Don't forget Thermal Bridging

[MortarThePoint]

We all put in a lot of effort considering different insulation types and thicknesses as well as airtightness and MVHR, but thermal bridging (aka cold bridging) could get neglected or forgotten at the design stage. I had a quick look and couldn't see a thread on the general topic of thermal bridging, so thought I would start one. I am far from an expert here, but wanted to draw together some of what I have learnt and prompt other more experienced members to pitch in.

 

According to Designing Buildings Wiki :

"A thermal bridge (sometimes referred to as thermal bridging, a cold bridge or thermal bypass) describes a situation in a building where there is a direct connection between the inside and outside through one or more elements that are more thermally conductive than the rest of the building envelope.

...

Thermal bridges can be categorised as 'repeating' for example where wall ties regularly bridge the cavity, or 'non-repeating' such as a wall junction or lintel."

 

'Total fabric heat loss' is the combination of heat lost through different areas of material (sum of each area multiplied by its U-value) and thermal bridges. Thermal bridges can account for a high proportion of the total fabric heat loss (e.g. over 20%). Neglecting thermal bridges at the design stage could undermine the effort you put in on the area based fabric heat losses (e.g. using wider cavities or triple glazing).

 

non-repeating Thermal bridges are typically at interfaces and so have a linear, rather than area, nature. Consequently, they are accounted for in SAP Assessments on the basis of their PSI values, which represents how lossy they are per metre, and their total linear length. It's similar to the area based fabric heat loss which is a sum across types of area multiplied by their U-value, but it is a sum over lengths multiplied by PSI values.

 

Some thermal bridges to consider include:

• Ends of cavities
• Lintels
• Junctions (e.g. between walls and floors, eaves etc)
• Holes for pipes
• Wall ties
• Poor window frame placement

 

Ends of cavities:

 

It is standard and required practice now to use cavity closers rather than masonry closure. This makes for a huge reduction in thermal bridging around windows and doors. Different cavities closers are available with different insulating materials and performance values.

 

Lintels:

 

A standard "Steel lintel with perforated steel base plate" can have a PSI value of 0.36 W/m.K. Considering all the windows and doors in a house design shows that there is a lot of length to lintels. A single metre of such a lintel looses more heat than 2m² of 150mm cavity wall (U-value 0.17 W/m²K). The difference in U-value between a double glazed and triple glazed window might be 0.4 W/m²K. That means for a 1m x 1m window, more heat is lost through a standard lintel than is saved by using a triple glazed window over a double glazed window (1.3m*0.36W/m.K > 0.4W/m²K*1m*1m).

 

Thermal break lintels can reduce this PSI figure by more than a factor of 5 to under 0.06 W/m.K. In the example of the 1m x 1m window that is an equivalent saving to using triple glazing over double glazing. Obviously triple glazing and a thermal break lintel would give a higher saving still.

 

Wall ties:

 

I can't find figures for these, please contribute if you have some, so here is a rough calculation. Wall ties are typically made of stainless steel and might have a cross section area of around 6mm². At a density of 2.5 ties per m2 they account for a very small proportion of the area (~6ppm), but have a much higher thermal conductivity that the wall insulation (e.g. ~500 times higher, 17 vs 0.032 W/mK). Consequently, their thermal effect can approach 1% (6ppm*500=0.3%). U-value calculations often ignore corrections that amount to less than 3% of the uncorrected U-value of an element (allowed by BS EN ISO 6946).

 

Useful links:

http://www.zerocarbonhub.org/resources/reports/thermal-bridging-guide

[Adam2]

Thanks, interesting reading. Thinking about this at the moment for my ICF build which helps in some ways but when it comes to doors/windows/roof and internal walls built off the (non-passive) slab I want to be sure we're doing as much as we can to maintain air-tightness and minimise the thermal bridging. Will be asking for details from suppliers soon to see what they propose and likely be asking questions on how to improve.

[MortarThePoint]

I can't correct it above, but made a mistake. If someone else has the power to correct it great, otherwise it's here.

 

Wall ties:

 

I can't find figures for these, please contribute if you have some, so here is a rough calculation. Wall ties are typically made of stainless steel and might have a cross section area of around 6mm². At a density of 2.5 ties per m2 they account for a very small proportion of the area (~15ppm), but have a much higher thermal conductivity that the wall insulation (e.g. ~500 times higher, 17 vs 0.032 W/mK). Consequently, their thermal effect can approach 1% (15ppm*500=0.8%). U-value calculations often ignore corrections that amount to less than 3% of the uncorrected U-value of an element (allowed by BS EN ISO 6946).

[MortarThePoint]

Mortar:

 

Another example of 'repeating' thermal bridging is the mortar joints in block or brick walls. Whilst you might be using thermally efficient blocks for your walls, the mortar needs to be considered as well. In a standard blockwork wall the mortar accounts for 6.6% of the area of the wall. A good thermal block may have a thermal conductivity of 0.11 W/m.K, but mortar has a thermal conductivity of around 0.72 W/m.K. Forming a simplistic average, the mortar increases the thermal conductivity of the blockwork leaf  to 0.15 W/m.K (0.11*93.4% + 0.72*6.6%), a 37% increase. Even for a less thermally efficient block of say 0.3 W/m.K, the mortar will increase the overall thermal conductivity of the blockwork leaf by almost 10%.

 

Thermally insulating mortars are available that claim thermal conductivities five times lower than standard. That would effectively eliminate the effect of the mortar on the thermal conductivity of a wall using good thermal blocks (0.11 W/m.K).

 

As most of the performance of a cavity wall comes from the insulation, the overall effect of the mortar is likely to only affect the U-value of the wall by 2-10%. For example a 100mm cavity filled with 0.032 W/m.K insulation with good thermal blocks (0.11 W/m.K) either side would have a U-value of 0.19 W/m²K, if it wasn't for the mortar thermal bridging. With standard mortar, this would be raised by 10% to 0.21 W/m²K. A similarly constructed wall with a 200mm cavity would have it's U-value raised by 6%.

 

I'm using a relatively simplistic approach here, if someone has more accurate figures or experience of using thermally insulating mortars then please comment.