I have often seen the thin green o-rings around the tops of the wbx cylinder sleeves blamed for fire ring leaks, so maybe a little reality check on that question is in order. In the ten years of my wbx rebuilding business, we tore down close to 200 wbx’s, and a lot, probably almost half, had at least one of those o-rings displaced or torn. They didn’t get damaged on head removal, it’s quite obvious it happened on installation, and for obvious reasons. When the heads are put on with standard pushrod tubes it’s impossible to slip them on straight, they have to go on at an angle and as such when the edges of the cylinder openings slide over the o-rings they easily catch the rubber and lift a little loop of it out of the groove. This loop then gets clamped in between the head and sleeve top and there it stays. Yet those same engines consistently show no evidence of fire ring leakage, no carbon staining of the sealing surfaces in the head or on the cylinder top. 

.

The cores I was getting were either good running engines removed for Subaru swaps (I got lots of cores from my buddy Mike Labate at RMW where they do lots of those jobs, so he mostly knew the owner history on the wbx and noted if it was a reliable runner or not), or they were cores returned by my own customers in exchange for their rebuilds. Nearly all of my customers were just looking for more power, only a handful were after a new engine because their old wbx has become unserviceable, and when that was the case they usually told me their story.  Plus the parts themselves always tell their own story on teardown.

.

Frankly, just looking at the design, I’ve never believed that the green o-rings actually did anything of value at all, and my observations confirm that to my own satisfaction. The engineering rationale was plainly to prevent crevice corrosion. You do see that often, as pitting on the vertical walls within the head’s cylinder recess, the surface against which the o-ring is supposed to seal. The irony is that, given the high incidence of damaged o-rings, and how many sleeves are nearly fused into the heads by iron rust above the o-ring height, their decision actually increased the incidence of crevice corrosion because when the o-ring seal is imperfect coolant is forced into the space the o-ring is meant to seal off, but is then trapped there by the same partial seal and as its anticorrosive additives are depleted, it attacks the metals. This is the same process that starts the corrosive process that undermines the water jacket seals wherever beads of excess sealant lie along the inner edge of the seals, but is even worse because there is cast iron and aluminum in close proximity, connected by a film of corrosive liquid, with an electrical circuit completed by the intended contact surfaces of the sleeve and head via the fire ring, a perfect set-up for rapid electrolytic corrosion, with the aluminum being the sacrificial metal. So there is both oxidation of the cast iron, at least until the dissolved oxygen is depleted, and electrolytic corrosion happening simultaneously in the same space. No wonder the sleeves get so glued into the heads. However, if corrosion above the head o-rings was to cause a fire ring leak, you would have to see the aluminum pitting advance to where it begins to undermine the fire ring, the same way it undermines the water jacket seals, or rust take away enough iron from the sleeve top to undermine the steel fire ring there, probably accompanied by rusting of the steel fire ring itself. And I have never ever seen the slightest evidence of any of that. I suppose given enough time either process could progress to that point, but not within the usual working life of the engine, which is pretty long. And given how many cylinders are nearly fused to their heads, with the green o-ring seen to be undamaged after pounding them apart, I’m not even convinced that they exclude coolant even when they are intact. 

.

Look at it another way: balance of pressures. When the o-ring fails to hold back the pressurised coolant, what effect can a viscous liquid at some 200-250ºF at only 1 bar pressure have against an interface that is holding back violently pulsing 900ºF gases with pressure peaks at over 1200psi numerous times per second? Little to nothing, I would say.

.

I’ve also seen it asserted with firm certainty that even if the o-ring seal was imperfect, there’s no way the liquid coolant could be forced in without the air having a way to escape, so the two fluids would simply oppose each other in equilibrium and no coolant could enter the space. Anyone believing this needs a much firmer grounding in the most basic aspects of fluid physics, because those fluids have a number of mechanisms by which they would easily trade places.

.

So, I actually concluded long ago that the green o-rings could be omitted without ill effect. I guess I just kept putting them in because there they are on my workbench, it takes under a minute to oil them and slip them in place, and having them there is no worse than not, so why not?

.

Unlike car aficionados, I’m not someone who has ever exalted engineers. I’m not a car person at all, actually, I’ve just made my living working on them for a large part of my life. I have also worked directly with engineers and know they can be just as flawed in their reasoning as anyone else. I know from experience, as any working wrench learns,  that companies’ design and engineering departments make plenty of glaring mistakes, and that any car has plenty of redundant and unnecessary components or features. Aficionados like to argue that every feature seen in an engineered device had to be put there for a reason, and I don’t disagree with that notion per se, but with experience you learn that the reason was not always a good one. That’s what I think accounts for the green o-ring; they had a theory as to why it should be there,  so they put it there. But their theory was pretty weak.