Boost leak in turbo? Any suggestions?

[boogyman]

Hi all~...

I drive a 2003 Tiburon 2.7, turboed with a Garrett GT30R turbo, with a 38mm TiAL wastegate.

The turbo was installed a little over a year ago, putting out 310 WHP at 0.75 BAR of boost (about 10~11 psi) on premium gas. The wastegate has a 0.6 bar spring, which I raise using a manual boost controller. At this setting, full boost was reached at around 3600 RPM.

A few months ago, boost started going up slower, and not reaching full boost until over 5000 RPM.

It's a big job to take everything out, because of the cramped engine bay, so I just drove the car easy for a few months, until last week, when I had to replace my clutch.

I decided to get to the bottom of the problem while I was replacing my clutch.

I found some exhaust leaks, which I thought was the root of the problem, fixed them, and assumed the car would be okay.

Well, it's not. Right now, at 4000 RPM, the car is only reaching around 0.3 BAR of boost, which is about 4~5 PSI, and to even get to 0.5 BAR, I have to go above 5000 RPM.

I thought that the boost controller may be the problem, taking it out, so that I could get 0.6 bar of boost from the wastegate, but this is not the issue.

I'm at a loss as to where to look for possible boost, vacuum, and/or exhaust leaks.

Can anyone give me some pointers on what may be the issue?

I'm thinking the turbine may be blown, but when I spin it by hand, it spins freely, with no noticeable shaft play, and there's no smoke coming out of my exhaust.

 

[MikeHalloran]

Have you measured the exhaust backpressure?



Mike Halloran
Pembroke Pines, FL, USA

 

[Ralf24]

The best way to check for boost leaks is to make some sort of air fitting adapter that you can use to couple the inlet of your turbo to an air compressor. Then pressurize the entire system with at least as much boost as you plan to run and listen for leaks.

Ralf

 

[boogyman]

Thanks for the tips.

Ralf, to measure boost leaks using an air compressor, I assume you mean to hook up the air compressor while the car is turned off, so that the sound of a leak can be heard?

 

[patprimmer]

You need to make sure all the inlet valves are closed when you do this. Not so easy if this is an OHC engine.

Check the blow of valve first. It is a likely source of inlet leak

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Ralf24]

yes engine off.

Pat, it's true you may get air leaking into a cylinder (you should remove the oil cap or somthing to make sure pressure dosen't build in the crankcase). I'm not sure I see the problem with this, maybe you can elaborate? usually it's quite easy to find all the leaks this way.

 

[patprimmer]

If an inlet valve and an exhaust valve are both opened on the same cylinder, like at TDC overlap, the boost will leak straight out the exhaust and the test will not hold pressure.

It needs to hold pressure to test if it only starts to leak before the desired boost pressure

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Mechaniac]

Here is a post I wrote some years ago decribing this method in detail. Most stock cams don't have overlap and it should work on a Hyundai. Soapy water works better than listening.

http://www.sccoa.com/forums/showthread.php?t=19149&highlight=leaks

Vernon

 

[franzh]

Mechaniac:
Sorry to disagree with your comment about stock cams having no overlap, but every OEM camshaft I have worked with in my career has some overlap.

Camshafts with little overlap tend to have better idle characteristics and lower end torque and is the reason many engines today have variable valve timing, to change the overlap at higher engine revs, and on a Hybrid vehicle, the overlap is increased during engine phase-in to assist in starting (Toyota in particular).

To further qualify, an engine with a single camshaft will have overlap while a dual overhead camshaft engine may have varying amounts of overlap.

Franz

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[patprimmer]

Most stock cams don't have overlap

Garbage.

Stock cams have only a few degrees of overlap, especially on 4 valve engines designed for fuel efficiency, but anything made in the last 80 years has some overlap.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[pipehead]

I thought the new Variable valve systems everyone was making have no valve overlap until the VVT kicks in and then they have a lot more than an old time big block making big HP.

 

[Ralf24]

I'm sorry, I still don't see the problem here.
This whole discussion about whether or not engines have very little overlap vs. some overlap is absolutely moot because as far as the OP is concerned, it should be extremely easy to find a crank position where both intake and exhaust valves are not open at the same time (ie. as long as no piston is within about 10-20degrees of TDC then you're fine) and even then, I really doubt the whole overlap thing is an issue to begin with. If a boost leak is indeed the root of the problem, it has to be big enough such that a GT30R struggles to make 5psi on a 2.7L engine! In which case, it must be a MASSIVE leak. I guarantee it will become quite obvious the instant the air is turned on regardless of whether or not any air is leaking out the exhaust!
Sorry for the rant, I just can't believe how overly complicated some people are making this simple problem.

Boggyman, good luck with finding your problem.

 

[boogyman]

Thanks for the suggestions, everyone.

Ralf24, if it is not a boost leak, what do you think the issue could be? Could you list a few ideas so that I can get an idea of where to look?

Is it possible that the turbo itself is starting to go bad? It's been about 12 months since I have been running it, so it's should still be going strong. As I said in my first post, there's no noticeable shaft play, the impeller spins fine by hand, and there is not a lot of smoke coming out of the exhaust.

When I disassembled the kit, I took out the wastegate, and pressed on the cylinder, and the spring seemed fine.

Any suggestions, anyone?

 

[MikeHalloran]

You don't seem to have eliminated the possibility of a clogged muffler or cat.





Mike Halloran
Pembroke Pines, FL, USA

 

[patprimmer]

Ralf

You are correct.

If you want a down and dirty really simple first check, no need for an adaptor, just wrap the air line with enough rag to jam it into the turbo air inlet with a reasonable seal and block the exhaust outlet with a piston at TDC overlap. The air will leak out somewhere. Also remove oil filler as there might be leaks around worn rings and worn valve guides.

However these problems are often not that simple, and with a large boost leak, often the engine wont even idle properly and blows black smoke out the exhaust as you try to accelerate. Something to do with a:f ratio going out the window as metered air escapes but the fuel for it is still injected. The OP did not mention these very obvious symptoms.

The one air leak that does not do this is an OEM blow off valve as the air is fed back downstream of the airflow or air mass measuring device. An air leak across the chamber and out the exhaust will make it difficult or impossible to measure the blow off valve effective opening point.

Having said all this, I think it most likely is the wastegate valve or spring that is damaged or worn or the muffler has a loose baffle that has moved and partially blocked the exhaust pipe.


Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[FoMoCoMoFo]

"it should be extremely easy to find a crank position where both intake and exhaust valves are not open at the same time" sorry but if it has more than one cylinder it will be extremely difficult to find a crank position with no valves open.

 

[boogyman]

Sorry to keep asking questions...

Regarding wastegate damage, how does one check if the wastegate valve or spring is damaged? For the spring, I did check it by pressing on the cylinder, and it seemed to be fine. Is there a more concrete way to do this?

Regarding exhaust, I have a 3 inch exhaust, with a huge resonator. The exhaust is only about a year old, which is one of the reasons I eliminated it as a potential problem. However, now that you mention it, my exhaust has become noticeably louder recently.

 

[patprimmer]

I just Looked up Tiburon.

It is a V6 24 valve quad cam with no V-tech.

It will most certainly have some overlap.

There will be plenty of stationary positions where the valves will be in an overlap position.

The turbo can be damaged but still spin freely and burn no oil. That damage might be in the form of burnt, bent or broken vanes on either side.

One problem with simple easy tests is they only find easy to find problems. One would hope that anyone posting here had already done the simple tests, and were here to seek advice beyond the obvious.



Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[patprimmer]

Very often the exhaust actually gets a lot quieter if a baffle moves to block a pipe or passage.

Waste gate test.

make sure there are no leaks upstream.

Make sure the blow of valve cant open and leak, probably by removing it and plugging the holes that leaves.

make sure one piston is at TDC overlap.

Block the exhaust outlet.

Blow air into the airduct at controlled pressure. Note the pressure when you start to hear airflow through the wastegate.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Mechaniac]

Quote:
"Stock cams have only a few degrees of overlap, especially on 4 valve engines designed for fuel efficiency, but anything made in the last 80 years has some overlap."

With a mechanical valve train maybe, but by the time you get this test set up and start checking most all hydraulic vavle trains will have bled off those "few" degrees of overlap. It's not "garbage" just real world experience. I guess I didn't specify as well as I should.

Vernon