Boost Leak Test Question - ?

[Traininvain]

At my wits end here trying to get my car to run right, hoping Zilvia can help me rule out one possible issue anyway.

- When boost leak testing, when I hook my tester up to the intake of the turbo, I get a steady stream of air coming out of the head/crankcase (ultimately it comes out the T fitting on the valve cover as that connection that goes back to pre-turbo/post-MAF on the intake track is off). Enough air such that it would blow a paper towel away or something, I don't have a way to measure it.

At first I thought this was simply air leaking past the rings into the crank case, but this is a fresh built motor. It has literally 1-1.5% leakdown and compression reads 180-182 on all 4. So then I tried moving my test point -

- If I instead hook my tester up to the hot side piping where it would exit the turbo outlet, the whole system holds air up to 25 psi like theres no tomorrow. Sure it will eventually lose pressure but its super super slow.

So I took the turbo off, put it on the bench, and capped off the outlet. When pressurizing in the inlet . . . that same steady stream of air comes out the oil fittings.

Is this normal? Should it leak any air at all? It sort of seems to me that it shouldn't leak any at all, as this would disrupt oil flow through the turbo? It's a 2871R.

[unicoder]

The turbo should hold boost on the compressor side when you pressurize it.

The seals that block off the center section might need servicing. Or you clocked the turbo and didn't tighten it down evenly and the air is escaping.
- Are you SURE its leaking, and it's not just spinning the shaft and you hear air moving from the hotside of the turbo?

By no means a trusted source, but you can read this if it helps:
SEALS

The main purpose of the seals at both the compressor and turbine ends of the center housing is to prevent the high-pressure gasses from entering the center housing and then to the crankcase! The fact that that the seals prevent oil from entering the housings is secondary.
https://www.ebay.com/gds/TURBOCHARGER-OIL-LEAKS-TROUBLESHOOTING-TIPS-/10000000178320282/g.html

[Traininvain]

Yeah so I just built a little better test setup -

Any air put into the compressor housing, when the outlet is blocked, immediately comes out the oil drain at . . . what seems to be the same rate I put it into the compressor housing. It holds no pressure whatsoever. I'm actually surprised it can flow that much, it acts like there is no seal there at all. The regulator on my compressor doesn't read well below 5 psi, but it does this the same at 5psi as 20psi.

It is only leaking from the oil drain (if I cover the oil drain it will try to force out the oil inlet, but that has a 3.5mm restirctor on it so it can'ta bleed off that fast). No air comes through to the exhaust housing at all or into the coolant lines

Fairly safe to say this may be why my car feels fine until about ~5psi and then completely breaks up? I'm assuming this means I'm pressurizing the crankcase almost inline with boost pressure, and its pushing that air back through the crankcase, out the T fitting, and then back down to the intake track . . . but seeing that much air come back post MAF has to disrupt the MAF readings before it I would assume.

This is 100% not normal - correct?

[Kingtal0n]

There must be a very bad seal between the compressor and oil supply.

It is normal for some compressor side liquid (including air as a liquid) to enter the oil supply, for example this is why Borg Warner Turbocharger engineers will say do NOT inject water/methanol pre-turbo (because it will mix with oil supply).

However there should be no detectable amount of flow (it should be so minimal that it isn't noticeable) where yours is clearly extremely BAD and you need a turbo rebuild/new turbo.
The minute you start to push pressure into the crankcase (from anywhere,really) the engine performance will dramatically suffer and oil will go pushing out all over everything.