Newsletter 11 – Insulate a Hot Water Cylinder

Newsletter 11.

A Tranquility House Newsletter

Newsletter Logo

How to Really Insulate a
Hot Water Cylinder


Mike Hillard

When designing the Tranquility hot water system it was instantly obvious there was an anomaly.
Consider this:

The temperature difference between the top floor of a house and the roof space in mid-winter is generally around 14° (loft at 4° and bedrooms at 18°) and the right level of insulation is 350mm.

The temperature difference between a hot cylinder and the space around it is about 50° (cylinder at 75° and space around it at 25°) yet the cylinders generally have 60mm and a maximum of 80mm around them. 350mm to protect 14° yet just 80mm for 50°! Clearly something is wrong especially given the top floor only has to protect that small difference for some months while the cylinder is losing heat all year. These diagrams show the problem:

diagram of temperature difference between a bedroom and loftdiagram of temperature difference between a hot cylinder and the space around it

We will go through how to upgrade your cylinder’s insulation step by step to match what has been achieved here, and I promise you are not going to be spending much. My prime objective is to achieve everything at minimal cost – it must be affordable – so let’s first see what has been achieved here in Tranquility. This is your target.

I am not aware of any other building anywhere that has the research data available here – always a prime objective if I was to know what had been achieved and therefore how to further improve it. T is a living, working test rig so here are the charts for a normal seriously well insulated conventional 140 litre cylinder – and the 140 litre one in T. Both were heated to exactly 76° and left for 42 hours. 3 sensors were placed near the top, middle and bottom – the red, blue and green plots:

Decay curves for the really well insulated normal cylinder

Decay curves for the really well insulated normal cylinder

Decay curves for the Tranquility cylinder

Decay curves for the Tranquility cylinder

I am sure you will spot the difference but please note:

  • No water was drawn from either; the connecting pipes were extremely well insulated but some heat was lost from them. More is lost through the T test because the cylinder is hotter. The savings shown below are therefore an underestimate.
  • The bottom sensor showed the very rapid loss of heat from the bottom of the normal cylinder, compared with the relatively slow losses from the T cylinder. The bottom of the normal cylinder dropped to 30° which, with that amount of heat being lost, was probably the temperature around the cylinder at that time.
    It took about 8 hours for it to drop by 36° to 40° whereas the T cylinder only lost 11° in that time.
  • The T cylinder only had about 80mm of fibre underneath it – so you can beat it.
  • The heat lost from the bottom of the normal cylinder was so great that even the middle sensor was showing a dropping temperature within that same 8 hours. In the T cylinder the middle sensor never dropped below the top one. The blip in the middle sensor was when I checked it! They are very sensitive and accurate so a quick sensor withdrawal instantly shows.
  • The normal cylinder loses some 21.55 kw.hrs of heat a week but the T cylinder only loses 7.99 kw.hrs a week (both figures include pipe losses). The potential saving is therefore more than 63% or about 2/3rds.

How does heat move?

(A repeat of this section in the ‘Underfloor Heating’ Paper)

In order to know how best to work with heat we need to understand some scientific facts – but don’t worry – we are not going technical.

  • Heat will always move from a warmer material to a cooler one no matter what the material is – solid, liquid or gas.
  • It only moves in 3 ways:
    • Conduction
    • Radiation
    • Convection

Conduction: Heat will simply move through any material from warmer to cooler parts of it no matter the direction which is important. There is a simple test you can do to show this. Boil a kettle and taking a saucepan (thicker bottomed will demonstrate better) pour some in. Quickly put your hand underneath and you will feel it warming, but be careful in case it gets quite hot. The heat moved from the top surface to the bottom very quickly didn’t it! Conduction at work. Downwards.

Radiation: All the heat arriving on Earth is radiation coming all the way from the sun, but hold your hand in front of a hot radiator and you can feel horizontally radiated heat. Lift a boiling saucepan (DON’T spill anything) and hold your hand close underneath. Feel the heat? That is radiated downwards so heat radiates in all directions equally.

Convection: Now we do have a heat transfer that only goes upwards but it is the only one that does. If you know of lava lamps (in which coloured oils move upwards from the light bulb), that demonstrates it, but for a bit of fun (and to teach any children) put some cold water in a glass jug, casserole or bowl that can be heated directly (Pyrex can be). Very carefully dribble a little food colouring down one side and let it settle on the bottom on one side, or better still get a tiny bit of potassium permanganate crystal from the chemist and drop that down one side (it dyes the water a lovely purple). Then warm the water mostly on the side with the colour and watch what happens. Convection at work, which shows it only goes upwards.

How to insulate your cylinder?

The total heat lost is simply what escapes from the outside surface, so that is what we have to minimise. The heat leaving that will only be the heat conducted and radiated away from it, which ignores the convected heat because it cannot convect away until the air around it has warmed up – and it only warms from the other forms of heat transfer.

To minimise the heat lost we therefore have to do just 4 things:

  1. Minimise the radiated heat lost – by reflecting back the radiation that is going out, using the highly reflective surface of Aluminium foil in the right place(s).
  2. Minimise the temperature of the outside surface – by using as much normal insulation as possible between the cylinder itself and the outer surface.
  3. Minimise the heat transfer away from the outside surface – by using a material that doesn’t transfer heat to the surrounding air easily.
  4. And remember the cylinder has a top, sides AND a bottom.

We will now see what the end product should look like so you may be happy to go for it from here, but as the cylinders are usually in awkward and confined spaces we will continue with some suggestions and a sequenced plan of action as it is easy to do something only to find you did it at the wrong time!

All around the cylinder we are looking to achieve what is shown in the diagram here:

Diagram 1

diagram showing the target insulation around a hot cylinderTarget insulation around
a hot cylinder

  1. Next to the cylinder is a layer of bubble wrap – any format.
  2. Immediately around that is a layer of Ali Foil – shiny side to the cylinder.
  3. Outside this is the insulation, preferably in two 80mm layers but two 100’s are of course better.
  4. Outside this is another layer of Ali Foil – also shiny side to cylinder.
  5. Outside this is a final layer of the bubble wrap.

Note: you may smile or panic at the idea of wrapping Ali Foil around it without puncturing or tearing the foil, but don’t worry too much. This is simply working as a reflector, so if 10% (or 1/10th) of the total surface were not there, it would still be doing 90% of the job. I would hope if you damaged 10% of the surface you would try again! When you have done this you should be reducing the heat lost by at least 63% which is a huge reduction from a really well insulated cylinder.

Guidance and some explanations

We will look at some technical information which will be in blue, but if you are doing what is suggested you don’t need to understand this so can miss it. It might though be interesting as all this (and more) determined the cylinder insulation solution in Tranquility and above.

The first thing to understand is that different surfaces have different reflectivities – not news if you have read Part 2 of the Climate Book Series. The rules for Planet Earth are the same as here. You will easily see this in materials you have: we need the highest reflectivity facing into the cylinder so heat being radiated out is reflected back.

But there is another extremely important thing going on – called surface emissivity. Some surfaces emit heat (radiate it) easily and some don’t and you won’t guess which does what. Ideally we need materials facing out that don’t emit it easily.

And yet another thing going on (working with heat isn’t as simple as we may have thought) called surface absorptivity. Some surfaces absorb much of the heat arriving at them – and some don’t absorb much at all. Solar Hot water flat plates can only be efficient if they have very high absorptivity and low emissivity, and this is taken into account when designing and manufacturing them.

We are therefore looking for materials facing out that emit little and absorb much, so here at least are some radiation (emissivity) comparisons for the materials involved – the lower the better:

  • Copper (the cylinder?) varies between 0.78 and 0.23. The 0.23 is a highly polished surface that no new cylinder has, and the 0.78 is for a dull oxidised surface which is the figure we need to use – unless you regularly wire wool and polish the copper!
  • Stainless Steel (the alternative cylinder) varies between 0.85 and 0.01 – no miss-type. The 0.85 is for heavily weathered external stainless; the 0.01 for mirror polished stainless (which you have surely never seen) and 0.075 is for normally polished stainless. The figure applicable depends on the degree of surface polish but 0.15 might be a sensible average, though it declines over time.
  • Aluminium foil ranges between 0.10 and 0.04. Take it the 0.04 is the shiny side and the 0.10 the dull side.

Clearly a copper cylinder is worst, but even the stainless version must be improved on.

The shiny side of the Ali Foil will reflect back about 96+% of the heat arriving at it; the dull side around 93%.

Problem 1: The cylinder has a steel outer case and you don’t know what is inside that:

There is probably 60mm of fibre inside so leave it as it is but continue outside exactly as in Figure 1 except you only ideally need around 100mm of additional fibre. If yours happens to have sheep’s wool insulation inside – it is no better than any other fibre (but costs lots more). The underside is very unlikely to be insulated so treat it as if it isn’t.

Problem 2: The position of the Cylinder:

Your cylinder is probably either sitting on the floor or on a raised plinth. If it is on a plinth there will be a cavity under the cylinder above the plinth which benefits from being filled if you can.

Diagram 2

diagram showing insulating under the cylinderInsulating under the cylinder

If you can gain access above the plinth – probably by drilling a smallish hole or chiselling gently through, be extremely careful – then you can inject expanding foam. But you will probably not do this!

If not, simply treat the underside of the plinth as the underside of the cylinder as shown in Diagram 2.

If the cylinder is on the floor, by either removing an odd length of board (be careful if sawing as there will be pipes under there somewhere) or by drilling a few holes through it around the cylinder – you will gain access. Then get as close as you can to the solution in Diagram 2 above.

The better solution is to raise the cylinder, which isn’t impossible especially with plastic plumbing, but will cost something.

The jobs in the best order

Job 1: Cover all surfaces with bubble wrap as that is the cheapest insulant and spacer I can find.

Job 2: Cover the bubble wrap with aluminium foil but don’t forget it is shiny side in. Two layers are OK if the first gets damaged, but there isn’t a thermal benefit.

Job 3: Fix the fibre to the outside of the foil as follows:

diagram showing the layers of fibre wrapped around a hot cylinder
  1. Start by shaping some to ‘level’ the top.
  2. Next, wrap one layer round the outside.
  3. Then do the circle for the top which now extends to the outside and tape these together. By doing it this way there is something to hold it up to.
  4. Repeat a second outer layer around the cylinder.
  5. Finish with the final large top ‘circle’.
  6. Finally tape the outer around the cylinder to the top to hold it up, and do seal round that top edge.

In the end so long as the cylinder is continuously covered and sealed – you have done the job.

Job 4: Don’t laugh, and isn’t difficult, but you now need the other layer of Ali foil around the whole thing, again shiny side in. By having both layers shiny side in, both are reflecting nearly everything arriving back in towards the cylinder, but the energy bouncing between the two will mostly be reflected in from the outer layer and then absorbed by the inner layer. Hold it by taping it to the top just as in 3 and 6 above, and finish with foiling the top.

Job 5: Nearly there, but while the foil on the outer surface may not be radiating much, it is a near perfect conductor so will warm the surrounding air which can then conduct heat away.

The final layer is therefore one of bubble wrap which will perform two jobs. First it is a bad conductor so will minimise the heat transferred to the air, but secondly it will protect that outer foil layer which is very delicate. Wrap a few layers around and tape together securely.

Job 5: Time for congratulations and maybe a small reward. Perhaps a glass of something you like in the knowledge you have done yourselves and the planet a big favour. For a tiny amount of money you will do what most products don’t – you will save buckets of energy and yourselves some cash every year.

Remember the cash saving from reducing the heat losses as shown in T would amount to 19.6kw.hrs a week or almost exactly £50/year, but that is starting from a very well insulated cylinder which you probably don’t already have.

The financial payback for doing this should be huge as the material cost is minimal, and you will have helped planet earth and our children by slowing carbon emissions. I am often asked what people can do as they feel impotent, so here is something you can really do.


  1. You will possibly or probably not have a clear 160mm around your cylinder and it may be in a square cupboard. Get as close to Diagram 1 as you can, filling the corners. Rigid polystyrene has the same insulation value as the fibres, so if it is easier to use that – do.
  2. Improving on 1 above: if there is a stud wall around the cylinder (a hollow timber framed wall) you can drill small holes in it and fill them with expanding foam so the wall is part of the insulation.

You can obtain large rolls of bubble wrap from commercial stationery suppliers for around £45, or huge ones (1.5m wide x 100m long) for around £55. The latter would do more than 10 very large cylinders, so if you can get a group together that will be the best way to acquire that.

One Final Message

It feels like a mantra here, but we must all be constantly aware it is always ‘Buyer Beware’. No manufacturer ever makes the best product they can – just the one they can most profitably sell.


This entry was posted in News and tagged , , . Bookmark the permalink.