Thermal Bridging & Derating

Download .zip Download .tar.gz


By default, TBD presumes an OpenStudio model’s opaque surfaces (maybe in the hundreds) refer to one (or very few) common multilayered construction(s) that reflect clear-field design intent. TBD subsequently derates these multilayered constructions on a surface-per-surface basis for energy simulation purposes. So while each construction has a unique clear-field Uo factor, each surface has a unique derated Ut factor (once TBD is done).

Depending on the extent of thermal bridging (due to each surface’s edge lengths and psi factors), surface-specific derating can range from barely noticeable to extensive (e.g. > 50%). One may start off with a single, common clear-field construction for all exterior wall surfaces, the latter will end up with sometimes radically different Ut factors.

Ut = Uo + ( ∑psi • L )/A + ( ∑khi • n )/A

This presumption is consistent with typical building energy simulation workflows. Yet determining what should be the initial (common) construction Uo factor may not be straightforward. What happens when designers are unsure of what initial clear-field Uo factor they should start off with, given façade layouts and thermal bridging design choices? Is it economically wise to aim for much lower Uo factors, as a means to compensate for major thermal bridging?

In some cases, building professionals may even choose (or are required) to achieve a maximum, area-weighted average Ut for all exterior wall surfaces. It’s the case for prescriptive requirements of the Canadian NECB 2017 and 2020 editions, e.g. a final wall area-weighted average Ut of 0.210 W/m2.K (R27) for climate zone 7 (NECB 2017). Depending on the extent of thermal bridging, the initial clear-field Uo factor for that single, common construction may need to be 0.160 (or much, much lower).

So in addition to derating construction Uo factors (to final surface Ut factors) for energy simulation purposes, TBD offers designers the option of first autogenerating required clear-field Uo factors (a process called uprating) to meet a given target, by reordering the above equation.

Uo = Ut - ( ∑psi • L )/A - ( ∑khi • n )/A

Uprating menu options (see Settings), are paired together for walls, roofs and/or exposed floors (let’s make things easy here by limiting the discussion to walls). The default value assigned to the “Wall construction(s) to ‘uprate’” pull-down menu option is “NONE”, disabling any uprating calculations for walls. TBD nonetheless pre-scans an OpenStudio model to retrieve referenced wall constructions in order of prevalence - referenced constructions covering a larger area are listed higher up in the pull-down list. Users can either limit uprating calculations to one (1x) such referenced wall construction, or to “ALL wall constructions” in a building model. The latter is an all-encompassing solution, overriding previously set construction assignments (the most prevalent wall construction is nonetheless retained as the basis for subsequent uprating - and then derating - calculations). Users can also set the desired, area-weighted Ut factor for selected walls (default factors are those of the NECB 2017 for climate zone 7).

TBD will log (and flash on screen if using the OpenStudio Application) the calculated clear-field Uo factor required to achieve the desired area-weighted Ut for walls (see Reporting).

An initial wall Uo of 0.162 W/m2•K is required to achieve an overall Ut of 0.210 W/m2•K for ALL wall constructions.

The uprating calculations are similar to UA’ assessments, yet in reverse order. In any UA-type exercise (like these uprating calculations), a significantly weaker component will have a disproportionate effect vs its area (as summarized in the very first paragraph of this guide, “In a nutshell …”). The degree of uprating may be quite reasonable for very efficient envelope designs, e.g.:

  • continuous outboard insulation
  • thermally-broken cladding clips
  • minimal fenestration
  • favourable façade aspect ratios

Yet it may be very challenging (and onerous) to meet such ambitious Ut targets when factoring-in weaker components, e.g.:

  • spandrels
  • poor detailing
  • lots of fenestration

We strongly recommend to first investigate this feature while relying on Apply Measures Now feedback (ideally UNCHECKing the Alter OpenStudio model option), to get a sense of how significant the uprating calculations may end up altering your initial designs. In some cases, EnergyPlus’ CTF convergence calculations may even fail with very thick (uprated) insulation layers !