NCC 2019: Spandrels are in, but lets keep it simple!

Over the last couple of years, a fascinating undercurrent has become more widely acknowledged within the fa?ade consulting industry. Slowly but surely, whether through an act of concern over under representing fa?ade performance design or a sensitivity to legal liability or maybe the correct sizing of mechanical services, this is the first time I have seen the industry push beyond wobbly and weak NCC Section J minimum compliance requirements.

No, this article is not about combustible cladding, although I have been bemused by its inclusion as a reference in the proposed Section J 2019 changes. I am referring to an absolute oversight in Section J until now…. Spandrel Performance.

To avoid an extensive write up on the subject, which I have already blogged on some time ago, the crux of the concern has been the significant under representation of thermal bridging at the spandrel. While we consider this bridging at the vision area of a window system, such as that below, this also occurs at the spandrel.

Simply put, our current ‘wall’ performance aspiration (in Victoria) of R 2.8 is blown out of the water when considering real world spandrel performance values, with the true design Total R-value (including frame) somewhere between R 0.5 and R 1.0 (depending on frame and insulation quality). Put even more simply, it is common today to ignore up to 65% of additional heat transfer related to the wall area, if it is a spandrel.

Flick back through the pages of NCC Section J 2016, and you won’t find reference to spandrels. But a quick search through the NCC 2019 proposed changes confirms things are about to change and these low performance heat suckers are about to become part of the design and construction process.

Defined as ‘the combination of wall and glazing components comprising the envelope of a building’, the inclusion of real world spandrel performance is a great move, a brave move even, one that needs support and is fundamental to bridging the gap between design aspirations and operational energy consumption.

Brave, as it cannot have been an easy move for the powers that be to accept that we have been overestimating wall performance so significantly that the majority of our curtainwall dominated offices, hotels, hospitals, etc, have not been designed correctly. But you have to wonder…if spandrel walls have been consistently underestimated to the tune of 65%, why has it not been more noticeable? Could the reason be that we oversize our mechanical services so significantly that a 65% difference in wall performance matters little in terms of heating or cooling of our buildings?

Touching on the detail, the proposed changes offer up two methods for identifying real world values, strangely in two units, System R-values and System U-value. First up ABCB, life is confusing enough, let’s stick with one unit of measurement.

The first method provides tables of indicative spandrels. While this a nice, quick and easy approach, it needs a lot more work to be useful. I assume the intent is to match your proposed spandrel design against one of the tables with your nominated insulation R-value in order to get a ‘framed’ Total R-value. However definitions are short and options are limited, so even a cursory glance at this approach generates numerous questions.

What are the definitions of thermally unbroken (bridged) or thermally broken (unbridged) frames? Today’s industry frame performance are highly variable, with frame U-values ranging from super systems (3.5 W/m2.K) to bracketed frames that support shading (30 W/m2.K +), so further definition is needed. What is double-glazed opaque or clear face? Are 50 mm air gaps un-ventilated or ventilated, reflective or not? Outside of the very limited tabled options, so I need to jump to method 2? To be useful and to avoid manipulation, we really need something as simple but more robust! 

As for the second method, for all intent and purpose, this is very similar to how we currently calculate glazing system performance. However, do we really want to add in further requirements for modelling given that spandrels show such a low level of variation as you increase insulation levels given they are limited by their typically high ratio of poor performing frame areas? With the additional increase in modelling requirements resulting from the drive for Performance Solutions and JV3 modelling in the case of energy efficiency, we are at risk of committing more of our budgets to modelling throughout the design process with little additional value.

In summary, I am really pleased to see this push. It cannot have been the easiest of these additions to have been considered and would no doubt have seen some challenges along the way. There are clear opportunities to make the inclusion of spandrels into minimum performance less of an arduous task and perhaps avoiding unnecessary modelling along the way. Get those comments into the ABCB by the end of next week, or we will all be scratching our heads once the dust settles in 2020!