Insulating cavity of existing beam and block sub-floor

[Dave Jones]

you also need to find out if the builder used block and beam due to ground conditions. e.g. heave.

 

do you have ceiling height to raise internal floors  ?

[embra]

On 13/04/2023 at 18:23, Iceverge said:

BEFORE

 

 

Without insulation. Assuming 100mm EPS in the cavity and full fill below ground level. 

 

 

 

 

 

AFTER

 

 

 

 

With 300mm of EPS under the floor. 

 

The U value drops significantly but given the effect of the thermal bridge not as far as you'd think. 

 

 

 

 

 

 

 

On 13/04/2023 at 18:34, Iceverge said:

 

Here's another option. 

Lay 75mm PIR boards over the block and beam. Carefully foam all joints, trim and tape. Then lay 2 x layers of 11mm OSB floating over the top, joints staggered, glued and screwed. If you wanted to bump to Phenolic insulation you could lower the U value to 0.24 from 0.28 w/m2K but I wouldn't imagine it's worth the cost. You can still see the effect of the thermal bridge at the wall floor junction. 

 

Then use large radiators for the ASHP. 

 

 

Personally I think that you could fill the floor void as is with EPS beads nothing bad would happen. I can't under stand how moisture would magic it's way vertically through 300mm EPS beads. 

 

 

 

Really interesting analysis, thank you for taking the time to prepare and share these. A great illustration that insulation detail and positioning can be more significant than improving thickness / thermal conductivity.

 

As a huge favour, would you mind running a couple more scenarios to add some data to ideas I've been toying with?

 

1) Repeat of the last scenario, albeit with a 50 mm only layer of insulation at a thermal conductivity of 0.019 W/mK (Kingspan K103)? Since I'm really constrained on stack height, I'm interested to see how the reduced thickness vs better thermal conductivity balances out. No insulation in the void.

2) As 1), but with a 120 mm layer of insulation at a thermal conductivity of 0.026 W/mK directly under the beam and block (to evaluate Q-bot application), as well as the 50 mm K103 insulation "inside" the house.

 

Most appreciated 🙏

[embra]

15 minutes ago, Dave Jones said:

you also need to find out if the builder used block and beam due to ground conditions. e.g. heave.

 

do you have ceiling height to raise internal floors  ?

 

Thanks for bringing this up, I am not familiar with assessing potential for heave - can you share any guidance or pointers? I'm going to have a look at this page to start with. Property was originally built ~1990 in case this rules out certain scenarios.

 

Internal ceiling heights are ~2.4 m from current floor finish, the larger issue with raising the floor height would be the effect of creating an unequal rise on the last step of the stair case from ground to first floor. Of course there is the option to install a new staircase which may prove to be simpler / more cost effective than sub floor insulation.

Edited April 15, 2023 by embra

[Dave Jones]

what about taking out the blocks in the floor and replacing them with insulated ones

 

https://www.cellecta.co.uk/product/tetris-beam-and-insulation-on-block-system/

 

your bound to have some that you cant as walls will be built but it would be a neater solution and no problems with messing with DPM.

[Iceverge]

1 hour ago, embra said:

 

 

Really interesting analysis, thank you for taking the time to prepare and share these. A great illustration that insulation detail and positioning can be more significant than improving thickness / thermal conductivity.

 

As a huge favour, would you mind running a couple more scenarios to add some data to ideas I've been toying with?

 

1) Repeat of the last scenario, albeit with a 50 mm only layer of insulation at a thermal conductivity of 0.019 W/mK (Kingspan K103)? Since I'm really constrained on stack height, I'm interested to see how the reduced thickness vs better thermal conductivity balances out. No insulation in the void.

2) As 1), but with a 120 mm layer of insulation at a thermal conductivity of 0.026 W/mK directly under the beam and block (to evaluate Q-bot application), as well as the 50 mm K103 insulation "inside" the house.

 

Most appreciated 🙏

 

Away from the laptop for a few days. 

 

At an educated guess 50mm of Kooltherm would be 0.35W/m2K. Typically it isn't worth the extra cost unless you really have to make a specific target for some regulations and you can't afford the space. 

 

Option 2 at an estimate would be about 0.22 W/m2K.

 

The thermal bridge at the corner is the killer. Some kind of insulated plasterboard on the wall would help things immensely. Unless you can get the floor down to a true 0.15W/m2K or ideally lower I wouldn't put in ASHP UFH. 

[saveasteading]

1 hour ago, Iceverge said:

The thermal bridge at the corner is the killer.

I wonder how much heat escapes from the centre of the room  in reality.  

If the UFH is kept away from the wall, and a thermal break is inserted there then the direct bridge is closed. Better still if it is on the skirting line and an insulated wall as Iceverge suggests.

 

The reality then is that the slab heat then all goes into the room, and some passes through that thermal bridge...but not a lot in the scheme of things.

[Sparrowhawk]

On 13/04/2023 at 18:23, Iceverge said:

What software are you using @Iceverge? Ubakus gives me nice theoretical numbers but this goes a whole way beyond by taking thermal bridging into account.

[Iceverge]

7 hours ago, saveasteading said:

I wonder how much heat escapes from the centre of the room  in reality.  

If the UFH is kept away from the wall, and a thermal break is inserted there then the direct bridge is closed. Better still if it is on the skirting line and an insulated wall as Iceverge suggests.

 

The reality then is that the slab heat then all goes into the room, and some passes through that thermal bridge...but not a lot in the scheme of things.

 

Interesting concept. I would love to be able to play with some software to get a real idea of this. I'll have a think of how I could manipulate Therm into doing this.  

[Iceverge]

4 hours ago, Sparrowhawk said:

What software are you using @Iceverge? Ubakus gives me nice theoretical numbers but this goes a whole way beyond by taking thermal bridging into account.

It's called Therm. It's free obviously because I'm a skinflint. It's a bit awkward to use initially but it's a good tool to get a feel for how heat flows work. I might be able to find a PDF of how to set it up if you're interested. 

[Sparrowhawk]

6 hours ago, Iceverge said:

It's called Therm. It's free obviously because I'm a skinflint. It's a bit awkward to use initially but it's a good tool to get a feel for how heat flows work. I might be able to find a PDF of how to set it up if you're interested. 

Thanks, if you can find that PDF I'd appreciate it

[Iceverge]

https://www.scribd.com/document/447244443/THERM-01-Beginning-Users

 

Heres one but its in ye olde units. 

 

http://passivehouse.nz/wp-content/uploads/2018/07/ThermInstructions.pdf

 

This is the one I used.