self build waterproof basement formwork membrane concrete insulation

Insulating a home with a basement.

I'm not selling anything here. This page is my conclusion and recommendation after years of going back to customers and asking them how what they had chosen worked out.

ICF did not work out well. Nor timber frames or SIPs. Heat pumps generally failed.

What worked best of all, and for less cost, was thermal mass and heat recovery ventilation with air tightness.

  self build basement

ICF basement house   In 2012 I built a basement and all the house walls with ICF.

These clients initially thought they had achieved zero energy bills for their new home with U Values around 0.20 rather than 0.10.

See a video I made in 2014, two years later, explaining this basement and its heating here.

These clients thought that their fuel bills would be zero because
  1. Photo Voltaics on the roof,
  2. Ground Source Heat Pump bringing heat in,
  3. Heat Recovery Ventilation saving some heat before it is lost outside,
  4. Air Source Heat Pump in the basement to extract even more heat from the air on its way out to heat their water,
  5. 150mm of ICF insulation above ground,
  6. Roof trusses stuffed with insulation and
  7. Excess electricity sold to the grid.
When I made the video in December 2014 the weather had been very cold already. They calculated that in that first month they spent just £26 on buying energy. The rest came from their GSHP. On an annual basis, after selling excess electricity in Summer, they thought their bills would be about zero.

But when I returned again in April 2017 to catch up on old times their bills were £1,000 a year for energy.

Something had gone wrong. Actually, two things in particular seemed to have gone wrong.
  1. The GSHP had got expensive. It no longer cost less than £26 a month to heat their house. It was costing 10 times that because the ground had become too cold.
  2. With no thermal mass because of the ICF insulation inside, they got too hot in Summer and installed air conditioning.

The guys selling ICF, timber frames and SIPs are all trying to sell you a building method that wastes the opportunity to store energy in thermal mass - in concrete.

True, it takes less energy to warm the air in a cold house. But why should the house be cold?

True, it takes less energy to cool the air in a warm house. But why should the house be too warm?

True, all these choices are more air tight than brick and block cavity walls used to be. But other methods of construction have upped their games since.

BUT NOTE: the ICF description only says "making it possible". Not that it will.

And NOTE: the timber frame description puts made from renewable woodland first. Not saving energy.

AND: the SIPS description only compares SIPs to "older technologies".

My analysis is that ICF, timber frames and SIPs all perform poorly compared to newer technologies and building methods.
I Googled "benefit of ICF / timber frame / SIPs. These came up first:

ICF basement

timber frame basement

SIPs basement

That still leaves the most common question I get asked: Sould you insulate inside or outside a basement? I answer that lower down.

The ground source heat pump and the air source heat pump guys are trying to let you believe that the efficiency in Summer is so good that even in Winter you save money. But put on the spot they can't tell you that you would save anything on the coldest days when you need the most energy.

It seems remarkably difficult to get a promise from any heat pump company that you will save much money when it is cold outside. So I tend to study their choice of words very carefully:

I read an article on page 11 in the Sept/Oct 2017 issue of Selfbuilder and Homemaker magazine hopefully still here by a Mitsubishi air source renewable heating expert. I was struck when I read "While this will need slightly more electricity in the colder months, the seasonal efficiency throughout the year will average out at 3.5 or more"

These articles never seem to go as far as to say that in the coldest weather heat pumps will still save you any money. That is why I suspect that claims of 3.5 times efficiency over the course of a whole year hide their failure to be as cost-effective as gas when you need heat most. Will the average still be 3.5 when you consider using 20 times the energy mid January compared to mid June? They don't say so, so no I don't think the average will be remotely 3.5. The average of 1 twenty times and 3.5 once is 23.5 divided by 21. Only a fraction better than not having it. A waste of money, probably, and a lot of noise all Winter as it struggles.

When I did the Google thing again for GSHP I didn't get a definition come up first. First I got ads then this box:

GSHP basement I was a bit intrigued that they express disadvantages first so I had a look around their site.

This is the same link to the same Icax page.

Firstly, they say Ground Source is better than Air Source.

Second they say Ground Source fails because it has not been designed properly.

I think that usually means you need many miles of coils so that you couldn't cool the ground enough for GSHP to lose efficiency before it gets a chance to warm up again next Summer.

But I think miles and miles of coils would be ruinously expensive to install. Even if you have enough land you would never get your investment back. I think that's why so many installations are poorly designed. If suppliers didn't make them affordable they wouldn't have any to install.

The Icax site promotes something else. Seasonal Thermal Energy Storage.

They tell us what we hear all the time, that it is very expensive to store electricity.

But, they say, it can be virtually free to store heat. What they seem to promote are two systems both with coils alongside each other buried in the ground.
  1. By Summer, solar energy is used to warm up the ground.

  2. By Winter, that same ground has the excess heat taken from it more efficiently by GSHP. If the GSHP takes more than was put there, that heat will get replaced next Summer. It still depends on enough coils under enough of your garden or else heating the ground up with more energy than you will need throughout the Winter.
From the Icax web site: "It is a characteristic of earth that heat only moves very slowly through it - as slowly as one metre a month."

Icax aren't talking about a house with a basement. If heat moves a metre a month they would want to bury coils at least 3.5m deep to store energy 3.5 months. Basement floor slabs are about 3.5m deep as well.

Here is how my journey got me to a similar place before I had ever heard of Icax or the term Seasonal Thermal Energy Storage.

This BRE information paper is over 20 years old and may no longer be considered accurate. However it suggests that an average size basement could have a U Value of 0.16 without any insulation. Just because it is buried.

Why, then, do we need any insulation under a basement floor when the floor is insulated from any cold air outside and above ground by 3m or so of earth at the edge, even more in the middle?

Two reasons come to mind.
  1. Above ground, an uninsulated concrete or masonry wall would be unthinkable because it would lose a huge amount of heat in Winter to cold air outside. So concrete has to be insulated.

  2. If the earth beneath a basement is 10°C and a basement heated to 20°C then a basement needs to be insulated to prevent heat loss.
But what if the earth beneath the basement floor was warmer than inside the basement? The basement would receive heat.
BRE U Values
   Page 1    Page 2    Page 3

I first thought of this when I was told at an exhibition that photovoltaics on the roof generated more electricity if they weren't allowed to get too hot. Basically, they need water cooling.

That then means cooling the coolant before it goes round again to cool the panel some more.

A car engine coolant needs cooling through a radiator. Photo voltaic coolant needs cooling as well. How? Where?

Some years ago, a commercial developer had the idea of putting pipes through the basement floor slab before the concrete was poured.

Heat generated from solar power on the roof could be stored in the basement floor concrete.

My thoughts are:
  • If you want heat to come out of the earth into the basement you don't want any floor insulation.

  • If the basement floor slab got too hot the basement would be unbearably hot as well:

    • Perhaps ventilate the basement in summer. But then you might not store much heat.

    • Try to drive the heat energy lower and deeper into the earth beneath. The heat would only come out when the basement got cooler than the earth - mostly in winter at night.

  • If you pump free, solar energy into the earth and some escaped, well, it was free and of no harm to the environment anyway.
basement floor thermal mass
In the photo above the plastic tube will end up just beneath the top layer of mesh because it will float up in the liquid concrete till the top mesh stops it floating to the top. This puts the heat in the concrete only about 90mm from the surface with more concrete below than above.

basement blinded So I think the tubes need to go beneath the lower steel mesh and to preserve concrete cover you might need an extra 25mm of concrete under the steel.

But, actually, why have the pipes in the structural concrete. Why not below the structural concrete?

It would probably be easier to snake the pipes all over your formation level and then smother them in semi-dry, lean mix concrete blinding. One note of caution, if you plan to use my rods to get your slab level, you will want to find a way to avoid drilling through your pipes. Perhaps cast threaded rod through your blinding then cut them to level? Or you could wind them out and put them back after your steel is down.

I would maximise thermal mass: all the floors, all the walls, the roof and earth under the basement; water cool photo voltaics to generate more solar electricity and push solar energy into the ground over Summer. I would have Heat Recovery Ventilation, excellent air tightness and continuous insulation all over the walls and roof just on the outside. I might heat the water with an air source heat pump taking heat from warm air as it is expelled.

Let us compare the kind of cost I am advocating here compared to the ICF house costs in 2012.

ICF House Concrete House
Photo Voltaic cells Similar Similar
Ground Source Heat Pump A lot of money Zero
Air Source Heat Pump for water Fairly expensive Might have one as well
Thermal Mass They didn't have any A much as possible, virtually free
Internal layer of insulation About £10 a m² None. No cost.
Excess electricity sold to the grid Some More because panels water cooled
Investment cost Very high Very low
Annual energy costs £1,000 Near zero

seasonal heating cooling envelope Continuous insulation outside the thermal mass walls and floors.

Including across the concrete flat roof.

But no insulation beneath the basement.

Note the coils of water pipe from solar panels on the roof buried in the blinding.

Store solar energy in the ground beneath the basement throughout the summer.

That energy will find its way into the basement whenever the basement is cooler than the heated earth beneath.

You might not like any of my thinking. Fair enough. I'm not selling it, I'm only sharing it.

Just one more piece of experience to share.

I almost forgot to mention that many clients put underfloor heating in their basements, and they don't turn it on. Ever.