Finding mystery leak in central heating system.

[frlrubett]

First winter aboard this boat and initially the central heating (three rads, towel rail and calarofier) were fed by a gravity system into a stove back boiler with thermostat. 

 

As this was noisy, inefficient we decided to remove the back boiler and fit a diesel heater.  We had coolant loss from the system quite a bit but I attributed that to getting the stove too hot to quick, happened quite regularly, could hear the water boiling in the pipes.

 

Now the stove has gone, the diesel heater installed and I have just checked the header tank again.  It was empty.  Have refilled with four jugs of pre mixed coolant, I guess about six liters.  Have spent the whole day looking for a leak but everything looks dry. There is a hole at the back of the boat, under the floor where you can see/feel the bottom of the bilge and that is also dry.

 

- this boat was built in 2011, I would assume any leak would arrive at the rear of the boat at the lowest point on the bilge?  We list to the rear, especially with an empty water tank so I any leaks would collect back here.

 

- Would a leak of this size be obvious?  I have checked around all the radiators and joins in the pipes which I can see. Around the heating pumps and calorifyer. 

 

- how else can I go about finding this leak, I am sure that losing six litres is definitely a leak?

 

Thanks

[MtB]

4 minutes ago, frlrubett said:

- how else can I go about finding this leak, I am sure that losing six litres is definitely a leak?

 

Thanks

 

Well the obvious way is to visually inspect every inch of pipe on the heating system. We know it is escaping from somewhere! 

 

But you write about filling it with 'coolant'. Is the heating system combined with the engine cooling system then? If yes then I'd be suspecting an internal leak in the engine e.g. hesd gasket or over-pressuring and discharge past the sprung filler cap.

 

 

Can we see a photo of this "header tank" of which you write please?

[Feeby100]

If you was running it with back boiler before and it was boiling the water that where it gone no leak 

[Tony Brooks]

Could it be  a weap/leak in the boiler heat exchanger that just vaporises?

[nicknorman]

Antifreeze doesn’t easily evaporate so maybe it isn’t actually leaking. Maybe the system got a lot of air in it when you changed to the diesel heater? Having run the system with an empty header tank, chances are there is air in the radiators even now. If you haven’t already done so, make sure the radiators are properly bled, then keep an eye on the header tank level.

[frlrubett]

8 hours ago, MtB said:

 

Well the obvious way is to visually inspect every inch of pipe on the heating system. We know it is escaping from somewhere! 

 

But you write about filling it with 'coolant'. Is the heating system combined with the engine cooling system then? If yes then I'd be suspecting an internal leak in the engine e.g. hesd gasket or over-pressuring and discharge past the sprung filler cap.

 

 

Can we see a photo of this "header tank" of which you write please?

 

So, the engine has a skin tank cooling system which is separate to the heating system.  BUT the engine is connected to the Calorifyer so when you run the engine you get hot water.  The heating loop with Diesel heater is also connected to the Calorifyer but I think it is a separate system.  I have never noticed radiators warming up whilst the engine is running, just hot water.  

I know that is known as a gravity fed system, so there is an open header tank (completely separate to the engine header tank) above the heigh of the radiators giving some pressure to the system. 

 

Thinking back to the summer, the coolant level in remained the same so I am right in thinking this leaks when its heated.  Would this be escaping as steam or will there be some water pooling somewhere?  

 

 

2 minutes ago, nicknorman said:

Antifreeze doesn’t easily evaporate so maybe it isn’t actually leaking. Maybe the system got a lot of air in it when you changed to the diesel heater? Having run the system with an empty header tank, chances are there is air in the radiators even now. If you haven’t already done so, make sure the radiators are properly bled, then keep an eye on the header tank level.

 

Good point, I will bleed the whole system tonight and then monitor from there onwards.  I have bled this a lot before, would that indicate water being pushed out during heating cycle and then air being sucked back in during cooling?  Thanks

[nicknorman]

You have adapted a gravity fed system to a pumped system, so the whole  thing was not designed as a pumped system at the outset. Not my area of expertise but I think if you have insufficient height difference between the return pipe and the header tank, air can get sucked in during operation.

 

Another though, does the header tank have some sort of overflow which could by some means allow fluid to escape unnoticed?

[frlrubett]

4 minutes ago, nicknorman said:

You have adapted a gravity fed system to a pumped system, so the whole  thing was not designed as a pumped system at the outset. Not my area of expertise but I think if you have insufficient height difference between the return pipe and the header tank, air can get sucked in during operation.

 

Another though, does the header tank have some sort of overflow which could by some means allow fluid to escape unnoticed?

 

Oh interesting.  I did wonder about the this.  The original gravity system did have a thermostat and a circulating pump on it, so just assumed that a Diesel pumped system would bolt on.  Yes there is an overflow, but that is at the top of the header tank.  Once the level drops below that then I do not know how any liquid will dissipate. 

 

 

[Tony Brooks]

5 minutes ago, frlrubett said:

 

Oh interesting.  I did wonder about the this.  The original gravity system did have a thermostat and a circulating pump on it, so just assumed that a Diesel pumped system would bolt on.  Yes there is an overflow, but that is at the top of the header tank.  Once the level drops below that then I do not know how any liquid will dissipate. 

 

 

 

The liquid will expand when heated and the volume f this expansion going into the header tank depends on the temperature rise and volume in the system. So with a tank filled to just below the overflow when cold will expel water as it heats up and the level will drop as it cools making it look as if there is a leak, but it will fill up again when the system is next heater. However the volume needed to top up is far in excess of what one would expect due to simple expansion.

[nicknorman]

9 minutes ago, frlrubett said:

 

Oh interesting.  I did wonder about the this.  The original gravity system did have a thermostat and a circulating pump on it, so just assumed that a Diesel pumped system would bolt on.  Yes there is an overflow, but that is at the top of the header tank.  Once the level drops below that then I do not know how any liquid will dissipate. 

 

On the tank overflow, I was wondering if it might be possible for the tank level to rise when the heater is in operation, push water out because air’s been sucked in, then when it’s not running the coolant settles back. It would be worth watching the tank level as the system heats up. Oh and just checking, is the tank in the heater flow or return?

[PeterF]

Building on Tony's and Nick's replies, if there is air in the radiators, then when things heat up the expansion will be more than with water, so the chances of overflowing are increased. Viscous circle, air in system causes more expansion and overflow, when it cools down, more air is sucked in, next heating cycle causes even more expansion, overflowing even more. Each cycle things get worse.

[Jen-in-Wellies]

1 hour ago, frlrubett said:

The original gravity system did have a thermostat and a circulating pump on it,

In which case it wasn't a gravity system.

[frlrubett]

3 hours ago, Jen-in-Wellies said:

In which case it wasn't a gravity system.

 

I think gravity system with added pump is best way to describe this system.  As gravity still plays its part in keeping pressure in the system. 

[Tony Brooks]

40 minutes ago, frlrubett said:

 

I think gravity system with added pump is best way to describe this system.  As gravity still plays its part in keeping pressure in the system. 

 

If you had a open header tank that is normally referred to as not being pressurised although one must accept the bare 6ft or less head must put some pressure at the bottom of the system. Even a circulating pump can't really pressurise such a system because the water expelled by the pump has a very easy path back into the pump. The pnly pressure the ump can build up is that caused by friction in the pipes etc and that won't be much.

 

I don't think it s very helpful to consider open vented systems as pressurised, leave that to closed systems that really are pressurised.

[frlrubett]

SO..... when the system heats up the water level raises in the expansion tank.  When it cools, it lowers. 

 

Is this air getting in, or water getting out or both! 

 

 

[system 4-50]

Water expands when heated.

[nicknorman]

20 minutes ago, frlrubett said:

SO..... when the system heats up the water level raises in the expansion tank.  When it cools, it lowers. 

 

Is this air getting in, or water getting out or both! 

 

this is normal within reason. What would be abnormal was excessive expansion such that the coolant overflowed the header, or shrunk back so the header was empty.

[TheBiscuits]

3 minutes ago, nicknorman said:

this is normal within reason. What would be abnormal was excessive expansion such that the coolant overflowed the header, or shrunk back so the header was empty.

 

It's the glass half full/empty thing.  An engineer would say the glass/header tank is the wrong size!

 

 

[nicknorman]

2 minutes ago, TheBiscuits said:

 

It's the glass half full/empty thing.  An engineer would say the glass/header tank is the wrong size!

 

True but I was starting with the premise that the header tank was adequately sized and some abnormal condition was causing the over and under flow.

Edited December 10, 2021 by nicknorman

[koukouvagia]

I have a Refleks heating radiators and calorifier via a 12v pump.  When it's going full blast, about 2 litres a day will evaporate which I relplenish and regard as normal.

One thing I did discover: when bleeding radiators - a chore which I have to do after the system has lain dormant during the summer - I have to turn off the circulation pump.  Otherwise air is sucked into the system.  It may be that the position of the pump could be changed, but since everything else works perfectly, I can live with it.

 

[Tony Brooks]

9 hours ago, TheBiscuits said:

 

It's the glass half full/empty thing.  An engineer would say the glass/header tank is the wrong size!

 

Providing there is no air trapped in the system as detailed in an earlier reply.

[koukouvagia]

3 minutes ago, Tony Brooks said:

 

Providing there is no air trapped in the system as detailed in an earlier reply.

I found fitting an automatic air vent valve mostly solves this problem.

 

 

[ditchcrawler]

2 hours ago, koukouvagia said:

I found fitting an automatic air vent valve mostly solves this problem.

 

 

Reading about your bleeding above needing the pump off, doesnt the autovent let air in when the pump sucks. I have an upstairs radiator in the house that does the same thing and considered an autovent but thought it may make matters worse.

[koukouvagia]

1 hour ago, ditchcrawler said:

Reading about your bleeding above needing the pump off, doesnt the autovent let air in when the pump sucks. I have an upstairs radiator in the house that does the same thing and considered an autovent but thought it may make matters worse.

I see what you mean, but it doesn't appear to make matters worse.  I've never taken it apart, but I assumed there was a one way flap of some kind that lets air out but doesn't allow any in. 

[Ex Brummie]

An auto air vent works in the same way as a float chamber on a carburettor (Remember those?). If air collects beneath it, then the float falls, the vent is opened until the level of water lifts the float. The problem you get with them is that in regular operation the exhaust port gets mucky and you have a persistent leak from the unit. If the pump causes the float to fall, then yes, air will be sucked in compounding your problem.