I assumed all of my turbos had over 100,000 mile on clock.
I don't know if it's still working fine.I found some of my turbos has minimal
shaft play when I gap on the shaft. I just switched oil weight from 5W-50 to 0W-40.I don't know if this related.What's the oil weight you're running with?Normally,
It's around 87-92 F. Degree in Thailand. My oil drain shoudln't be the problem.
stock one came up with a very big Diameter.only thing I haven't checked yet is
the oil pressure.the Crankcase Breather System,I put all the hoses of and cleaned
up everything.There're 2 valves.one,Regulating valve.another,one way valve.It
will stay shut under boost and blow by when it's under vacuum.If this valve is blow by all the times.It can push much of oil to the intake tract.I replaced it!.
Valve stem seals have been finished last 2 years.When I cold started in the morning,There's no puff of smoke out.I need to wait until the engine s warmd up
then it'll be getting to smoke.So I think if the valve stem seals is bad.There's will
be puff of smoke since It has been started up.because the oil in the combusion chamber will be burnt out since that.What 's Idea? I've already taken photos of my turbo shaft and rings.I will shoot you an email soon.
Regards,
POP

hi a quick one a bought a new holset hx40 for my 24c 6clyinder vw engine conversion as im unsure if i should be useing a oil restrictor as if i rev the rev to approx 4000rpm i can bits of oil coming from the snail part of the turbo //boost hose outlet ,as exhaust is dry ,inlet is dry as well with very minimal play as im unsure what the correct size is ,due to not wanting to damage the turbo bearings ,,,

good info

Sure glad i read this. Makes sense of my intermittent smoke i am getting from my rear turbos.
Here i thought i had a bad oil control ring.

great reading :icon_tup:

thought i might post pics of another turbo killer i encountered firsthand, boost surge. caused by your BOV's not venting enough pressure when you let off the gas, all the pressurized air backs up and rapidly slows the impeller wheel while the turbine still has exhaust pressure pushing against it. this causes a twisting motion of the shaft and eventually, it will snap. mine was due to a vacuum line popping off of the BOV without me knowing

first picture was the primary of the stock twins
second is the turbine wheel and shaft banzai found in my midpipe
third is them flushing out my intercooler
fourth is the pieces of my impeller wheel that made it into the intercooler

on my FD, i replaced engine. and reused my known good turbos. it was smoking, and leaking oil bad enough that oil was dripping from turbos waste gate cover. So i though my turbos got bad from previous engine blowing apex seal. when i replaced the turbos, it was smoking a lot less, on a first test drive. second day i had it running, at idle, it started to smoke just like it was with the first set of turbos i had, and again dripping from waste gate cover. What would that mean? Could it be from vacuum lines being rerouted wrong way, passably?

I am still too damn busy @ work to address this ticking time bomb issue that is have. Just knowing and prolonging this will only end in one result:(.

I hope to i get to cure this before me time is up.

BTW I am seeing more and more oil in the cold side pipes:(

bad ju ju man, it doesn't take too long to swap out the turbos. make sure you have something to crimp the coolant lines though so you don't have to drain and refill your coolant system. I used needlenose vise grips

im 6 natalie gates hi :)

Edit: oops sorry, that was obviously my daughters first rx7club post, lol

glad someone posted this, im getting tired of seeing people toss turbos onto their cars with no oil line restrictors and wondering why there are no mosquitos for miles. and also why i hate that those same shops also sell oil pressure mod upgrades to those same customers and send them on their way billowing clouds of smoke under boost.

^ Very true.. my car had a Gt 40R dual b/b turbo when I bought it... NO Oil RESTRICTOR!!!! and a bloody kinked oil drain.. I immediately re did my drain line and put a restrictor in...but despite my efforts..The turbo went out yesterday in the canyon..Billowing smoke..and oil all over my exhuast mani. Can't complain though.. put 8k on it after It finally went out. Time for an upgrade!!!

sorry to hear, forcing oil past the oil seal damages the seal and shortens the turbo life so it was just a matter of time. :(

my turbo seal went bad but it looks maybe due to old age and hard usage.

how can you tell if your seals are bad? oil in the inducer side?

usely signs of eccsive oil in the intake , possible push off blue smoke on start up and a puff of blue when changing gear in boost, if its dry , be fine

good info....i saw smoke coming up where the turbo is located...now i want to see if its a line or a leak from the turbo itself..gots some work ahead of me..i bought the car a year ago...88trubo II, and took intercooler off and almost everything to see if maybe the car didnt trun on becuase of a wire that wasnt conected..found the problem.. and finally got it running.. changed the fuel pump and cut the wire that goes to the injectors and turned on my rx7 after 5 yrs of just setting in someones back yard..i cant wait to register the car and take it for a spin....=D

Does it cause your car turbos to smoke

hey guys.

ive got a issue here with my newly lapped, ported and dowelled 12a large extend port. The car is running a TO4R with 1.15 rear a/r and has approx 1500 miles from new.

The car was running fine a few weeks ago when i found a short in under the dash from old head unit wiring..

I sorted that out and withing 10min of firing it back up the car was chugging white oily/ fuely smoke out the tail pipe and had lots of oil in the rear muffler.. The plugs seem clean (pic below).

Attachment 704792

So today i decided to do some analysis and i removed the oil feed from the turbo and ran it back into a can and let the car run - it was running and wasnt running the best - but didnt seem overly clean.

Shaft play on the turbo seems minimal and i dont know what else to do..


http://s1017.photobucket.com/albums/...t=IMG_1285.mp4

Any advice guys??

If theres oil in the exhaust it will take a long time to burn out. You could try something like this to run it N/A for a while to see if it clears up.

https://www.rx7club.com/attachment.p...1&d=1313541507

thewird

hahahah ^^^

this is exactly going to be my next step.. I want to pull the wastegate off to check and see if there is oil on the valve - if not then i guess im clear on the oil control rings and turbo related..

keep you's posted.

If you plan on doing this, make sure you lock the compressor wheel so it won't spin without oil.

So i removed the wastegate and found this.. Unsure if this is now related to the tubro or not :(

Attachment 704793
Attachment 704794
Attachment 704795

pull the turbo off and see if there is any oil in the manifold or if the oil trail starts directly after the turbine wheel.

oil in the manifold leading to the turbo = bad oil control rings or poor port work into the oil seal tract.

oil just past the turbine wheel into the exhaust = bad turbo turbine carbon seal.

I have an oil issue on my single turbo FD, I have a 60-1 turbo /journal bearing/ with .84 hotside. what would be the proper restrictor size for this turbo or if I wanted to replace this turbo and go for ball bearing what would be the closest turbo to this one? I have to rebuild the turbo again due to bearing failure. It might have to do with crankcase pressure being to high but I have a atmospheric vent on the oil filler neck cap on my car so I don't think that is the issue. Any thoughts anyone?

thoughts = needs vacuum.. i keep typing it,, and those that ignore it keep on burning up turbos


what is so abhorant to all of you who keep on persisting with only a vented system or catch can ?



the factory puts a vac on it .. with stock blowby
/// you have more boost and more blowby .. so why do you think you now magically dont need a vac to pull it out ?


its fu%*& simple to add the vac line.. or to simply not delete the one you was given from factory

when you get bigger boosts,, its also not hard to tap a 5/16 or 3/8 fitting into the filler cap and use the larger diameter hose that common sense tells you you will need

From the neck the vacuum comes from the upper intake manifold and the primary turbo on twin turbo cars. the primary turbo is constantly providing vacuum. when you go to a big single turbo is a different story. that vacuum is not as constant as the primary stock turbo from the twins. since when you are on boost the pcv valve going to the upper intake manifold closes and all the vapors are getting sucked up by the turbo's intake side. Also, 95 RX7s don't the vacuum set up to the upper intake manifold since the primary turbo is in constant vacuum. I had set up a system like that before where I had both the uim vacuum and also a line going to the intake of the turbo. Since the turbo was pulling more boost that stock it was basically pulling oil into the turbo due to the fact that oil tends to creep up the oil filler neck on FDs. the vent I had installed had a check valve which vented when pressure was too much and then it returned to normal when the UIM was doing all the vacuum once pressure was down.

The correct way to vent a single setup is with a port on the oil filler neck and the twin turbo OEM oil drain on the rear of the engine to a catch-can with a breather filter on it. If you don't want to empty the oil, you can put a drain back into the oil pan via a port on the front cover.

thewird

@thewird-the set up that you just mentioned is the set up that I'm currently gathering the parts for. I'm actually purchasing RE-Speed oil neck with -10an fitting, remote oil breather tank, secondary turbo drain fitting, and lines.