HW Cylinder - lose heat through convection

[fixitflav]

It was most likely my thread!..

It just seems a bit daft, assuming this is/was a common thing, to blame cylinder heat loss on losses through the cylinder insulation.

Found this, but I don't think it's much help in explaining yours

C-Plan System boiler and tank location

Hi, My daughter has just purchased a house with a C-Plan boiler system installed. The boiler and tank used to be next to each other in cupboard, but the pervious owner has moved the boiler to the loft. My understanding is that the hot water works on a gravity feed, but with the boiler in the...

www.diynot.com

 

[Harry Bloomfield]

Found this, but I don't think it's much help in explaining yours

As said - it's a bit of a curiosity, which appears to be defying the laws of physics, unless - the convection flow, is being driven by the boiler itself?

If the flow sections of my Vaillant 18Kw boiler, is lower than the return sections, warm water could be drawn down the return, cooled in the boiler, then cooled water appear at the flow, circulating back to the cylinder coil, to be reheated by the HW in the cylinder.

 

[Alec_t]

I'm grasping at straws here.
In a hot water cylinder the water at the top is usually warmer than at the bottom. If the pump and boiler are inactive, the water at the top cools a bit faster than the water at the bottom (loss of heat being proportional to temperature excess) and so will sink towards the bottom. Likewise the water in the coil will also cool and sink. Since the water circuit is complete then I think the water will tend to circulate, albeit very slowly, leaving the tank coil at the bottom and re-entering at the coil top, i.e flowing from the cylinder via the boiler return and re-entering via the boiler flow. This would make the return warmer than the flow.

 

[Harry Bloomfield]

In a hot water cylinder the water at the top is usually warmer than at the bottom. If the pump and boiler are inactive, the water at the top cools a bit faster than the water at the bottom (loss of heat being proportional to temperature excess) and so will sink towards the bottom.

I accept it will cool at the top, but in my opinion it will always remain warmer than the water below it, therefore it will remain at the top, rather than sinking down.

My thinking at the moment, is that the reverse flow, is most likely due to the layout of the boilers heat exchanger. The convection flow, due to the way the boiler is cooling the water. It near impossible to quantify how much heat it's wasting - I can measure the temperature drop across the boiler, but not the reverse flow.

 

[fixitflav]

It just seems a bit daft, assuming this is/was a common thing, to blame cylinder heat loss on losses through the cylinder insulation.

I just felt my flow and cylinder return pipes at the boiler, and they're warm, guess 30-35°C. Rads return stone cold. But I don't know how long ago there was a HW cycle. I'll try again sometime. They're 28mm pipes as it was converted from a gravity HW system.

 

[Harry Bloomfield]

I just felt my flow and cylinder return pipes at the boiler, and they're warm, guess 30-35°C.

You may be suffering a similar waste of heat to me. The only to know which way the flow is going, is by accurately measuring the flow and return pipes, at the boiler.

Is your a standard type 3-port, which always parks in the HW open?

 

[Alec_t]

Are you sure you are detecting heat loss via convection, rather than conduction? Water-filled metal pipes will be quite good at conducting heat from a hot cylinder to a cool boiler. The heat flow may be unequal in the flow and return pipes due to diferences in length, bends, insulation, hence giving the temperature difference you measured.

 

[fixitflav]

You may be suffering a similar waste of heat to me.

Too early to say. Been out for a while, just felt them again and they're quite hot, so it must have recently finished a HW cycle. I'll keep trying.

Is your a standard type 3-port, which always parks in the HW open?

Yes, it's an either/or valve Honeywell V4044 on a W-plan system. I've ranted before about not seeing significant advantage of Y-plan, with less reliable mid-position valve, or S-plan with more complicated wiring, over W-plan.

 

[fixitflav]

Are you sure you are detecting heat loss via convection, rather than conduction? Water-filled metal pipes will be quite good at conducting heat from a hot cylinder to a cool boiler. The heat flow may be unequal in the flow and return pipes due to diferences in length, bends, insulation, hence giving the temperature difference you measured.

I hadn't thought of that, but it seems unlikely. Have you done an estimate of heat flow? I'll do it when I have a minute.

 

[fixitflav]

You may be suffering a similar waste of heat to me. The only to know which way the flow is going, is by accurately measuring the flow and return pipes, at the boiler.

Felt them again, and the pipes are same temperature as the rads return, cold. So I haven't got the problem. I have pipes about 6m each, flow and return, horizontal, and about 400mm from top of boiler to bottom of HW cylinder

 

[fixitflav]

Are you sure you are detecting heat loss via convection, rather than conduction? Water-filled metal pipes will be quite good at conducting heat from a hot cylinder to a cool boiler. The heat flow may be unequal in the flow and return pipes due to diferences in length, bends, insulation, hence giving the temperature difference you measured.

I calculate with 5m of 22mm copper pipe. 0.9mm wall thickness, hot end 60°, cold end 20°, heat flow by conduction in the pipe wall = 0.2 watt, and in the water 0.0015 watt i.e. negligible.