GTX3076 leaks air into CHRA

[SpeedRebirth]

I was diagnosing where my boost leak was coming from and found this weird problem...
Boost is somehow leaking into CHRA housing and getting into the oil feed and return lines! Is there some seal there that is shoot or something?

www.youtube.com/watch?v=XF5FwSfWrHc

My AFR's are in the low 10s at 6k rpm so there is a alot of air lost somewhere and this was not the place i expected to find a leak!

Here is my everything connected during leak test:


Tagging a few knowledgeable folks: @Lex; @Justin@Freektune; @rfinkle2; @ms3blackmica;

[Brian MP5T]

The CHRA has several seals weather it is is journal bearing or a ball bearing.

[rfinkle2]

ATP says that is common...

Take that fwiw.

[ms3blackmica]

The 30R's have an orangel gasket on the compressor cover.

The 35R's do not have any gasket at all.

Some air escaping Is completely normal however it should not be entering your CHRA.

The normaly escaped air exits from the outer edge of the compressor backing plate.

[rfinkle2]

I didn't read thoroughly and goofed. If it is getting in your return lines that is a nono.

[torquemaniac]

Not sure this is the best way to test for leak...the intake is never under pressure, and the pressure buildup in a centrifugal-type compressor is highest at the edge of the housing (the compressor wheel pushes air on tne side of the housing...the seals at the center should not see full pressure)

[Yatta]

Originally Posted by torquemaniac Not sure this is the best way to test for leak...the intake is never under pressure, and the pressure buildup in a centrifugal-type compressor is highest at the edge of the housing (the compressor wheel pushes air on tne side of the housing...the seals at the center should not see full pressure)

I believe also you would have oil present on one side of that seal (wet seal versus dry) I don’t know how much pressure you would/should see there but those seals are designed to keep oil in.

How much pressure are you placing on these for testing and how much is leaking?

[Lex]

The seal from the shaft into the CHRA is a piston ring type seal with a gap to allow for some thermal expansion. You do have a small amount of leakage here. If you are worried about the amount of leakage it is best to get in touch with Garrett's tech line and ask what is acceptable.

Here's a good article on this:

Turbo Torque: The Turbo Blues

[SpeedRebirth]

@Lex; @Yatta; @torquemaniac; @rfinkle2; @ms3blackmica;

I got a respons from cpe's "garrett engineer":

"Charge air seal leak rate is approximately 30cfh for Garrett GT25-35r CHRA, so the observation that air leaks from the compressor housing into the oil drain is accurate. On a typical GT35r turbo the total airflow rate is about 1000cfm, 30cfh is .5cfm. So the gas seal leak rate is far less than 1 percent of the total airflow rate, in this example it is .05% - enough for you to hear leaking into the oil drain but not enough to affect fuel mixture, AFR etc.

The gas seal leakrate is easily measured with the proper volume flow measuring equipment, I would be happy to measure the gas seal leak rate of this turbo for you. However, I would continue to examine other avenues as I do not expect the leak rate to be out of specification if the turbo has very few hours on it."

Kelly from cpe said that my test is not accurate because i don't measure how many % of the air that is actually leaking. That all turbos leak a little, but that they don't think this is my issue.

Maby the leak will be smaller when the turbo is at operating temp? And my rich afr values are from another leak
Advice? If you haven't watched the video there's a link in OP

[Jackis]

At operating temp and conditions there is an oil under 3-4 atmospheres, rotating in the CHRA like crazy, with huge centrifugal force, and it creates reall barrier for an air to leak into CHRA. Thats fine and I believe your leakage is somewhere else, cause I had exact the same kit in the past and my AFR were always fine.