Another back boiler post

[towpathrider]

I have read the other back boiler posts, but I know most systems have their individual quirks, so wanted to post ours and get some feedback. Our system so far just has the 28mm loop with a substandard header tank arrangement, so I'm looking to upgrade with a couple of rads.

 

Spcifically there are a couple of things I wanted to check:

1. Do I need two pipes going to the header tank? If so, why? (header tank will be vented, and unpressurised)

 

2. I will have to place rads either above of below pipework, (can't really put them over the top?) will the rad below in the bedroom work?

 

[Mendip-Locks]

I have read the other back boiler posts, but I know most systems have their individual quirks, so wanted to post ours and get some feedback. Our system so far just has the 28mm loop with a substandard header tank arrangement, so I'm looking to upgrade with a couple of rads.

 

Spcifically there are a couple of things I wanted to check:

1. Do I need two pipes going to the header tank? If so, why? (header tank will be vented, and unpressurised)

 

2. I will have to place rads either above of below pipework, (can't really put them over the top?) will the rad below in the bedroom work?

 

Ok not sure if your drawing was as you would pipe it or an impression of where the rads and pipe work is? But the hot needs to enter the rad at the top and the return to the boiler at the bottom both can't be on the bottom.

1. Yes one feeds the water to keep system topped up the other over the top into the tank for the expansion.

2. Not how you have drawn it, the return won't rise up and get back to the boiler needs to flow sloping down to boiler. It's a gravity system hot water rises to start it can't then rise again after it's been cooled by convection from the radiator.

Hope that helps, there ar many ways to achieve a system though as you will see trawling through posts on here.

Good diagram here from Smiley Pete.

 

http://www.canalworld.net/forums/index.php?showtopic=72184&page=2

[Mendip-Locks]

Ignoring the 2 left rads pump etc and all the 15mm pipe work apart from the 15mm expansion down to return!

[towpathrider]

Thanks for the diagrams, I had seen them, but my problem as I mentioned is I can't put the rads between the main pipes, they're currently only about 20cm apart, and pass through numerous bulkheads. So I'm trying to plan a system where the rads either go below, or above the pipes. In all cases, the top of the rad will be above the hot (top) back boiler outlet.

 

Here's a re-drawn diagram.

 

[spadefoot]

Rad 2 won't work. Rad 1 will but you need a vent pipe from the highest point of the pipework, feeding into the header tank.

Instead of rads, you could put some (several) copper pipes to connect between the hot & cold pipe runs. Use compression fittings & a smaller bore, 22mm perhaps. You will still need a vent pipe from the highest point on the loop into the header tank

[Tony Brooks]

Ok not sure if your drawing was as you would pipe it or an impression of where the rads and pipe work is? But the hot needs to enter the rad at the top and the return to the boiler at the bottom both can't be on the bottom.

1. Yes one feeds the water to keep system topped up the other over the top into the tank for the expansion.

2. Not how you have drawn it, the return won't rise up and get back to the boiler needs to flow sloping down to boiler. It's a gravity system hot water rises to start it can't then rise again after it's been cooled by convection from the radiator.

Hope that helps, there ar many ways to achieve a system though as you will see trawling through posts on here.

Good diagram here from Smiley Pete.

 

http://www.canalworld.net/forums/index.php?showtopic=72184&page=2image.png

 

 

But on my ex hire boat that is exactly what they did as do/did single pipe domestic central heating systems. It worked well. The pipe leaving the boiler at the back ran the full length of the boat just under the gunwale. it then dropped down to just above floor level and returned to the boiler. There were three radiators fitted with both feed and return from each in the lower pipe. This was a gravity system with no pump.

[towpathrider]

Am I right In thinking Rad 2 won't work because the cold outlet has to rise in order to get to the return pipe?

 

Would the siphoning effect work if the cold outlet is above the level of the cold inlet on the back boiler?

 

Lastly, would it be weird to mount a radiator like rad 1, though in the bedroom it would be on the wall. Would the heat distribute throughout the room much less efficiently than one mounted further down? LIke this:

 

[towpathrider]

 

 

But on my ex hire boat that is exactly what they did as do/did single pipe domestic central heating systems. It worked well. The pipe leaving the boiler at the back ran the full length of the boat just under the gunwale. it then dropped down to just above floor level and returned to the boiler. There were three radiators fitted with both feed and return from each in the lower pipe. This was a gravity system with no pump.

I think that the way this would work wold be through convection, rather than thermosyphoning, and I'm led to beleive thermosyphoning is more efficient, as the water is effectively pumping itself around the system.

 

[WotEver]

I think that the way this would work wold be through convection, rather than thermosyphoning...

My understanding is that thermosyphoning happens via convection.

 

Random Google find: http://www.sciencedirect.com/science/article/pii/S0378778810000502

[towpathrider]

My understanding is that thermosyphoning happens via convection.

 

Random Google find: http://www.sciencedirect.com/science/article/pii/S0378778810000502

Yes, but you can have convection without thermosyphoning right?

Edited August 13, 2016 by towpathrider

[Tony Brooks]

Well gravity circulation is thermo-syphoning, that is why the water travels around the main pipes. With a connection at the top of the rad you are relying on water cooling in the pipe and dropping down into the radiator while with the connection in the bottom you rely on the hot water in the pipe rising up into the radiator displacing the cold water back into the main pipe by the other connection.

 

Of the two I would expect two bottom connections to be slightly superior in getting the circulation through the rad running but at the expense of feeding cooler water to the succeeding rads.

Edited August 13, 2016 by Tony Brooks

[WotEver]

Yes, but you can have convection without thermosyphoning right?

If you have convection in a closed loop then surely the water must be circulating. Isn't that the definition of thermosyphoning?

[David Mack]

Thanks for the diagrams, I had seen them, but my problem as I mentioned is I can't put the rads between the main pipes, they're currently only about 20cm apart, and pass through numerous bulkheads. So I'm trying to plan a system where the rads either go below, or above the pipes. In all cases, the top of the rad will be above the hot (top) back boiler outlet.

 

Here's a re-drawn diagram.

 

 

I think rad 1 will work with the hot connection at the top or the bottom of the radiator. Rad 2 won't work at all in either configuration, as the less dense hot water will just stay in the pipes above, and never get down to the radiator.

 

I would strongly advise that you should have a separate feed and vent pipes to the header tank, with a continuous rise from the boiler to the vent pipe discharge above the tank. But it can be done with a single combined feed/vent pipe, but again this must rise continuously from the boiler. So the arrangement connecting to the bottom pipe, as in the drawing quoted, will result in airlocks which cannot be cleared. Stick with a layout more like the first one, but with the feed/vent pipe connecting to the highest point of the main loop.

 

All of these are pretty poor compromises. You really want the out and back pipes of he loop to be spaced apart vertically, with the rads in between. If you are stuck with the current pipe layout, then a pump may well be needed. You will still need a continuous rise from the boiler to the vent, to vent air when filling, and to allow steam to escape if the boiler boils (which it might if the pump fails).

[Tony Brooks]

I would not like to go to the expense of trying it but if those who advocate taking the hot feed from the top pipe into the top of the rad are correct in that it works then what is the difference between that and rad 2 in the diagram. I suspect it may well work.

[towpathrider]

All of these are pretty poor compromises. You really want the out and back pipes of he loop to be spaced apart vertically, with the rads in between. If you are stuck with the current pipe layout, then a pump may well be needed. You will still need a continuous rise from the boiler to the vent, to vent air when filling, and to allow steam to escape if the boiler boils (which it might if the pump fails).

 

What do you think of the third diagram? I don't understand how hot water circulates into a rad, if the hot connections go down into the rad. But this how hot water feed is connected in smileypete's design, with the rads in between the pipes.

 

I should also mention that my diagram is inaccurate in that the hot feeds for the rads will not rise and then go down, but instead rise steadily to the rad inlet. This way hopefully each rad should thermosyphon, as in the solar diagram above?