Leak down test reveals HG leak, but no overheat or bubbles

Hey y'all....sorry for the long post then the question, but if you all know the entire history, then you guys can offer better help and suggestions

I purchased an engine from a builder last year for $2300. It was bare bones long block and had TB kit WP and all other related gaskets in new boxes for me to install. I had to use my oil pan, valve covers and plastic covers. My buddy and me had a endoscope camera and we inspected every crevice it fit into and everything looked legit. Pistons where sparkly new, cross hatching on cyl bores gleamed back at us and valves were clean. All seals were also new. I babied this car during the breaking in period and changed oil religiously.

Builder said that decking the block and heads along with Cometic MLS head gaskets raised the compression up a little so he advised 91-93 octane. I've worked on cars for over 35 years, built over 20 SBC aluminum headed 350s so I knew what he was talking about. For a while I did run 93 but switched to 87 octane when I had to once (gas station ran out of 93) and I didn't notice a difference in power. These cars ECU really do a good job adjusting timing and knock. So I ran 87 for like 6 months then the engine developed a P0303 code a couple of times and a shaky idle. I knew an exhaust valve got shot. I adjusted valves and that cyl 3 was indeed a little tight at .006 when it should been at .009. Changed plugs and went back to 93 octane and no more CEL..... but idle still shakes

did compression and a leak down test

Compression Test:

CYL 1 --- 175

CYL 2 --- 183

CYL 3 --- 90

CYL 4 --- 178

Leak down test on the engine was done a little warm, not up to temperature. The test confirmed my suspicions on the exhaust valve either burnt or stuck. I would have to drop manifold to see if valve guide collapsed.

The percentage loss between 1, and 4 are at 7% and 5% and CYL # 2 was oddly less than 4%

but of course cyl #3 was close to 50% loss.

SO HERE IS THE WEIRD QUESTION. How can I have a HG leak on #2 CYL with it being with the most cylinder compression?

Radiator started bubbling during #2 cyl test

It's been almost a month and the engine is still running normal.....not hot, no smoke, no odd sounds, no external leaks and NO BUBBLES in coolant reservoir.

Tried to contact builder but he moved out of state. I really cannot say if the gasoline octane grade caused the CYL #3 problems but not following the builders advise and my experience leads me to believe it did. I believe these engines are around 10.0 CR. I really doubt he managed to get 11 to 1 compression to require the 91-93 octane. I know someone might say I got what I paid for but I was happy for a while. I lost the build sheet to see what HG thickness was used. I am familiar that the Viton coating on these gaskets sometime fails.

What your thoughts guys?...should I wait till problems start happening or tear the engine down now? Should I use Subaru Turbo MLS gaskets? part # 11044AA642

Funds are tight, I have 2 kids starting college soon. But the costs should not be that much since everything on the engine is a little over a year old

2007 Gun metal grey Legacy Wagon Limited 2.5 NA

160K miles on body, 13K miles on engine. Syn Oil after 5th oil change

I love this BIT*$%

CYL 1 --- 175

CYL 2 --- 183

CYL 3 --- 90

CYL 4 --- 178

 

SO HERE IS THE WEIRD QUESTION. How can I have a HG leak on #1 CYL with it being with the most cylinder compression?

I am confused by the weird question. According to the compression test results, cylinder 1 has the second lowest compression.

When you did the test - at which position was the piston for that cylinder?

In theory it could be a cylinder wall crack a bit down if the piston was at the bottom of the cylinder.

But looking at the #3 figures it looks more like that cylinder would be the real problem and that the engine is in need of some serious work anyway.

Leak down test confirmed # 3 loosing compression through exhaust valve. The hissing through exhaust manifold was obvious. I even went through an engine rotation just to make sure there was no leaking past the rings. Leak down test was done BTDC. I had to wedge my breaker bar so the 100 psi air pressure wouldn't push cylinder back. When I pull the engine out I am hoping to just get the heads done, checking for block warpage and clean up mating surface.

I've been reading a lot regarding Subaru MLS HG's. I have been researching for a while now. My car is still acting normal, but i know its a ticking time bomb. The HG will fail eventually.

I send emails and questions to those who are running Subaru MLS and I haven't received anything negative. Many have said they were told by their mechanics and also after they tear down themselves they find that the HG coating was wearing off.

Just don't buy too many expensive parts until you have pulled apart the engine so you know what you need.

The devil is in the details.

Thanks ehsnils...you are right, definitely I'm not buying the HG's and parts before the engine is out. It's tough to imagine a cracked cylinder bore though. I baby this car, I haven't let her "open up" yet. I'm quite certain that when I purchased the rebuilt engine the guy just shaved the heads and did a once over on them. I don't think the heads had a complete job on them, so the heads might have a lot more of miles on them. The block looked internally and externally worked on. The heads were already on the engine and so I couldn't inspect them thoroughly.

Question is still beyond my knowledge of these cars:

How can the cylinder with the most compression, leak air into the coolant passages during the leak down test?

Why hasn't the engine start over heating like the classic HG problems the 2.5i have?

Could it be that the Cometic MLS Headgaskets do not fully seal until the engine is up to temperature?

I'm just gonna have all the parts saved up on online shopping lists. When that temperature needle goes past 9'oclock, I pull out the engine, remove the heads send them out for a full valve job and get my 500 watt flood light out and inspect the living crap out of the block.

I'm gonna take your advise Ehsnils to not buy anything yet

Leak down test was done BTDC. I had to wedge my breaker bar so the 100 psi air pressure wouldn't push cylinder back.

Doesn’t sound like you were at TDC if you had to wedge a breaker bar to keep from rotating. The trick that worked for me was to stick the open test hose in my ear as you rotate to TDC. You can hear and feel the air very easy. You get a short pause at TDC, before you start hearing a vacuum.

Thanks d22597, I will try your method. I guess I am so used to Chevy V8's and how I test piston rings. My reasoning behind putting the compression hose right before total TDC is so there is more air volume pumped into that cylinder. With more volume the air can hiss louder.

I am still bothered with air seepage into the cooling passage on the other cyl head. The engine acting normal (other than the cyl no.3 issue) and it's not heating up on highway driving, I fear that when that HG goes it will be sudden. And of course with the luck i have it will happen when my wife is driving the car. She wouldn't know whats wrong until the car over heats and dies.

I know Head gasket sealant doesn't work for internal compression leaks, there is no smoke coming out the tail pipe or anything so I'm at a loss.

I'm in CT (freakin expensive AF) so what would be a fair price for a machine shop to charge for new valve guides, valve lapping and quick resurface/polish of heads?

I can buy a pair of V25 AVLS heads fully re-done ready to go for around $850. So I don't know if I should go that route.