M1 5w30 / 2006 Legacy GT / 7165mi

[hooligan24]

is there a way you can pressurize the intake system to see if you have any intake leaks? on my Neon, I hook one side of a 2" rubber coupling(for plumbing) onto the turbo and a cap and valve on the other side. I then pressurize the system up to around 20psi with air and start to listen for leaks. with all the IC pipes and such in the intake system, there are alot of places to be loosing boost and causing particle ingestion while in vacuum.

Im not sure of how a subie ecu works, but in my car, even if there is a boost leak, the ecu will adjust the wastegate actuator to add boost until it see's the target boost. Who knows, if you have a leak you might just get better MPG's and a better UOA if you fix them.

hard to see. this is the ruber coupling attached to my turbo inlet. the other side has a PVC cap with a hole and a valve stem mounted into the hole. I clamp it down tight and then set the regulator on the compressure to 20psi. pressurize and listen for leaks, or spray all joints with a dish soap/water mix and watch for bubbles.

 

[Schwinn]

hooligan24 - I am familiar with this method of intake-leak testing. I don't have a setup built to do this, and it's not that easy to access the turbo outlet in this manner, but it hasbeen /can be done.

The stage2 setup uses a boost pressure of 18psi, and due to the small size of the turbo, it drifts down as the revs rise, because it just can't deliver the volume at this pressure. For this reason, if I get 18psi, I would imagine that I am ok. What's more, comparitive 1/4mile runs also show that my trap speed is comparable to others (around 100mph)... further showing that the car is not likely to be losing boost.

So, I agree the test might be a good thing to do, but I'm not so sure it's worth the effort at this time...? Again, shouldn't I see more Al in the mix (as others have mentioned) if it was really sand/dirt coming in?

 

[JoeFromPA]

Schwinn - This is pretty extreme....but any chance you are still running the stock TMIC? I know they tend to blow around 17-18 PSI, and I'm wondering if you could have a very minor leak which is inhaling air into the IC when the system is in vacuum vs. pressurized.

Joe

 

[Schwinn]

Joe - yes, it's the stock TMIC. The tabs are holding, but there is some oil showing up on the passenger-front corner of it... possibly from the blowby on the PCV circuit?

 

[edhackett]

Quote:
I did forget to note that this oil sample was stored in a glass pasta-sauce jar before it was sent it. I cleaned the jar, but maybe it still had "something" left on it that the oil picked up?


This could be a possible source. That pasta bottle is about 50% Si. SiO2 will dissolve into basic liquids, like motor oil with a remaining TBN.

When doing chemical analysis, it's always best not to collect your sample in a container made from one of the elements you are going to analyze.

Don't feel bad, I get PhDs who want SiO2 analysis done on quartz filters and carbon analysis on Teflon filters. Polyethylene is your best bet. You can find new, empty PE bottles at the grocery store.

Ed

 

[hooligan24]

I know when I did my first leak test ever on the car, it wouldnt test over 4psi before I heard a pretty big leak. Before the test, I was getting stock boost levels on the gauge even thought I had a few leaks in the system. My brothers 03 Evo was the same thing.....stock boost levels on the gauge, but still had leaks. If your ECU runs off a map sensor, your boost will be adjusted to meet what the ECU wants to see.....leak or not. After I fixed the leaks, I boosted to 19psi until the ECU saw what was happening and adjusted back down to the stock 15psi. Thats why we need upgraded ECU's, map clamps, piggy backs, standalones...ect to fool the ecu if you want to run more boost.

Its really worth the effort, especially since you are modded. might as well get the most from the mods you paid for. And, could also help out your UOA's

 

[Schwinn]

hooligan24: Ok, I hear you. I'll stop being lazy and build one up.

edhackett: Yes, I know better than to use the wrong container, but I did it anyway because I didn't think the glass would "come out" enough to affect the reading, nor did I expect the reading to be that sensitive to it. I won't do it again! (Bad engineer! *slap*)

 

[Schwinn]

For those interested, the next sample and post-thread are here: http://www.bobistheoilguy.com/forums/ubbthreads.php?ubb=showflat&Number=1347432#Post1347432