i doubt the leaks on exhaust or turbo as its under pressure so will cause bubbles to flow up through the rad, doesnt allow for water to flow back up and over the exhaust header and into the cylinder. i doubt its the head gasket as you're getting 200psi compression. the leak has to be somewhere on the intake side for cylinder no 2. you said this head was ported? its possible it was done over enthusiasticly and a chunk of water jacket on the intake port has broken off. a crack isnt going to allow much flow of coolant... unless its split right open...
My suggestion was in relation to your question "How can I definitively and accurately diagnose this problem?" As you are now in the 'acceptance phase', start the repairs by removing the induction manifolds.
Sure If the crack/leak is in a water jacket right next to the exhaust port on cylinder two, it would dump coolant directly into the exhaust manifold. Couldn't be on the inlet side of things or he'd have a hydro-locked motor. My thinking is it's quite close to the exhaust valves hence a small amount is backing up and entering the combustion chamber. Also a compression test might be the only time water is getting into the combustion chamber, with no exhaust gas pushing the coolant into the exhaust manifold. Could be well wrong but in my mind this seems to make sense.
brisz, when do you think you'll tear it down? I'm genuinely interested (from a technical perspective).
And the winner is .... me ! FAARRRRRRK Dropped the exhaust and it was full of water as expected. Pulled the engine out and removed the head, gasket looked fine, after many minutes looking and head scratching, a very vague water mark on the floor of the centre exhaust port, on very close inspection a pin hole, maybe smaller. A gentle probe with a small screwdriver and it opened up easily. Was this corrosion or "over zealous porting" ?, well inspecting other components no heavy corrosion is apparent. No evidence of water in number 2 despite high compression test, even though the pin hole was in number 4, the water must of pooled in the exhaust manifold droop between 2 and 4 and some how a very small amount of water was taken up into number 2. Here is a picture.
That's not good news John, but I guess it didn't come as a surprise. The prize for correct diagnosis goes to Rollin & Egg by the looks of it, well done guys!
Bugger. Well, at least now you know the deck surface wasn't damaged by a blown head gasket. Bores haven't rusted I hope? I assume you'll just reuse the bottom end and just slap on another head? Or will you get another head ported to match? I don't know much about these heads specifically, but it might be just bad luck. Usually cast in water jackets on cylinder heads have a fairly rough dimensional tolerance. The sand core may not have even been seated correctly when it was poured.
The engine is only 35k since detailed rebuild, yeah bottom is perfect, that was the thing, it started and idled no worries, only minute water in number 2 and I think it was probably only when plugs out doing comp test, plug and cylinder show zero signs of water. Interesting thing is it used very little water till you stopped hot, exhaust pressure must of minimised the loss and when stopped with some pressure in cooling system it flowed down the exhaust where it was being vaporised and coming out the pipe as steam when running. I was going to get a place to look at doing a weld repair and ultrasonic testing but another smaller problem on a cam journal appeared so its for the recycling bin. I have no doubt its what I called it for.
