Have you been wondering how U values for windows are calculated, and why different suppliers and different countries seem to use different numbers?
Let's start by looking at the basics, as getting a handle on the fundamentals is critical.
What is a U value?
According to wers.net, a “U-value measures how well a product prevents heat from escaping. It is a measure of the rate of non solar heat loss or gain through a material or assembly. The lower the U-value, the greater a window's resistance to heat flow and the better its insulating value.”
Also, a U value is the opposite of an R value. The R value measures thermal resistance of an element as opposed to its conductivity. Hence, an element’s R value is simply 1/U, and conversely, U = 1/R. It therefore follows that higher R values are better thermal performers.
That’s a bit confusing in Australia, because we use U values for windows, and R values for walls and roofs and floors (if you are including insulation in the floor, which in the colder parts of our country, you really should be doing).
In Europe, however, they use U values to describe walls as well as windows.
Furthermore, a U value is different from an SHGC value. The SHGC (solar heat gain coefficient), or g value measures solar heat gain and is expressed as a percentage. The higher the percentage, the higher the heat gain through the element.
In hot parts of the world, you don’t want a high SHGC – you want a lower one. But you also want an effective (lower) U value to stop the heat transferring into the building (and to stop the cool temperatures moving out of the building).
A U value must be calculated or assessed for both the window frame as well as the glazing itself. Because a frame is made of different material from glass, two U values must be determined. The frame's U value is often referred to as the Uf value.
The glazing-related value, called the Ug value, relates to the glazing unit part of the window, which includes the glass itself as well as the gaps between the panes (if double or triple glazed).
Therefore the goal is to combine the two separate U values (the Uf and the Ug) to determine the U value for the entire window as a whole (or the Uw value).
Frames typically provide higher U values - or worse performance - than glazing, particularly if you are looking at double or triple glazing. That means that if you have a small window to calculate, with a Ug of 0.6, the small window will come out with a worse Uw rating (higher U value) than a large window simply because the small window has more frame in proportion to the glass.
This is an important point to note because it makes one aware that there is no such thing as a standard U value for a whole window setup (Uw) regardless of its actual size. You will always need to calculated a Uw value for the exact size and shape of the window unit as the proportions between frame (Uf) and glass (Ug) will always change depending on the window’s size and configuration.
If you are ordering 10 windows of different sizes, you may end up with 10 different Uw values for your project, even though the exact same glazing and exact same window frame profiles are used throughout.
The next confusing issue is that of measurement units themselves: imperial vs metric. Australia and Europe follow the metric system, so the U value is described as watts per metre squared multiplied by degrees Kelvin (W/m²K).
In reality, most European windows come with lower U values (higher performance) than Australian windows because European windows are constructed differently, with different materials.
Imperial measurements are utilised in the US, which causes a lot of confusion with the rest of the world. Indeed, one green building journalist in the US wrote about our Superpod certified passive house, saying that he could not understand how our building was high performing, because he thought that the insulation was so thin. The issue was that European U values are not the same as US values and he had failed to make the proper conversions.
A detailed review of the US vs EU standards is found in the paper International Window Standards, Final Report/April 2014 for the Homeowner Protection Office, branch of BC Housing, by RDH Building Engineering Ltd in Canada. The paper says that “there is no straightforward way to compare North American and European product performance.” It reviews in great detail the differences between window rating systems in North America, Europe, and for passive house.
The mathematics behind U values can get quite complicated. Suffice it to say that there is a rule of thumb for converting metric and imperial U values, and that is to divide the metric U value by 5.6783, in order to get the imperial equivalent. This means that a U value of 1 in the EU will be (roughly) equivalent to a U value of 0.174 in the US - assuming the same testing methodology for that window.
Let’s now consider that last point regarding testing methodologies. There is a voluntary window organisation in Australia called WERS: the Window Energy Rating Scheme. It is run by the Australian Window Association (AWA). If you are a member of the AWA, you can participate in WERS, and you can see WERS ratings for different window companies specified on their website.
The WERS website says members must obtain energy ratings for their products from a rating organisation that is accredited by the Australian Fenestration Rating Council (AFRC), which is an incorporated entity and not a government body. The AFRC states on its website that it “will develop, administer and approve the only uniform, independent, comparative rating and labelling system for the energy performance of windows.”
In a similar way, European made windows must be tested against European standards. For a certified passive house, your windows must be accompanied with a certificate to the EN Standard. EN Standards are documents that have been ratified by a European Standardisation Organisation for the purposes of EU Regulation. There are also DIN standards in play throughout Germany. There are literally thousands of EN standards for things as varied as cork floor testing and electromagnetic compatability emissions standards for industrial environments. The US has its own standards and compliance testing facilities.
As noted above, most Australian windows have higher U values than European windows because of the different materials used. This in turn is partly due to more stringent legal requirements. Germany, for example, has mandated low U values for some time (generally under Uw 2).
In contrast, according to WERS, "U-value ratings generally fall between 2.0-10.0 W/m2.K for Australian products."
But there’s another thing: if a window company says that all their windows have Uw values equal to 3, they are not providing accurate or practically useful information. There would be different Uw values depending on the actual size and configuration of each and every window.
Furthermore, if you want your house to be certified as a passive house, the detailed and structured building physics modelling program (the Passive House Planning Package) requires proper EN standard compliance regarding U value reporting, and checks the sizes of each window carefully. Consequently, the passive house program requires the specific Ug and Uf values to be separately nominated. If you really want to model the heat loss of any window, you need to calculate the heat loss through that size of glass as well as the size and shape of the frame, to consider the whole effect of that particular window.
Also, U values say nothing about the installation methods (which the Passive House Standard also requires to be nominated and verified), air leakage, quality of production, or tolerances. How many of us have watched those "Grand Designs" episodes where the windows are 10 millimetres out? Your window should fit. In Australia, our building code tolerances may allow for things that are just not going to cut it if you want the building (and the windows) to actually perform.
So there is much to watch out for. If we understand what is being reported when numbers, ratios or percentages are being bandied around, we will know better what is being assumed, and what is being accurately calculated.