April 17, 2019
by John Karras, PE, and Thomas F. Chmill
Imagine this nightmare scenario for a design professional. Just after the construction of a sleek rooftop amenity space, the building management reports widespread leaks into the structure’s upper floor, where multimillion-dollar penthouse residential units are about to be occupied. Then the question arises: Did the project design drawings and specifications sufficiently address potential amenity roofing challenges?
Rooftop amenity spaces have become common selling points of luxurious multifamily, office, and hotel buildings in competitive real estate markets. These inviting spaces are frequently included in new construction, but are also gaining popularity during retrofits of existing roof spaces. Certain amenities, such as water features, swimming pools, playgrounds, outdoor kitchens, and elaborate landscaping, can add value to a building from a variety of perspectives, including the addition of usable square footage and generation of premium lease rates.
These amenities, however, carry many inherent waterproofing and associated drainage complexities that significantly exceed those encountered in conventional low-slope roofing design and construction. Without a coordinated design among all relevant parties, coupled with carefully sequenced construction operations, rooftop amenity spaces can plague project stakeholders and property managers with leakage and water management problems that are formidable to address once the building is in service. At times, these spaces can inadvertently commit the building owner to a waterproofing/drainage maintenance program requiring unintended confined space procedures, or worse yet, demolition, to resolve.
Roof design coordination
Coordination among design team members and consultants is not a new concept, so what makes amenity roofs such a unique challenge? In part, the endless combinations of materials and features specified for rooftop amenity spaces are limited only by the imagination of the designer and the project budget. Moreover, the coordination process involves multiple design professionals (architect, structural, plumbing, and civil engineers, and landscape architect) and is complicated by key amenity features (e.g. pool and fountain systems) that are traditionally deferred to be executed as delegated designs by supplying contractors. In the authors’ experience, coordination could be managed by systematically recognizing the challenges and risks posed by specific amenities and their associated design details.
Some overbuilt rooftop amenities, such as pools or sun decks, require a walking surface elevated above the structural roof deck and waterproofing assembly, resulting in an interstitial space below the raised deck (Figure 1). The raised deck is supported by structural elements, such as concrete or masonry knee walls (often referred to as ‘stem walls’), that subdivide the interstitial space into discrete roof zones (Figure 2).
In the authors’ experience, the primary building waterproofing is usually located at the structural deck elevation, and the zoned nature of the resulting roof presents complications to drainage design. On a traditional low-slope roof where the membrane surface is unobstructed, the square footage of a particular roof area drives the size and quantity of roof drains. However, the drainage design for an interstitial space subdivided into zones should be developed in lockstep with the layout of the stem walls. The design team should also hold a coordination session between the structural and plumbing engineers to communicate viable drain locations relative to the stem wall layout. During this process, one must endeavor to locate a dedicated internal drain in each zone. It is important to note, continuous stem walls completely enclosing each zone will not only warrant numerous dedicated drains, but may also present unforeseen consequences, including limited ventilation and service access to the interstitial space.
Often, project teams find it is infeasible or impractical to provide dedicated internal drains in each zone. Additionally, as the plumbing and structural engineers coordinate the design of overflow drainage, the waterproofing and stem wall design must incorporate alternative paths for water flow between zones (e.g. discontinuities in stem walls or ‘block-outs’ at the base of the stem walls). Block-outs are a reasonable approach to communicate water between zones, but only when the waterproofing design strategy places the membrane continuously below the stem walls. This is because providing competent four-sided waterproof flashing through the block-out is significantly more workmanship sensitive and likely less reliable in service, if the waterproofing instead extends up and over the stem wall. Placing the waterproofing membrane continuously below the stem walls also has construction schedule benefits (faster drying-in of the structural deck by the waterproofing contractor), but the design team must confirm the membrane is suitable for this purpose, durable, and has a predictable service life comparable to the materials installed above it. Further, each penetration of the vertical reinforcing bars engaging the stem wall to the structural concrete deck requires flashing. However, the flashing operation of round bars is straightforward and it benefits in service from protection by its encasement in cast-in-place concrete. With that said, it is important to coordinate the location of block-outs so they do not coincide with the reinforcing steel.
If block-outs or stem wall discontinuities are incorporated, in the interest of limiting the tortured nature of the path between water entry into the interstitial space and the nearest drain (as well as the associated higher risk of water leakage), it is advisable to coordinate the block-out locations and slope of the waterproofing membrane so water never has to travel through more than one block-out on its way to an internal drain. Additionally, the block-outs should be sized and constructed with materials minimizing the likelihood of clogging or impeding water flow in service. In the authors’ experience, approximately 203 mm (8 in.) tall x 203 mm (8 in.) wide block-outs, formed by rigid insulation boards that are removed after curing of the concrete stem wall outperform small polyvinyl chloride (PVC) pipes that are sometimes cast into the bottom of stem walls with the intent of serving as block-outs. Aside from the undesirable clogging tendency of the small pipes, the round section of the pipe result in water ponding on the upslope side of the pipe (Figure 3).
Once the drainage design at the waterproofing level coalesces, or in parallel with this process, coordination with the raised deck drainage must take place. The nature of the raised deck drainage system will initially be driven by the specification of the wearing surface (e.g. open-jointed deck systems versus hardscape). If the raised deck requires unit drains, the plumbing engineer will select either a dedicated downleader system (i.e. independent of the waterproofing-level drains in the interstitial space) or specify bi-level drains depositing raised deck drainage into the drains at the underlying structural deck elevation (Figure 4). There are multiple viable methods to address both raised deck and structural deck drainage. However, it is critical to coordinate the specified drain type (e.g. promenade for hardscape, bi-level for split-slab, and clamping ring when a waterproofing membrane must integrate with the drain) with its adjacent assembly of waterproofing and/or overburden materials.
Amenity feature utilities
Interstitial spaces provide designers with a seemingly convenient location to discretely route utilities servicing rooftop amenities. Some examples are piping for recirculation of water and landscaping irrigation, natural gas conduit for fire pits or barbeque grills, and electrical conduit for light fixtures. If design for the support, type, and routing of these utilities is not coordinated with the interstitial space waterproofing or considerate of facility maintenance needs, and haphazardly installed with an ‘out-of-sight, out-of-mind’ mentality, the resulting waterproofing vulnerabilities may be difficult or impossible to repair without relocation of utility services or even demolition.
The support of piping and conduit in the interstitial space is often achieved with hangers suspended from the raised deck, or brackets attached to the stem walls or bearing on the structural deck. Depending on the configuration of the waterproofing membrane in the interstitial space (e.g. continuous on structural deck versus up and over stem walls), the various methods to support utilities each carry performance risks, and they should be evaluated in that light. Where possible, it is best to avoid penetrating the waterproofing with utility supports altogether. At a minimum, it is best to provide several inches of clearance below and adjacent to horizontal utility lines to permit future waterproofing repairs (Figure 5), support the utility hangers with round stanchions with flush post-installed anchor heads to facilitate flashing, and always avoid attaching brackets with irregular section profiles directly through the waterproofing (Figure 6), as the fastener penetrations are blind and unable to be reliably made watertight. Additionally, wherever piping or conduit must penetrate the waterproofing, it must do so a minimum of 102 mm (4 in.) away from walls and curbs to permit proper flashing installation. Finally, to the extent permitted by plumbing and electrical regulations, it is advisable to avoid the use of flexible corrugated piping/conduit at the penetration locations, since waterproofing materials cannot predictably fill the corrugations and eliminate the numerous pinholes that are inherent to the corrugated shape during the flashing operation.
A future maintenance plan for interstitial space utilities, the waterproofing membrane, and drains should also be considered during the design phase and communicated with the building owner, since these items will require maintenance, repairs, and eventually replacement when in service. Some panelized raised deck systems (e.g. wood decking if permitted by local code) inherently permit future access by virtue of their removable planks, while other monolithic upper-deck systems (e.g. cast-in-place concrete on metal form deck) warrant more careful consideration of the quantity, size, and location of access hatches that should be included in the design. Provisions for natural ventilation (if the space above the waterproofed structural deck is unconditioned) through the interstitial space are also prudent to include in the design from the standpoint of minimizing moisture buildup and minimizing the potential need for safety engineering controls, such as forced-air ventilation during maintenance activities. Ultimately, developing a design for an interstitial space that does not later surprise the owner by necessitating a costly permit-required confined space program to perform routine maintenance below the raised amenity deck should be the goal.
Rooftop landscaping and guardrails
Incorporating landscaping in the design of rooftops is a process that, in recent years, has benefitted from increased understanding of vegetative roofing in the architecture, engineering, and construction (AEC) fields. The lack of coordination between waterproofing details and drainage with landscaping features, however, continues to haunt some project teams with many unintended problems.
One example of a risk-avoidance design opportunity relates to the specification and design detailing of a popular amenity feature—steel-walled planters. These products—comprising of vertical steel wall plates, horizontal steel base plates, and steel gusset-plate stiffening elements—require fastening into a structural substrate through the base plates. If the waterproofing membrane is located on the structural deck and the base plates are designed to bear directly on the deck, the post-applied bolt anchors securing the base plates create blind penetrations through the waterproofing that cannot be reliably sealed by simply applying excess fluid-applied waterproofing material or sealants over the bolts. Alternatively, specifying a concrete curb, either with regularly spaced discontinuities or with intentionally located block-outs (similar to the interstitial drainage/stem wall drainage discussion earlier) with continuous waterproofing below the curb is the preferred practice, as it decouples the planter bolt penetrations from the waterproofing.
Glass guardrails, commonly specified on amenity roofs, present similar risks and coordination opportunities as the steel-walled planter. Whereas traditional steel-pipe-stock guardrails can be specified with round stanchions to facilitate the detailing and reliable base flashing detailing, glass guardrails, particularly those with the desirable continuous/frameless appearance, require a continuous aluminum ‘shoe’ to capture the glass lite. Most stock glass guardrail shoes are designed to be anchored through the bottom into the substrate. Like the steel planter base plate fasteners, aluminum shoe fasteners should not penetrate the waterproofing in a blind configuration and rely on sealant. It is advisable to employ the concrete curb approach described earlier.
Traditional rooftop landscaping elements such as concrete-walled planters can also be at risk of missed coordination opportunities. The authors visited a project site where the intended drainage features at such a planter were rendered inoperable due to lack of coordination. The architectural drawings and specifications appropriately and consistently showed rooftop concrete planters lined with, from interior to exterior, a hot-rubberized asphalt waterproofing membrane, root barrier, and drainage composite. The plumbing specifications appropriately indicated a bi-level planter drain product with openings at the waterproofing membrane level. However, the planters were filled with several inches of a concrete overlay, blocking both the waterproofing-level drain opening and the vertical drainage composite installed at the bottom of the planter wall. Rainwater collecting in the planter would have only one outlet—after ponding to a depth of several inches, it would eventually reach the upper openings in the vertical standpipe of the drain, but only after creating unnecessary water leakage risk and other unintended consequences, such as compromising the health of the plantings and attracting mosquitoes. In this instance, the landscape designer—with prudent intentions—had specified the sloped concrete overlay to improve planter drainage, but the concrete overlay was never represented (or therefore reviewed) in a coordinated design detail that included the waterproofing, drainage composite, and application-specific drain.
For pedestrian comfort or the provision of Americans with Disabilities Act (ADA), rooftop amenity spaces commonly require ramps linking varying wearing surface elevations. A common design strategy for ramps on an amenity roof is for the ramp wearing surface to be cast-in-place concrete, where the concrete is placed on a sloped stay-in-place geofoam form. The perimeter of the ramp will include downturned curbs bearing on the waterproofed structural deck. Since rapid and unobstructed drainage of the waterproofing membrane reduces leakage risk and prolongs the life of the membrane, the presence of the geofoam warrants the specification of a drainage composite layer to be included between the waterproofing membrane and geofoam fill. Additionally, the footprint of the ramp must be considered as a ‘zone’ of the amenity roof interstitial space, and the drainage recommendations outlined above should be implemented in coordination with the structural and plumbing engineers.
The authors recommend the use of geofoam as a method to construct and support the wearing surface. Though not unreasonable for a ramp, it is not preferable to be employed to support large areas of the raised deck on an amenity roof. This is because, even if a drainage composite is provided between the geofoam and waterproofing and the waterproofed zones include a sufficient quantity and layout of drains, future access into the interstitial space (e.g. to repair or maintain the waterproofing) would be restricted by the presence of the foam.
Child care and playground surfaces
Child care and playground facilities are incorporated, for their functional and business opportunities, on some larger-scale amenity rooftops. With child care and playground facilities, designers often specify a poured-in-place rubber-wearing surface, not only for comfort and minimal maintenance, but also for compliance with playground safety standards such as ASTM F1292, Standard Specification for Impact Attenuation of Surfacing Materials Within the Use Zone of Playground Equipment. The use of monolithic rubber-wearing surfaces, however, requires close coordination with the overall drainage design strategy of the rooftop amenity space. The authors investigated a building with a rooftop rubber-wearing surface experiencing water leakage into the lobby below during heavy rain. The authors’ found the waterproofing at the structural deck level appropriately included drains integrated with the waterproofing, but the presence and relative impermeability of the rubber-wearing surface over the drain products precluded surface water from readily reaching the drains (Figure 7). Since the perimeter base flashing around the amenity roof area only extended 203 mm (8 in.) above the waterproofing elevation (which was approximately 15 mm [1/2 in.] above the wearing surface elevation), ponding rainwater against the perimeter wall easily overtopped the hot-rubberized asphalt base flashing. Although the base flashing would have some ability to resist hydrostatic pressure, the seams of the self-adhered membrane weather barrier above it were no match for exposure to standing water.
In this case, the provisions for water to bypass the rubber surfacing and reach the drains should have been included in the design, and the base flashing detailed to reach a reasonable height (e.g. 203 mm [8 in.]) above the wearing surface, not the structural deck. Regarding surface drainage, designers should carefully navigate the design options. Simply specifying a bi-level promenade drain, though appropriate for similar applications (e.g. split-slab construction), may be inappropriate for the child care/playground application given the ‘hard point’ (and associated risk for injury) associated with the exposed body of the drain frame and grate. One option to consider would be to slope the rubber surfacing (to the extent permitted by applicable regulations) away from areas of the playground accessible to children and below a guardrail that permanently cordons off the exposed components of the drain product to playground occupants other than maintenance personnel.
Athletic-wearing surfaces, putting greens, and pet relief areas, also popular selling points of amenity roofs, warrant similar design attention from the standpoint of providing unimpeded surface and structural deck level drainage. Additionally, these amenities warrant design attention toward protecting the waterproofing-level drain inlets of the drain product from clogging (e.g. from crushed rubber athletic surface overburden) using a filter fabric with appropriate permeability to wrap the perforated drain extension.
Coordination among design professionals is vital in the context of amenity roofs. Going forward, this coordination will become increasingly challenging, as some jurisdictions begin to adopt regulations directly applicable to amenity roofs such as stormwater retention performance and vegetated areas (already common in the authors’ market of Washington, D.C.). Such regulations will cause design professionals to further broaden the scope and duration of coordination exercises, especially since the amenity features described above are conventionally procured with delegated design (i.e. very late in the design phase if not during the construction phase) and will have disparate design development schedules with amenities like stormwater/green area ratio, that require finalizing of design prior to project permitting. Additionally, hydraulic-related regulations such as stormwater retention are bringing new parties like the civil engineer to the rooftop amenity design process. Finally, intense weather events, which impart heavy demands on the complex waterproofing and drainage systems associated with amenity roofs, show no signs of abating.
John Karras, PE, is a senior project manager at the national engineering firm, Simpson Gumpertz & Heger (SGH). Karras’ expertise encompasses building envelope design, consulting, and construction phase services. Karras is well experienced in roofing/waterproofing consulting and creatively collaborating with construction teams to navigate complex field challenges. He can be reached at firstname.lastname@example.org.
Thomas F. Chmill is a senior staff member at Simpson Gumpertz & Heger (SGH). He is experienced with the investigation and remediation of existing low-sloped roofs including the design of repairs and construction oversight for a variety of owners, architects, and general contractors (GCs). He can be reached via e-mail at email@example.com.
Source URL: https://www.constructionspecifier.com/amenity-roof-design-challenges-coordinating-waterproofing-and-drainage-with-luxury-features/
Copyright ©2020 Construction Specifier unless otherwise noted.