Passivhaus Training – Windows, Bridges and Air Changes - ECD Architects - Passivhaus Training – Windows, Bridges and Air Changes

It’s a common misconception in the UK that achieving the Passivhaus standard is simply a matter of matching a few increasingly familiar yet often misunderstood principles. The usual culprits are “U-values of 0.15 W/m2K, triple-glazed windows, an airtightness of 1.0, heat recovery ventilation . . .” and so on. “Not too difficult . . .” someone only familiar with the Code might think. Well, think again. Above all, the CEPHD course teaches just how much more sophisticated is the interplay of Passivhaus design and its components, and how success lies in the attention to detail that verification via PHPP demands.

In the first part of the course we focussed on the need to balance a dwelling’s energy flows – this is true with any building of course but in a Passivhaus the input energy is limited to solar gains, minimal internal heat gains and top-up heat from the ventilation system (or its equivalent from another heat source). It’s therefore vital that designers have an accurate prediction of the fabric’s heat loss so a more thorough investigation is demanded. One of the complaints about so-called low energy houses is that they haven’t delivered the performance predicted of them  - indeed, this design/reality gap was one of the reasons Passivhaus was developed. One reason for this disparity is that the impact of thermal bridging is often grossly underestimated in conventional building models (figures of 30-50% have been suggested as the shortfall) so a good understanding of techniques to avoid them form a key part of the course.  Thermal bridging is the phenomenon whereby heat flow out of a building is accelerated by penetration of the fabric, say from structure, party walls or cantilevered balconies. We talk about the Ψ(psi) value indicating the linear thermal bridge loss coefficient and for Passivhaus these have to be shown to be either insignificant (ie less than 0.01 W/mK) or have to be individually added to the heat loss total. SAP 2009 software now allows for linear thermal bridges to be assessed but in practice I suspect this will be hardly used - instead the generic ‘Y’ value will be applied as is current practice. ‘Y’ values add a percentage to the overall area-based heat losses to account for thermal bridging and designers will have to commit to following both Enhanced and Approved Construction Details – one look at these through CEPHD-trained eyes and you can see that they fall well short of Passivhaus standards though.  One of the significant differences from SAP is that PHPP measures the external surface area of the thermal envelope, not the internal surfaces as SAP does. At a stroke this overcompensates for the geometric thermal bridges, encouraging the designer to properly account for them as, if designed correctly, they may have negative values which would reduce the calculated fabric heat losses. And accounting for the heat loss from features such as balcony supports makes you really consider the detail from a new perspective. Without this level of consideration to thermal bridge losses, you really aren’t designing to the Passivhaus standard whatever claims for super low external wall U-values you might make.

As the coursework moved on to window design, it underlined that there is a world of difference between specifying ‘triple-glazed windows’ and their PHI-certified equivalents. Windows are a crucial component in a Passivhaus, typically accounting for about 43% of the total heat gains but as much as 50% of the fabric losses. The ambition is to achieve a net positive energy balance ie over the course of a year more heat energy enters the windows than escapes. Real window U-values are derived by factoring in the separate thermal conductivities of the glazed area, the frame area and the Ψ-value impact of the junction between glass and frame. How many UK window manufacturers could readily supply that data breakdown? Continuous warm edge spacers are essential to minimise bridging losses and the overall Uw-value can be no worse than 0.8 W/m²K in order to eliminate cold radiant surfaces and any associated down draughts, referencing back to those all-important thermal comfort criteria. Even then, the effect of the method of installation has to be accounted for, with an additional Ψ-value applied to every window before the final heat loss figure is derived. Contrast this with the basic way fenestration is accounted for in SAP!  In a Passivhaus, the windows effectively constitute the building’s primary heating system so great attention has to be paid to the size and orientation of the window as well as the g-value, or solar transmittance, of the glass. Too little south-facing glazing and the shortfall in heat demand cannot be met by the ventilation system. Too much and the building is at risk of overheating. We measured the impact of reveal depths on the solar gains due to their shading on the glazed area - even the impact of dirt on the window is factored in.  All this has a profound effect on the way you perceive windows henceforth. For example, it becomes clear that rooflights are hugely problematic and, frankly, best avoided. Think about it - they are always positioned way outside the thermal insulation and are inclined to maximise heat loss on cold days and solar gains in summer! There is one window feature that doesn’t change though, despite what detractors would have you believe – you can, and must, be able to open the windows in a Passivhaus!

Of all the criteria, the one that seems to have most concerned the designers and builders of the first UK Passivhaus projects has been achieving the requisite airtightness of 0.6 n50. That’s measured in internal air volume changes per hour, by the way, not the UK metric which tests the average leakiness of the envelope. Airtight construction is sadly still a bit of a new concept in our industry and the Building Regulation standards in this country are derisory. Passivhaus design seeks very low air leakage for a number of reasons, not least of which is the associated heat loss. Playing with different n50 performance in PHPP shows a dramatic increase in heat demand as more preciously-warmed air leaks through the envelope. Other associated problems from air leakage are less to do with energy losses and more related to fabric damage risk and thermal comfort. The comfort problem is straightforward to understand – greater air leakage equals internal draughts which in turn increases discomfort. German research suggests avoiding air speeds greater than 0.1 m/s reduces the acceptable air temperature by 2˚C. The other risk is all about moisture penetration into the fabric, another poorly understood science in UK construction. The narrowest of slits in the air barrier can lead to enormous increases in moisture transfer from the internal air volume. If this moisture then meets a cold surface along the way, you have a serious risk of substantial interstitial condensation, so Passivhaus design takes this very seriously. Low air leakage is tricky to achieve as the CEPHD course visit to one of the exemplar dwellings on the BRE’s Innovation Park pencils revealed. Under depressurisation from a blower door, smoke pencils revealed that external air was pouring in from the most insignificant of gaps, even the entrance door keyhole. Yet, with care, we know it can be done. The Welsh and English Passivhaus pioneers all achieved excellent results, well below the 0.6 n50 threshold and often with the contractor attempting the standard for the first time. But we desperately need more expertise on the techniques involved in the UK – a demonstration in membrane taping at the recent AECB conference by Ecological Building Systems was mesmerising for the skill and tradecraft involved. Time for a new specialism on site?

So accurate accounting for thermal bridging, real window U-values and properly rated airtightness are all essential PHPP inputs and are therefore dealt with thoroughly on the CEPHD course. It is this attention to detail in the fabric design that makes a Passivhaus so much more than the sum of a few headline principles. More in the final part . . .