Asphalt-colored water is dripping through the wood deck below.
Can a waterproofing membrane leak if it has no holes? Unfortunately, yes, under fairly common circumstances. Consider an asphalt-based membrane installed with little or no drainage, so that water stands on it. This may occur on a low-slope roof, or more commonly, on balcony and terrace decks with porous finishes, such as concrete or tile. If drainage is absent or impeded, water can slowly migrate directly through the membrane.
The assembly of the leaking terrace deck shown here comprised a concrete topping slab applied directly to an asphalt-based membrane over a wood structural deck. Since a drainage mat was not installed to remove water from the low-slope assembly, it would be wet much of the time. Asphalt-colored water would drip through the wood deck below, as though a waterproofing membrane was not present. Liquid water was also found between the plies of the membrane, even though the seams were sealed.
Very little is published on the subject of water seeping through intact waterproofing membranes. This column touched on the issue 25 years ago in the “Deck Drain That Doesn’t.”
“Water trapped below the walking surface can cause serious problems…a continuously wet membrane is more likely to leak than one that drains freely. Further, tar-like constituents of some bituminous membranes can boil or float up with the trapped water (the column appeared in the March 1994 issue of The Construction Specifier).”
The aesthetic detraction of oil floating up to the deck surface is caused by exactly the same mechanism that allows water to seep down through the membrane: specific gravity. If the density of the membrane constituents is less than the water above, they will both try to migrate to resolve the imbalance—heavier elements go down, while lighter ones come up. It may take a while, but gravity always wins—the unrelenting force controlling the motions of galaxies will eventually push standing water down through a less-dense membrane. Some materials like pond liners may be engineered to handle this imbalance, but they are not typically used in roofs and terraces.
Manufacturers’ literature, building codes, and industry standards universally require good drainage from membranes, but they rarely mention why. Since drainage has obvious benefits it may seem unnecessary to warn against a secondary mechanism that only comes into play if good practice is ignored. However, an understanding of this mechanism is important in failure analysis. Leaks may be suspected due to membrane installation defects or punctures rather than seepage due to faulty drainage design.
The opinions expressed in Failures are based on the authors’ experiences and do not necessarily reflect those of The Construction Specifier or CSI.
David H. Nicastro, PE, F.ASTM, is the founder of Building Diagnostics Inc. and Engineering Diagnostics. They specialize in the investigation of problems with existing buildings, designing remedies for those problems, and resolving disputes arising from them. He leads the research being performed at Building Diagnostics’ testing center, the Durability Lab, at the University of Texas in Austin. He can be reached via e-mail at email@example.com.