Skin Tank Cooling

[davidk65]

For sometime a friend has had problems with loss of coolant. Apparently after a couple of hours running he has to stop and when the engine has cooled sufficiently top up with water. How much water is needed I am not sure. The mystery is where does the missing water go? The bilges and everything else are dry but the expansion tank for the engine is empty.

The calorifier was suspect. But if the heating coil in the calorifier was leaking inside the hot water cylinder I would have thought any hot water drawn off at the taps would in some way be tainted with engine coolant but this, I am told, is not the case.

As well the calorifier, all the usual suspects have been checked i.e. Head gasket. Water leaks from system. But all to no avail.

The latest thinking is could it be a skin tank problem? Of the two hoses connected to the skin tank, the top hose gets hot quickly. The bottom hose remains cool for quite a long time but eventually heats up.

I think of a skin tank like a central heating radiator. i.e. If the rad is warm at the top but cold at the bottom, it nearly always turns out that the radiator requires bleeding. Could this be the case with the skin tank? I understand that the tank does not have any means of bleeding air from it.

Does anyone have any other ideas on what it could be?

The engine fitted to the nb is a very old Canal Star diesel.

[nicknorman]

For sometime a friend has had problems with loss of coolant. Apparently after a couple of hours running he has to stop and when the engine has cooled sufficiently top up with water. How much water is needed I am not sure. The mystery is where does the missing water go? The bilges and everything else are dry but the expansion tank for the engine is empty.

The calorifier was suspect. But if the heating coil in the calorifier was leaking inside the hot water cylinder I would have thought any hot water drawn off at the taps would in some way be tainted with engine coolant but this, I am told, is not the case.

As well the calorifier, all the usual suspects have been checked i.e. Head gasket. Water leaks from system. But all to no avail.

The latest thinking is could it be a skin tank problem? Of the two hoses connected to the skin tank, the top hose gets hot quickly. The bottom hose remains cool for quite a long time but eventually heats up.

I think of a skin tank like a central heating radiator. i.e. If the rad is warm at the top but cold at the bottom, it nearly always turns out that the radiator requires bleeding. Could this be the case with the skin tank? I understand that the tank does not have any means of bleeding air from it.

Does anyone have any other ideas on what it could be?

The engine fitted to the nb is a very old Canal Star diesel.

 

It is normal to send the hot coolant to the top of the skin tank and return it from the bottom. This is so that the hot (and therefore lighter) coolant takes as much time as possible to get through the skin tank, so is able to dump the maximum amount of heat. A good skin tank will take a long time to start returning hot water at typical canal engine power, so if the engine is not overheating it would seem that the tank cooling circuit is working satisfactorily.

 

Without knowing more detail it is hard to know what the cause of loss of coolant it. If the loss has just started without any preceding maintenance activity, engine boiling etc, it does point to a leak that he has yet to find.

 

If he has for example recently replaced all the coolant, and the problems started after that, as you suggest it could be just a lot of air in the skin tank. Not easy to bleed a bad installation but typically one can sometimes do it by undoing the top connection to the tank - engine stopped and cold of course. A good installation will have a constant rise from the skin tank top connector to the engine so would pretty well self-bleed during (slow) filling.

[steamraiser2]

I would be suspicious of the engine itself. A weeping head gasket or a crack in the head would be my guess. It would explain the water loss without necessarily showing itself with clouds of steam. A water system pressure test would be a good idea. Canal Stars (Lister Alpha) are a great engine and are not prone to water traumas but easy enough to repair.

[NBMike]

I've just removed my oil cooler and found out where my water has been leaking to. The gearbox oil has about 50/50 oil and water. Its been leaking from the coolant side to the oil side of the oil cooler heat exchanger. Its an old Bowman which I am now going to do some experiments with to see if I can stop the water seepage.

Mike

Edited February 5, 2012 by NBMike

[RobinJ]

For sometime a friend has had problems with loss of coolant. Apparently after a couple of hours running he has to stop and when the engine has cooled sufficiently top up with water. How much water is needed I am not sure. The mystery is where does the missing water go? The bilges and everything else are dry but the expansion tank for the engine is empty.

The calorifier was suspect. But if the heating coil in the calorifier was leaking inside the hot water cylinder I would have thought any hot water drawn off at the taps would in some way be tainted with engine coolant but this, I am told, is not the case.

As well the calorifier, all the usual suspects have been checked i.e. Head gasket. Water leaks from system. But all to no avail.

The latest thinking is could it be a skin tank problem? Of the two hoses connected to the skin tank, the top hose gets hot quickly. The bottom hose remains cool for quite a long time but eventually heats up.

I think of a skin tank like a central heating radiator. i.e. If the rad is warm at the top but cold at the bottom, it nearly always turns out that the radiator requires bleeding. Could this be the case with the skin tank? I understand that the tank does not have any means of bleeding air from it.

Does anyone have any other ideas on what it could be?

The engine fitted to the nb is a very old Canal Star diesel.

It only needs a very small leak from a hose or joint to loose a considerable amount of water, if the engine is hot when it happens, it may only leave a small stain if your lucky.

Air trapped in the system will also expand and displace the water without loosing any, but this will depend on when it was last flushed or filled.

If the head hasket has blown, it creates an opportunity for water to leak into the oil, or the cylinders being expelled during starting as white vapour.

Similarly it can get into the gearbox oil through any leak in an oil cooler.

Also remember that an air gap is required below the filler cap on most systems to accomodate expansion, if you fill it right up, it is likely that excess will be expelled into the engine tray and the cap seal will deteriorate.

 

Is the system supposed to be pressurised?

Have you isolated the hot water system to proove it is not the calorifier heating coil?

[bizzard]

I've just removed my oil cooler and found out where my water has been leaking to. The gearbox oil has about 50/50 oil and water. Its been leaking from the coolant side to the oil side of the oil cooler heat exchanger. Its an old Bowman which I am now going to do some experiments with to see if I can stop the water seepage.

Mike

This is not uncommon with PRM boxes.

 

A simple way of checking for a head gasket or internal pressure leakage is to fill the header tank right up,start the engine and run it until normal running temperature is reached and then keep revving,blipping the the throttle sharply watching for any bubbles or a constant stream of fizzy bubbles at the header tank filler.As the water heats up it will overflow at first and if all is ok the water level should settle down at a more or less steady level with no rising and falling and no bubbles or streams of bubbles.

[NBMike]

This is not uncommon with PRM boxes.

 

A simple way of checking for a head gasket or internal pressure leakage is to fill the header tank right up,start the engine and run it until normal running temperature is reached and then keep revving,blipping the the throttle sharply watching for any bubbles or a constant stream of fizzy bubbles at the header tank filler.As the water heats up it will overflow at first and if all is ok the water level should settle down at a more or less steady level with no rising and falling and no bubbles or streams of bubbles.

Aaaah but mine's a Self Changing Gears box.

[NBMike]

Okay So I took the oil cooler home, stripped it down and tried to discover where the water was leaking from. I think it was due to putting some sealant on the mating faces when I last rebuilt it. This was just enough thickness to make sure the oiginal seals weren't compressed enough to seal against the core section. (Think dowty/bonded seal washers) Now that I have cleaned all the sealant off and bolted the flanges up evenly, there is no leak I capped off one end of the oil cooler and connected a mains pressure hose to the other end. Nothing leaked through to the water side over about a half hour period.

But the biggest own goal was that I connected the Johnson water pump up back to front. So all this time I have been running the cooling system backwards. This meant that the engine was always running cool until I fitted the 'Redshaw' thermostat housing. This meant circulation from the engine (hot) went straight through the gearbox oil cooler so making sure it got hot really well and really quickly and caused massive loss of power. So I have now made a Spaghetti junction around the pump to get the flows reversed. Similar to what I had to do with the gearbox a few years back. I tested it with a couple of clear hoses connecting the oil cooler water and oil pipes together and the water is now going the right way, the water flows through the by-pass as it should and the thermostat opens and allows hotter water through to the skin tanks. I'll get the oil cooler fitted again this week and I hope that now I have solved all my engine coolant problems.

Mike