
| Main Menu | ||||||
|---|---|---|---|---|---|---|
|

Home
Rebulding a marine engine
Rebulding a marine engine | Rebulding a marine engine |
|
Cooling without corrosion SOME engine manufacturers say that after a few years use, it's pointless to convert marine engines from sea to freshwater cooling - but of course, they're in the business of selling engines so have little interest in increasing an engine's life.. What flows though a raw-water cooled engine is warm seawater, which is extremely corrosive. Replacing this aggressive solution with a 50-50% mix of freshwater and anti-freeze will virtually stop the process. When we bought Alston Prince, our 26ft Colvic Sailer, back in 1995 the engine was 10 years old and just starting to show corrosion problems. Some blistering of the paint and a crust of white powder on the number one cylinder head joint, hinted that things weren't as they should be. Further investigation revealed that the aluminium cylinder head was being eaten away by the saltwater - despite the sacrificial anode having plenty of zinc left on it. Clearly, this engine wasn't going to last us another 10 years so something had to be done about it - and preferably without too much expense. Going it alone I became interested in a ready made kit conversion, complete with header tank and electric pump (price £328) which I would have bought, had it not been for the electric water pump, which I thought would compromise reliability. I didn't want a situation where an electrical problem - say, a blown fuse or worn out brushes on the pump motor - would render the engine useless. So I decided to go it alone and make my own sYstem. At fIrst it seemed an immense task, until I broke the system down into its basic components, and concentrated on each one in turn. Here's how I went about it: Heat exchanger I bought a second hand core unit - the bit that has all the copper tubes - for a fiver, while the outer tube came off a lorry oil heat exchanger which I found in a scrap yard. The core, which I think had never been used, fItted perfectly into the tube, and could be sealed with a couple of new 98mm '0' rings. Because the outer tube is cast aluminium, it was necessary that the salt water should flow up the centre, which of course is seawater resistant copper. But to make sure, I fitted a sacrificial anode (same type as the engines) in the bottom, which also doubles up as a drain hole. The seawater connection adapter plates were made out of 3mm thick 316 stainless steel sheet and the 10mm copper pipe work was joined to it with silver solder. Incidentally, silver solder is an excellent material to join copper and stainless. It has good flowing properties and a high resistance to salt water. Because I was unsure of the heat removal capacity of the unit, I mounted it vertically to allow convection to improve its efficiency. I arranged the saltwater outlet (hot) to exit from the top and the engine water (cold) to exit from the bottom. This really worked well. If you put your hand on the side of the casing, you can feel the temperature difference. Circulating pump This took a bit of thought. 1'd already decided that it would be belt driven off the front pulley, which already drives the alternator and Jabsco seawater pump. The problem was to fmd a commonly available car water pump, compact enough to fit below the engine on one of its mountings. Eventually, I opted for one off a Ford Fiesta; which was when I discovered that with all these pumps you only get half of it - the other half being part of the engine block! This I would now have to design and make (see above). It was important to ensure that the new pump would run fast enough to work with the engine just ticking over. To enable hand cranking, the front power take off shaft - to which the pulleys are attached - runs at only a third of the speed of the crankshaft. This meant that with the engine is at tick over (1200rpm) the drive pulley speed is only 400rpm. As I already needed to make a new pump pulley, I made this as small as I could - 77mm diameter. This gave me a pump speed of 670rpm at tickover - enough, I thought, to pump water around a closed circuit. Coolant plumbing To help circulation, I made the pipe work diameter as large as I could and its total length as short as possible. I used domestic 22mm diameter copper tube and bends for most of the connections, coupled to 25mm ID flexible rubber tube which I bought from a car accessory shop. I arranged the pipework carefully, so that it was self-bleeding. This means that if any air gets into the system, it would bubble out into the coolant reservoir tank. Starting from the cool side of the heat exchanger, water flows downhill into the Fiesta pump. This drives the coolant along underneath the engine sump and up where it splits into two, just before it enters the bottom of the cylinder blocks. An ideal copper fitting to split the flow is a 22mm 'T' with two 15mm connections. All four of the engine's water fittings had rust inside their bores - which I drilled out to 13mm to remove any potential bottleneck. The water recombines on top of the cylinder heads, where it flows through my specially designed thermostat housing and into a home built calorifier before returning to the top of the heat exchanger. There's a small bleed on top of the thermostat housing, which allows any air in the system to escape into the coolant reservoir tank. The system isn't pressurised because I don't believe the block core plugs are designed for it. If one did 'push out', the consequence would be disastrous, due to loss of engine coolant. A pressurised system is intended to raise the water's boiling point, which stops it from boiling when the engine is turned off. Another way to do this is to run the coolant about 15°C lower than normal - say 71°C, which is plenty hot enough in a confined engine space. Seawater plumbing There are few re-routing mods to make here. The seawater tube that originally fitted to the engine block was moved to the bottom of the heat exchanger. The flow then continues up and out of the top fitting, and into the bottom of the exhaust manifold, as the existing plumbing did. Unfortunately, the exhaust manifold isn't included in the fresh water circuit, because the heat exchanger isn't big enough to cool this as wellnot a problem, because, the cast iron casting has thick wall sections, which won't rust through in a hurry. To improve the reliability of the Jabsco sea-water pump, I've removed the mild steel bends tm top of it - the source of many overheat" ing problems - and replaced them with copper. Thermostat housing This was straightforward to make and fit. It contains a commonly available thermostat set for 71°C, and also a temperature sender, again easily obtained. To ensure the system can both bleed itself of air, and maintain a small rate of circulation when the thermostat is closed, an outlet fitting was fitted on the engine side of the thermostat. This enables a small amount of coolant/air to be circulated back to the reservoir tank and also allows hot coolant to reach the thermostat to open it at the right operating temperature. To control the bleed flow rate, an M6 stainless steel grub screw, which has a 2.5mm diameter hole drilled through it, was fitted into the 90° bend on top of the thermostat fitting. Coolant Reservoir Tank Th |
| < Prev | Next > |
|---|

