I'm looking into what engine to put in my boat. Marinised engines generally have several parts water-cooled that are usually air-cooled on an auto engine. The most extreme of these is the exhaust manifold - in a car, there is constant air flowing over this to cool it, but if you just put the engine in a boat with no cooling air, it can easily glow red hot. So it's usual to use a manifold with a water jacket around it. This is added to the engine's existing water cooling loop.
Some engines have water jacket manifolds available for them but most car engines don't. So it's not that unusual for people to make their own. This is usually made by building a box out of sheet steel or aluminium, cutting holes in it where the manifold pipes will enter and exit it and then welding it around the manifold. I'd like people's thoughts on something a bit different.
A water jacket could be made by packing a layer of something water-soluble around the manifold and then laminating glass roving over this with a high-temperature epoxy. An epoxy such as this might be suitable: https://www.permabond.co.uk/1-part-epoxies-ch9o. It could be cured just by running the engine under enough load to bring it up to temperature. The mold material could then be gradually flushed out be squirting water into it, and eventually running water through it.
How durable might this be? My particular worry would be the join between the steel manifold and the epoxy. It would be hard to predict how hot this would get; not as hot as an air-cooled exhaust manifold, but potentially still well over 100 degrees in places a little distance from the water. It's worth remembering that the water is not kept as cool as possible; the thermostat only switches the heat exchanger in at a certain temperature to improve engine efficiency. The epoxy is marketed for applications like bonding heatsinks, so presumably it can maintain a bond through repeated temperature cycling, but how long exactly is it likely to last?
This could perhaps be improved somewhat by replacing the studs that bolt the manifold to the engine block with longer ones and cutting another piece of sheet to cover the edges of the outlet ports; the ends of the glass roving could then be sandwiched between this plate and the manifold ports. This would reduce the dependence on the strength of the epoxy somewhat by backing it up with a mechanical clamp on the roving.
I'd like some informed feedback on this. As far as I can tell, no-one has tried this before. Is that because it's a stupid idea?
Bear in mind that a boat's engine is a fairly safety-critical piece of kit; you don't want to be stuck in the middle of the ocean without a working one. And a sudden leak in the cooling system is a serious failure for an engine.