DWM-logo.gif (6532 bytes)

October  2004


Enhancing Window Performance
Canadian Research Shows How Improved 
Spacer Bar Design Can Do Just That
by A.H. Elmahdy

Poor thermal performance of insulating glass (IG) units increases heat loss through windows and causes condensation that can lead to mold growth and deterioration of windows and wall sections, especially in cold climates. Typically, aluminum or steel spacer bars separate the lites in an IG unit. The high thermal conductivity of these spacer bars results in high heat loss through the bars and the surrounding area of the window (known as the edge-of-glass region) and increases the potential for condensation. The edge-of-glass region is the perimeter of glass about 60 mm (2 ˝ inches between the edge of the frame or sash and the point where the glass surface temperature is the same as that of the center of the glass. IG units with higher glass surface temperatures in the edge-of-glass region are less prone to condensation. 

The temperature in the edge-of-glass region is a function of factors such as spacer bar design and frame material. The overall thermal performance of a window is determined by the spacer bar type, the glazing and the frame (edge material for fixed windows) or sash (edge material for operable windows). So, while a window with a high temperature in the edge-of-glass area will perform well in terms of condensation resistance, it will not necessarily show good overall thermal performance.

Those spacer bars that conduct less heat than conventional metal spacer bars in the edge-of-glass region employ warm-edge technology (WET). WET spacer bars have either a thermal break in the spacer assembly or are constructed of materials with low thermal conductivity. WET reduces the potential for condensation by reducing the heat flow from the warm side to the cold side of the glazing. This is especially true of high-performance IG units with low-emissivity coatings on the glass and heavy gases, such as argon or krypton, in the sealed cavity for improving insulating value. 

A number of innovative spacer bars have been developed in the past decade, but there was no procedure for comparing their thermal performance. To provide a basis for comparison, researchers at the Institute for Research in Construc-tion (IRC), National Research Council of Canada, carried out studies to assess the performance of these new spacer bars. The IRC study assessed the performance of ten types of spacer bars (nine WET and one conventional) in IG units without window frames (unmounted) and with window frames (mounted) of various materials.

IG Units without Window Frames 
Each of the ten IG units  was fabricated by a different manufacturer. All units were nominally 6 by 48 inches, air-filled and made of clear glass. The units were tested for heat loss, with surface temperature readings taken at the mid-height of the glass.

It can be seen that unit IG8 had the highest glass surface temperature at the edge-of-glass region, making it the best in terms of reducing condensation. IG7 had the lowest glass surface temperature in the edge-of-glass region. Although the temperature difference between the best- and poorest-performing IG units was only about 11°F in the edge-of-glass region, it could have a onsiderable effect on condensation resistance. Figure 2 also shows that the further away from the edge-of-glass region (from the effect of the spacer bars), the more similar the glass surface temperatures, for all ten IG units. 
IG Units Installed in Window Frames 
The ten types of spacer bar IG units were then tested as part of a complete fixed window assembly by assessing temperatures at different horizontal planes on the warm side of the window. Four different types of frame specimen were used: FR1 (redwood), FR2 (vinyl), FR3 (thermally broken aluminum) and FR4 (foam-filled fiberglass).

WET spacer bars on the temperature measured at 3/8 inches from the sight line for each type of IG unit and frame specimen when exposed to a temperature difference of 70°F. The combination of FR4 and highly insulated spacer bar IG8 offered the warmest glass-surface temperature at the 3/8-inch plane (and at almost all the horizontal planes in the edge-of-glass region). Conversely, the combination of FR3 and hybrid spacer bar IG6 produced the lowest glass-surface temperature in the edge-of-glass region. 

The frame specimens used in these tests were not intended to be representative of commercially available frames, but rather to determine whether the trend in spacer bar performance was affected by the type of frame. The tests confirmed that the trend in spacer bar performance was the same regardless of the type of frame used. 
R-Value Performance

The overall R-value is dependent on the type of spacer bar, frame material, glazing and particularly the thermal properties of the frame material. However, even with non-conductive frame materials, poor design may reduce the thermal performance of the window.

In general, wood has a higher thermal resistance than the other materials studied and therefore the FR1 specimens performed best in terms of overall R-value. In fact, for all the specimens tested, the FR1 specimens had the best overall R-value regardless of which spacer bar was used– the only exceptions were spacer bars IG4 and IG6, where the R-values for FR2 specimens and FR1 specimens were about the same. 
When used with the poorest-performing (with respect to R-value) WET spacer bar (IG6), the FR1 specimen had an R-value only slightly lower than when it was combined with the best performing WET spacer bar (IG8). 

Concluding Notes 
The IRC research shows that commercially available warm-edge technology spacer bars can have a beneficial effect on both the edge-of-glass temperature (condensation reduction) and the overall R-value (capability to reduce heat loss). Manufacturers can use the results as a benchmark for choosing suitable combinations of spacer bars and frame materials to further enhance performance.

While the research conducted on these ten spacer bar designs provides a performance comparison, it must be stressed that the results reported are specific to the spacer bars and frame specimens tested and cannot be extrapolated to other window configurations without testing.

© Copyright Key Communications Inc. All rights reserved. No reproduction of any type without expressed written permission.