Horizontal cylinder in a 'cold' loft. Bad idea?

[Post and beam]

I did think about a horizontal cylinder in the loft. A discussion elsewhere advised against on the grounds of the loft being un heated.

 

Now, it is true there will not be any heating in my loft but the rafters have 160mm+ of Kingspan and are designed at U = .11

 

Is a horizontal cylinder a terrible idea?

[crispy_wafer]

So within the thermal envelope of the building?  If so, then getting it up there, removal if and when, and Structural elements aside I don't see why you couldn't.

[Post and beam]

1 hour ago, crispy_wafer said:

So within the thermal envelope of the building?

Yes.  I have had different opinions on what constitutes a cold roof.  Hence me saying the space is unheated but well insulated.

 

An issue put forward is that the horizontal nature of the tanks makes mixing of the hotter & colder water less efficient.

[Iceverge]

2 hours ago, Post and beam said:

I did think about a horizontal cylinder in the loft. A discussion elsewhere advised against on the grounds of the loft being un heated.

 

Now, it is true there will not be any heating in my loft but the rafters have 160mm+ of Kingspan and are designed at U = .11

 

Is a horizontal cylinder a terrible idea?

Unheated is not a problem so long as it's within the heated envelope. 

 

What is your preferred distribution system? Radial or trunk and branch? Are you planning a hot return loop?

[JohnMo]

I utilise a loft section as my plant room.  The room is all within the thermal envelope so insulation is at roof level.

 

Have an MVHR Extract in there, so any heat loss from the cylinder is recovered. 

[Post and beam]

32 minutes ago, JohnMo said:

 

Have an MVHR Extract in there, so any heat loss from the cylinder is recovered. 

Good idea, thanks

 

1 hour ago, Iceverge said:

What is your preferred distribution system? Radial or trunk and branch? Are you planning a hot return loop?

Dude, i am not a plumber i have very little idea. I am just trying to accommodate a cylinder in the least worst location.

[Iceverge]

My 2pence worth. 

 

Put the UVC as near as possible to the kitchen tap. 

 

Run a radial system with narrow bore pipes to the wash basins. 

 

Put the cylinder over a tanked floor with a level floor drain such that if you even spring a leak then the house won't be ruined. 

 

Don't trust plumbers just because they've got a ticket. Ours made a mess of our house. 

 

 

 

 

[SteamyTea]

2 hours ago, Post and beam said:

An issue put forward is that the horizontal nature of the tanks makes mixing of the hotter & colder water less efficient.

I have never seen it modelled, so hard to tell.

 

Take a cylinder with a 0.5m diameter ad a 1.2m height.

Assuming a U-Value of 0.2 W.m^-2.K^-1 

 

Volume will be 235 lt.

Surface Area will be 2.28 m² 

Base of cylinder temperature, once settled, 36°C, top of cylinder temperature 60°C.

Ambient Temperature 10°C.

 

If one assume a temperature gradient of 20K.m^-1 (about what mine is), then the power losses, when vertical, will be the sum of the top end, plus the area of the diameter (hoop), then the sum of the hoops, and finally the sum of the last band and the base area.

Using a course 0.1m down the cylinder, the power losses are 18W, or if the cylinder is unused for a day, 0.44 kWh.

 

Now lets turn the cylinder horizontally.  Working out the surface area is not quite so easy here as for every 0.1m loss in height, the end area and the hoop area do not scale in a linear fashion, so I sketched it up in CAD, sliced it, triangulated it, then worked out the dimensions.  Accumulative errors was between 1 and 8%, so shall use 4% as the error.

The cylinder power losses are now 20W, 0.49 kWh.day^-1.  A difference of 0.05 kWh.

 

The above is on a static model, but there will be some turbulence.

With a mean temperature of 48°C for the vertical temperature, the mean density of the water is 988.7 kg.m^-3, at the top of tank temperature, the density is 5.53 kg less, 4.48 kg more at the bottom.  A total of 10 kg.m^-3 difference

The horizontal cylinder only has a 12°C temperature difference (because I used the same temperature gradient of 20K.m^-1), so the density difference is only 5.6kg.m^-3.

Now without getting into Reynold Numbers and tangential surface areas, a simple way to model it would be to look at the difference in stored energy and the difference in mass as energy is the ability to do work, which can be reduced to moving a mass a distance.

The vertical cylinder will have 233 kg of water in it, the horizontal one 232.4 kg, so 0.6 kg less.

To move 1 kg of water, 1 metre, will take 1 joules of energy.  So to move 232.4 kg 1.2 metres will take 279.6 J in the vertical cylinder.  There is only 0.5 metres of height in the horizontal cylinder, so 116.2 J, so the turbulence losses will be in thee order of 42% less.

 

So I would not worry about the cylinder orientation.

 

I am going outside to sand some wood now the glue has set.

 

 

 

 

 

Edited August 10, 2023 by SteamyTea

[Post and beam]

57 minutes ago, SteamyTea said:

So I would not worry about the cylinder orientation.

Thanks dude, i am so impressed. 

[SteamyTea]

8 minutes ago, Post and beam said:

Thanks dude, i am so impressed. 

Not idea if it is right, but seems to make sense.

 

Basically, with good insulation around the cylinder, and keeping the top of cylinder temperature as low as practicable, the losses are probably the same.

Could be why there is no information on it.

[Adsibob]

3 hours ago, Post and beam said:

Yes.  I have had different opinions on what constitutes a cold roof.  Hence me saying the space is unheated but well insulated.

 

An issue put forward is that the horizontal nature of the tanks makes mixing of the hotter & colder water less efficient.

The loft will be warm, because the heat from rest of the house will heat it.

[Iceverge]

Do you have space for 2 x  slimline cylinders where you were thinking of putting the square tank? 

 

Alternatively if the loft is big a V large cylinder would give excellent performance. Heat pumps like large volumes at a low temp. Make sure you have space to access it to service/install etc. 

 

 

 

[SteamyTea]

2 minutes ago, Iceverge said:

Heat pumps like large volumes at a low temp

May need a bit of ceiling strengthening 

[Post and beam]

1 hour ago, Iceverge said:

Make sure you have space to access it to service/install etc. 

The loft hatch size is always going to be the main parameter of concern, followed closely by the weight of the thing i guess.

 

1 hour ago, Iceverge said:

2 x  slimline cylinders

Its more a case not only of space in the utility but the copper around it. SWMBO is nervous of what it might look like so doubling up wont help the situation.

 

If i could convince myself that a 200l cylinder would suffice then the pre packaged one mentioned before fits the bill. It does just look like a fridge after all. Does anyone know the size of the old short round fat copper cylinders we used to have in the airing cupboards as kids?

[JohnMo]

A 210L slimline is 450x1950mm ish depending on make.

 

My thoughts are there are generally 2 of us in the house, so 210L will be more than enough for us. The cylinder has an immersion, so if we need more capacity I will flick on the immersion heat the cylinder to higher temp and leave on while people are showering.

[sharpener]

+1, we have 210l, it is fine for four showers in the morning when we have visitiors. I think the key is good stratification, will not be so good with a horizontal cyl.

[SteamyTea]

3 minutes ago, sharpener said:

+1, we have 210l, it is fine for four showers in the morning when we have visitiors. I think the key is good stratification, will not be so good with a horizontal cyl.

Except for the same top of cylinder temperature you can store more energy.

Stratification is a bad term to use as it implies a boundary where the temperature makes a discrete step.  That is not the case in a cylinder, why I use the term gradient.

There will also be a temperature gradient between the sides of the cylinder and the middle.  That gradient will 'flip' between heating mode, cooling mode and usage mode.

[Adsibob]

1 hour ago, Post and beam said:

followed closely by the weight of the thing i guess.

Yep! We had a 300L Telford tempest and the structural engineer was fairly relaxed about me installing it next to an external wall of the house, as minimum moment there on the joists. When I suggested it would be better in the middle of the floor, some 6m from the external walls, he said we would need to strengthen the floor.

[Post and beam]

11 hours ago, SteamyTea said:

a boundary where the temperature makes a discrete step.

Off topic but we used to use this to hide under in a Submarine, somewhere in the north adlantic.

12 hours ago, JohnMo said:

My thoughts are there are generally 2 of us in the house, so 210L will be more than enough for us. The cylinder has an immersion, so if we need more capacity I will flick on the immersion heat the cylinder to higher temp and leave on while people are showering.

 

11 hours ago, sharpener said:

we have 210l, it is fine for four showers in the morning

Thanks for these 2 comments. There are 3 of us in our house. I am thinking more of the future if we ever sell to someone with a house full. I am coming around to the idea that the 200l ' fridge' option will suffice in the corner of the utility.

[JohnMo]

Have a look at this size guide and go from there.

 

[SteamyTea]

1 hour ago, Post and beam said:

Off topic but we used to use this to hide under in a Submarine, somewhere in the north adlantic

Thermocline boundary layer.

More to do with salinity than temperature. The melting Greenland icecap is going to change that quite a bit in the coming years.