While pressure testing my complete intake system tonight, I found a significant leak in a strange place. I am pressurizing from the compressor housing inlet at 10-12 psi. I did find quite a few leaks elsewhere and repaired them. Afterwards, I still heard a hissing coming from the turbo area. At first I thought it was the air going through the turbo, but after spraying a soapy solution it was easy to see the leak. The compressor housing itself is leaking quite badly where the compressor housing bolts to the center section with the 13 mm bolts. All bolts are tight, but it starts leaking around 10 psi. Is this normal or a big problem? I have a feeling there is a seal shot, but not sure. I had this apart a long time ago to measure the wheel, but I can't remember much about it. The red arrow in the picture shows where the leak is. This is the line where the compressor housing meets up with the center section. It leaks all the way around from what I can tell.

 

By the way, if you have not performed this test and think/assume you don't have leaks, think again! I would have bet a lot I had no leaks with nearly 18" of vacuum at warm idle but I was wrong. This is such a simple test if you have a MAF setup, can't believe I have not done this sooner. I picked up the parts at HD for less then $10.

 

After an hour of web research I found that certain T3 housings came with an O-ring seal and others did not. Without an O-ring seal I think this will be normal. Now to find out if this model had one or not...

 

I promise to use bigger pictures next time! Damn I F'ed these up, they are ridiculously huge for some reason. I only used medium from the Iphone, normally not this big.

 

I've heard equal amounts of stories from around the web - one side says it's OK to pressurize the compressor housing, and the other side says it's NOT ok to pressurize the compressor housing.

When I did my pressure tests I opted NOT to pressurize the housing. Figure under normal operation the ENTIRE housing is NOT pressurized - just the exit side of the compressor wheel is. I pressurized from the turbo end of the first hardpipe... and still found a ton of leaks.

 

Jim:

Yes, I found varying opinions as well, but silicone intakes which is the leading supplier of the "store bought" pressure testers recommends connecting to compressor inlet. In Van's video he uses the MAF side of the factory J-boot which also pressurizes the compressor housing. I am sure the entire "snail" is seeing the elevated pressure as it is open. I am sure it varies from point to point and lower closer to the inlet, but the area in question with the leak is definitely pressurized. I would feel better if it were not leaking here. It is not a minor leak and is leaking on the entire circumference.

 

in the 911TT world, it's always pressure test at compressor inlet...air leaks are air leaks....
always best to pressurize to a few pounds over 'max' boost to reveal any leaks that might occur during a boost spike

 

Hmm..I've always hooked up to the compressor outlet, not inlet. Mainly because I build my pressure tester for a 2.5" pipe, not 3". I do know my T04 has a gasket on the compressor housing.

 

I am going to call Turbonetics who made my turbo and see what they say. I too don't think it should leak, but if designed without an O-ring I see what it leaks. I found some pictures I took when I disassembled the turbo and definitely no O-ring present.

 

OK, just got off the phone with Turbonetics, short answer design flaw!!! Original design does not use an O-ring for a seal. Redesigned down the road after many complaints. Recommended solution:

Remove compressor housing and apply very thin coat of Honda or Yamabond to both surfaces, reclock and let sit for 24 hours. I hope I can achieve this in the car. Will increase boost response and much more stable at set boost pressure. We will see.

 

First look at removing compressor housing while still mounted to pedestal doesn't look good. There is one 13mm bolt on the bottom that looks impossible to get to. Has anybody attempted this before? Looks like a lot of work for what might be very little gain. It has been like this for years and I have never known about it. Might be a big headache over nothing.

 

Had exactly the same leak on my LR Super 75 T4, no seal on the flange.
Gobbed some sealant in there and all good.
Maybe not helped by one retaining bolt removed, by design, for clearance issues.
I somehow managed to remove the inlet housing whilst still bolted to the car.
Good luck !

 

Well, that is good news! Maybe mine has that bolt removed as well? I could not see or feel the one at the bottom near the pedestal. I either need to look from the bottom or get a mirror.

 

COuld you remove stuff around it, and loosen it enough to clock it so you can get to it?

BTW......I wish I had done this a long time ago. When I put everything back together, I swapped my idle and CCV hoses. Blew the rear main seal out on the track. Then in August, one of hte nipples on the intake was loose enough that under boost, it pushed out, but off boost, the hose pushed it back in place. Only found it after taking off the intake to put in my rebuilt wiring harness. I have the stuff to do it, did it when I was selling some big intercoolers I bought from an old Trans Am team.

Doing this test could have saved me a lot of lost track time.

And I cannot for the life of me think why a turbo couldn't handle a measly 15PSI of pressure on the inlet side.

 

Harry, thanks for the input, your opinion is always highly valued and welcome. I guess I am in for a long night then. I am pissed at myself for not doing this sooner! I have zero miles on the engine since pulling the intake a few weeks ago to replace, ISV, every harness, TPS, rebuild throttle body, and verify all vacuum lines for leaks. I was trying to fix an idle issue with little to no help. Now I get to do it all over again.