I would concur, the water is probably being blown out of the header tank overflow, to check put the overflow pipe into a recepticle.

That's when things have got really bad! In the early stages, the beginnings of head gasket failure will just suck an undetectable amount of coolant into the cylinders and out through the exhaust, requiring only an occasional top-up. One can delude oneself that's acceptable.

Although known for gasket failure, I think part of the reason is that the bolts which hold this odd engine design together are far too long. After years of vibration, expansion and contraction, I think they stretch, failing to squeeze the head onto the block as tightly as they should.

If I had one of these dreadful old engines which started to lose coolant, I would undo the bolts by 90deg, then by another 90deg, then retorque them. It would be an interesting experiment to mark them and see by how many more degrees they would retighten from their original position. If you try this, please let us know the result!

Good question. In an unpressurised state I doubt if half a turn loosening would allow the rest of the seal to be broken. Even if it was, the weight of the head should maintain continence. IIRC it's difficult enough to break the seal when all the bolts are out and you are bashing it with a wooden mallet trying to get the damn thing off! If you disagree, try this experiment - fill a glass half full (to preclude vacuum retention), place a thin card over it and invert onto a smooth worktop. None of the water will escape.

Though noted for h/gasket failure a simple dye test will confirm/refute

Ebay sell them for less than a tenner,

Look for block tester.

Idiot proof test.

Dye tests and gas sniffer tests aren't 100% absolute. The coolant is sometimes sucked into the engine.

I had a Hyundai Santa Fe 2.2 CRDi which was slowly losing coolant - despite umpteen checks the dealer couldn't pin-point the problem - after a terminal boil-over event towing a caravan down the motorway, the dealer got Hyundai to authorise a partial engine strip-down but even then it took an eagle-eye to spot the minute evidence of leakage across the cylinder head gasket.

After the rebuild, the engine felt more powerful and definitely more economical.

I only kept on at the dealer as I'd had a similar issue on a previous Vauxhall so I knew it was possible, however unlikely.

edit, old thread dragged up again, note the OP never did reply, quelle surprise.

I wonder if a moderator could send OPs an email asking for a response to our collective suggestions when the outcome might prove interesting? Could it be automated after, say, five reponses along the lines of....'please let us know the outcome of your problem?'