replacing existing floor with new slab - minimum excavation

[Bemak]

Hi All, 

Hoping to kick off a project I posted about here previously at the end of the year. It's been delayed for multiple reasons but hopefully it will get a bit of momentum in the next few months. House is an old 2-storey farmhouse, 140 years old with 600mm thick rubble stone walls. The house is in good shape considering it's rarely used and structurally very sound. I had been mulling over whether to dig out the slab at all as the existing floors are relatively sound. In the end, I've decided to dig out the existing floor as it will allow me to install UFH, a radon sump and membrane, and insulation. 

 

I was talking to a very experienced contractor about it recently as I was concerned about the potential knock-on of digging out the existing floor. He suggested that I keep the excavation to the minimum and suggested a build-up of, 100mm compacted hardcore, blinding, Radon barrier, 150mm insulation, 50mm liquid screen with UFH. 300mm overall. his argument is that the existing subfloor would already be well compacted and that the suggested buildup would minimise the potential of disrupting existing walls. I agree but I just wanted to put it out there to see if anyone has done something similar?

[Lears]

How did you get on with this in the end? I'm in a similar situation, old property, want to retrofit UFH etc and slightly worried about the state of the foundations although we haven't dug any tests yet to know if there'd be an issue or not. ..

[JohnMo]

There are two ways to do UFH, loads of insulation or no insulation at all - the middle ground between the two means high running costs. 

 

Both systems work using different mechanisms.

 

Loads of insulation stops downward heat losses, as the downwards heat flow breaches the insulation it never come back, as heat will allow try to move towards cold and the ground will always be cooler than a heated floor, so you want really good insulation. You can run intermittently if you need to.

 

No insulation uses the the ground as a buffer for heat, the zero insulation allow what goes down to come back up. However you just let the heating tick away all heat season. If you switch off the heat is dispersed and you need to rebuffer it all again.

[Gus Potter]

13 hours ago, JohnMo said:

There are two ways to do UFH, loads of insulation or no insulation at all - the middle ground between the two means high running costs. 

 

Both systems work using different mechanisms.

 

Loads of insulation stops downward heat losses, as the downwards heat flow breaches the insulation it never come back, as heat will allow try to move towards cold and the ground will always be cooler than a heated floor, so you want really good insulation. You can run intermittently if you need to.

 

No insulation uses the the ground as a buffer for heat, the zero insulation allow what goes down to come back up. However you just let the heating tick away all heat season. If you switch off the heat is dispersed and you need to rebuffer it all again.

What a belter of a statement.. heretic I say! That said, what John is presenting is based on sound arguement, although some may not want to hear it. 

 

Here is a bit of background. If we have a large industrial building that is to be heated to say an office (domestic) environment temperature then the first thing we look at is the ground. We ask..what is the ground.. I'll pick clay as it is the best performing.

 

Next.. is it pretty impermeable clay that has negligeable ground water cross flow movement? Any significant ground water cross flow will take any stored heat away. If the water flow can be neglected then we have what we call a "dumpling" in the middle of the building that we can heat and then later recover that. To get your head around this.. think of Adobe clay bricks.. the clay has insulating properties, normally associated with hot climates but the same principle applies as the clay does not know from which direction the heat is coming from.

 

Now we know heat can leak out the sides of the floor.. and if we heat the dumpling heat we need to know how much heat will leak out the sides, call that a thermal bridge.. we now have PSI values say that look at this effect in detail.

 

But on a large building the perimeter  / floor area ratio is small so much less heat loss out the sides. On a small extension much more heat gets lost out the sides in relative terms.

 

If you know a bit about large industrial floor design then often you'll just see the edges of the floor insulated. I've been dabbling in this for a long time as has @saveasteading. A lot of the modern raft slabs, passive ones also are based on techniques we were using since the 1980's, maybe before. 

 

Now on a smaller floor area we can adapt large building floor slab design by for example insulating the perimeter walls down to the founds and capture the heat in a clay dumpling. 

 

This basic understanding can be invaluable when upgrading old buildings.

 

Great comment John!

 

 

 

 

 

 

 

 

[Roger440]

On 27/07/2025 at 14:04, Lears said:

How did you get on with this in the end? I'm in a similar situation, old property, want to retrofit UFH etc and slightly worried about the state of the foundations although we haven't dug any tests yet to know if there'd be an issue or not. ..

 

I did this. Foamed glass wth concrete/limecrete (limecrete in my case)  over. Total depth from finished floor height, 225mm if my memory is correct. 

 

There were virtually no foundations, so took a cautious approach around the sides with a 45 degree taper.

 

Installed UFH as the same time. End result was fantastic.