Here is a head gasket solution I picked up from another board that I thought might be appropriate to post here...
___________________________________
From: Gene Goldenfeld <genegold@h...
Date: Thu Jan 13, 2005 8:06 pm
Subject: Re: [outback] Head Gasket failure info on Subaruvanagon mailing list
dyerlytle wrote:
There is an interesting discussion about head gasket failure in 2.5L Subaru engines over on the subaruvanagon mailing list. (This is a mailing list for people with VW Vanagons who have swapped in a Subaru engine to replace the VW water boxer.)
Indeed. In the interest of getting the discussion over here (and providing me a copy to archive), I've copied Al Wick's posts on head gasket failure from the Subaruvanagon group. I let him know.
Gene
-------------
I was in unusual situation where I was able to find the head warp "gasket fail" problem before the dealers were getting any failures. Many years ago I adapted a new 2.5 (10k miles) to my airplane. I have sensors galore on the plane, so when the head first started leaking, I was able to look at the data on my laptop and see that the head leaked pressure to coolant system 3 seconds after I hit full throttle. It then dissipated 5 seconds later. I really learned a lot and deliberately operated at full throttle for long periods (hour or two). All the time logging the pressures and temps every few milliseconds.
So here's the deal. All failures are caused by air in the cooling system. No air, no problem. If you have air bubble AND you operate at high throttle settings, the head will warp. It takes a long time to show up. So if there was trapped air 6 months ago, then you might now start to see discolored coolant, overheating. Or it might only occur climbing a hill in hot weather. Only the 2.5 has this marginal condition. All other Subaru's bullet proof.
My flying partner makes a living replacing gaskets on 2.5's these days. Number one repair item. It appears that the 2.5 has an area at center of block/head interface which doesn't have enough coolant flow. When a bubble passes by, it boils locally. This eventually causes head to warp, gasket to give out. Subaru has tried 3 different style gaskets, even adding coolant conditioner to improve heat transfer. Still a problem. But absolutely all failures caused by trapped air in system. All models of 2.5 liter the same.
The solution is very simple. Just drill and tap your coolant crossover pipe and add a fitting that allows air to leave engine and rise to your swirl pot. You will never have a problem. I operate my engine full throttle for hours at a time. Fabulous engine.
-al wick
--------------
I got a couple other private messages regarding the same. I'll try to
clarify...
The crossover tube is the aluminum coolant tube that lives under the intake manifold. It's rectangular in shape and the main coolant hose attaches to one end of it. It's the one everyone reverses. I'm unable to say there is a "best place" to add a fitting to the tube. Likely it does not matter where you place the fitting. Just somewhere in the top of this tube. The fitting needs to be on this cross over tube because this is the highest point in the ENGINE cooling system. Your goal is to purge any air that happens to be in the engine. This is different than purging air from your radiator or from your heater core. Air in the engine causes head warp. Air in other components just reduce their efficiency. We imagine that air flows with the moving coolant, but actually it only does to small degree.
Drill and tap this tube. Install a fitting, and run hose from that fitting to your swirl pot. No dips in hose, it must gradually rise to swirl pot. As long as the swirl pot is above the engine it will automatically purge all air from the engine. This results in a "robust" cooling design. Robust meaning it handles unusual conditions. If your brother in law borrows your vehicle, blows a hose, and doesn't realize he needs to bleed air, no problem. Because your system automatically purges all air from engine.