U-value of wood?

[richardindorset]

I'm looking to build an upstand for a roof lantern. It will be about 300mm tall.

I was going to just use joists and construct a solid 'wall' all the way around the lantern opening rather than an open trussed structure.

Then I thought about possible cold bridge issues so i started looking into the U-value of wood.

My googling found that 1 inch of ply has a U-value of 0.80.

Softwood is slightly better and hardwood worse.

So am I correct in assuming that two 47mm wide joists, side by side would have a U-value of about 0.20?

I was a bit surprised that the U-value of wood is so good. It appears to be marginally better than insulation foam???

If 25mm of wood is 0.80, then 100mm of wood should be around 0.20.

100mm of celotex is 0.21!

The steelwork that the upstand is sitting on is 100mm wide and the base of the lantern is 90mm wide so aiming for an upstand of 90mm seems reasonable.

If using solid wood (so 145x47 and 170x47 on top, doubled up all the way around) would get my U-value under 0.28, it seems the easiest way to go.

Am I missing something?

Cheers!

 

[freddiemercurystwin]

Yes wood is a pretty good insulator though I've not checked your figures, in anycase its always best practice to try and incorporate say a 25-50mm sheet of insulation around the perimeter of the upstand to minimise the bridge and maintain the insulation bubble.

 

[^woody^]

Don't confuse high insulation values with the potential for thermal bridging and condensation risk. ie achieving a certain u-value in timber does not mean it won't experience condensation.

 

[richardindorset]

Further investigation reveals I was talking ******!

The 0.62 R value for ply is US not Metric!

12mm ply is actually 0.109

With this in mind, with two skins of 25mm ply, I would need 70mm of celotex to get to below 0.28

This would take the thickness to 120mm which is a bit wider than I'd like...

 

[cjard]

If 25mm of wood is 0.80, then 100mm of wood should be around 0.20.

100mm of celotex is 0.21

Am I missing something?

Cheers!

100mm of plywood (circa 800kg per cube) has a U of 1.06. Not sure what R value (or lambda) you used for your wood but it should be 8.5 ish for the R or .12 to .14 ish for the lambda

If you want .3 (regs) you can have 25 mm ply backed with 25 mm battens to max 10% surface area, infilled wih PIR then a 50mm PIR Over the top and plasterboard. Note, no more than 9 screws per square metre before eu start having an impact

Why not use a 100mm SIP? .24. Bridging is what kills your other options

 

[^woody^]

Why would you want the u-value of the timber upstand to be significantly less than the glass which covers most of the lantern?

 

[cjard]

Why would you want the u-value of the timber upstand to be significantly less than the glass which covers most of the lantern?

It's a wall. They have a u value target under regs. OP may not be able to mitigate this elsewhere

 

[freddiemercurystwin]

Why would you want the u-value of the timber upstand to be significantly less than the glass which covers most of the lantern?

It's a wall. They have a u value target under regs. OP may not be able to mitigate this elsewhere

It's a wall in that it's vertical but in reality build a timber upstand with some suitable sized timbers, infill any voids with celotex, affix a sheet of 25 or whatever inside and outside depending upon the roof finish/rooflight/flashing etc and jobs a goodun, you're overthinking it. But it does need to be thought about/well detailed beforehand.