By Brad Johnson, MBCI
As an architect, you’re required to design a building’s wall to meet the code-required R-value (or U-factor) in the International Energy Conservation Code. So you design the wall and add up the manufacturer-stated R-values of the components. Done, right? That method only makes sense if walls have no joints, seams, windows, or doors! Let’s think about this.
Accounting For Thermal Discontinuities
The manufacturer-stated R-value of an insulated metal panel (IMP) should really be the R-value in the center portion of the panel, if the manufacturer uses terminology consistent with ASHRAE 90.1. However, a wall is made up of many IMPs, and there are joints between the IMPs. We’ve all seen the infrared photos showing the heat loss at joints between panelized anything - plywood, insulation boards…and IMPs. The joints between each and every IMP are thermal discontinuities, commonly called thermal bridges. These are locations where the R-value is not what you read in the manufacturer’s literature. There are also metal clips and attachments that reduce the R-value of the IMP wall system. If you’re designing a wall system, don’t specify the R-value of the panel and assume it is the R-value of the wall system!
Calculating The R-Value Of A Complete IMP System
A building owner deserves a wall that meets or exceeds the code-required minimum R-value or U-factor. The mechanical engineer needs to properly size the building’s mechanical systems based on the "real" characteristics of the building envelope.
Let’s put some numbers behind this idea. Let’s consider a 42 inch-wide panel, 2 inches thick, with a stated R-value of 12. The outer surface of the panel is close to the exterior temperature - say 30 degrees. The metal wraps through the joint, decreasing the temperature of a portion of the metal on the backside of the panel everywhere there is a joint. Clearly this reduces the overall R-value of the IMP as a system. Let’s estimate that the thermal bridging effect of the joints reduces the R-value 5 inches along the edges of the panels to an R-6. That means 30 inches of the panel has an R-12, and 10 inches of the panel has an R-6. That calculates to an average R-value of 10.5 for the panel overall, which is more than a 12% loss of R-value. This is why blindly using the famous equation of R=1/U is dangerous. That equation is only true if the R-value and U-factor involved are consistent with how thermal bridging is or isn’t represented.
U-Factor Testing For Higher Accuracy
It’s clear that the panel joints are thermal bridges, but the extent of loss is really an educated guess. But there is a solution! The forward-thinking IMP manufacturers are performing U-factor testing and finite element modeling, and that includes joints between panels. The U-factor testing is a more accurate determination of thermal resistance.
As an architect designing the wall system, if you use stated R-values, recognize that you’ll need to account for the loss of R-value because of the joints. Or, simply specify panels whose manufacturers are determining the U-factor for their IMPs!
This posting is one of many by Brad Johnson and other MBCI experts, covering a wide range of topics, that can be accessed on the MBCI blog: Click Here.