Buildings Move, Buildings Leak: Revisiting the Critical Link Between Engineering Mechanics and Enclosure Performance

No v e m b e r 9 , 1 0 , 1 2 , 1 6 , 1 7 , 1 9 , 2 0 2 0 | 2 0 2 0 I I BE C B u i l d in g E nc l o s u r e S y m p o s i u m P o r t e r | 3 1
Buildings Move, Buildings Leak:
Revisiting the Critical Link
Between Engineering
Mechanics and
Enclosure Performance
Jonathan Porter, PE, AAIA
Kraus-Anderson Construction Company
501 S 8th St., Minneapolis, MN 55419
(612) 336-6420 | jon.porter@krausanderson.com
ABSTRACT
SPEAKER
Jonathan Porter, PE, AAIA
Kraus-Anderson Construction Company | Minneapolis, MN
Jon Porter is the director of building science for Kraus-Anderson Construction (KA). In
his role at KA, Porter focuses on building performance, particularly the exterior enclosure
and interior finishes. He serves as a technical resource to project teams in relation to means
and methods, constructability, building materials technology, workflow and sequencing,
proper installation techniques, and testing and acceptance protocols. A licensed professional
engineer in the state of Minnesota, Porter has more than 24 years of design and construction
experience, including roles in forensic investigation of design and construction failures and
as an owner’s representative.
An enclosure that is appropriately responsive to its environmental loading conditions is one of the most fundamental
measures of satisfactory performance for the built environment. Some loads are directly experienced by
enclosure components and materials themselves, while other loads are experienced because material deformations
or deflections influence those components and materials. While the relationship between structural movement and a
structure’s usefulness to its intended purpose has been well understood throughout the history of design and construction,
that understanding has not always translated well into satisfactory enclosure performance. Drawing on
experiences in post-construction forensic investigations, troubleshooting during construction, and efforts to influence
design detailing, this presentation will discuss key factors in applying engineering mechanics for the benefit (or detriment)
of enclosure performance. Specific aspects to be shared will include the cross-party dynamics in design and
construction that give rise to current challenges, case studies of failures as a result of insufficient consideration, and
areas for improvement across the design and construction industry.
3 2 | P o r t e r 2 0 2 0 I I BE C B u i l d in g E nc l o s u r e S y m p o s i u m | No v e m b e r 9 , 1 0 , 1 2 , 1 6 , 1 7 , 1 9 , 2 0 2 0
INTRODUCTION
Within the past 80–100 years, the practice
of understanding and applying building
construction technology has grown in
both breadth and depth. This has been
particularly true with respect to construction
materials and how they respond to
the environment they are placed in. This
growth has followed a natural progression,
where knowledge is initially leveraged at
the material level, over time is elevated to
the assembly level, and then ultimately
reaches the systems level.
This growth of understanding and
application has benefited from historical
feedback loops to facilitate continuous
improvement. The cycle of study, propose,
implement, observe, and adjust within the
building design and construction industry
has closely followed the principles of the
scientific method and the Deming Cycle of
Plan, Do, Check, Act (Figure 1).1 Evidence
of this cycle can be seen in the structural
codification of design requirements as natural
events inform future code provisions.
A more striking example of this is the evolution
of seismic design requirements that
follow a clear progression directly tied to
lessons learned by the structural engineering
community after significant earthquake
events.
Similarly, evidence can be observed in
the evolution of building enclosure performance
as past and present diagnostic efforts
inform future best practices. A clear example
is the evolution of what is considered
reasonably achievable for whole building
enclosure airtightness. Using the residential
sector as an example, while expecting
a value of five air changes per hour (ACH)
under a pressure differential of 50 Pascals
(5 ACH50) may have been considered
aggressive 15 years ago, today, a value of
less than 3 ACH50 is considered customary,
and values in the range of 0.5–1.0 ACH50
are considered completely attainable.
However, within the past 15–20 years,
other forces in the industry have at times
worked in direct opposition to the cycle of
continuous improvement for the performance
of building enclosures. Often these
forces might be exhibited by the trend of
working in “silos,” where separate parties
in the design and construction process are
not incentivized to focus beyond their own
set of responsibilities. Other times they are
characterized by “a race to the bottom,”
where competitive pursuit of projects can
motivate parties to reduce their service fees
in order to increase their chances of being
awarded a project. These forces can regularly
influence project outcomes, despite
the favorable attention that communication,
collaboration, and integration receive
when discussing improvements to project
delivery.
In project-specific situations where
there is downward pressure on professional
fees for design and technical services,
or there is a drive to increase profitability
without additional payroll expenditure,
one consequence is that technical design
and detailing might be reduced, either in
contract documents or in delegated design
submittals. If critical design considerations
or detailing are overlooked as a result,
building enclosure performance can be
negatively impacted. Other times, when
compensation for design services might be
adequate, early focus can be disproportionately
focused on schematic planning of
aesthetic-based design activities without
the important accompaniment of technical
consideration on those aesthetic decisions.
This can result in enclosure performance
issues that are more difficult to solve later
in the project. This focus can additionally
reduce available resource allocation for
technical development and detailing of the
project. Another significant force in the
past 10–12 years has been the transition
within the design industry to construction
documents being prepared by an entire
generation of workers who are proficient
exclusively in the use of Building Information
Modeling (BIM), yet might have little to no
direct exposure to the actual fit-up process
that occurs out on the jobsite to produce
Buildings Move, Buildings Leak:
Revisiting the Critical Link
Between Engineering Mechanics
and Enclosure Performance
No v e m b e r 9 , 1 0 , 1 2 , 1 6 , 1 7 , 1 9 , 2 0 2 0 | 2 0 2 0 I I BE C B u i l d in g E nc l o s u r e S y m p o s i u m P o r t e r | 3 3
Figure 1 – The Deming Cycle illustrating continuous improvement.
completed assemblies and systems. In this
scenario, it is possible for individuals generating
construction documents to assume
that the construction sequence closely
follows the way in which the building has
been modeled, and so detailing decisions
that affect constructability can be driven
more by BIM workflow knowledge than
knowledge of the actual building assembly
process. This in turn increases the chances
of constructability issues that can negatively
impact enclosure performance.
This development can compound the
tendency to focus on schematic design.
Young designers are out in the field less
often during construction activities, often
because the architects’ fees have been used
up by then. This robs young designers
of a critical aspect of their development,
resulting in a dwindling number of designers
with field experience. To understand
the impact of detailing decisions on the
built project, it is important for designers
to have the opportunity to observe the
detailed assembly of building systems
under construction.
Another negative force is the aggressive
advocacy used to market products
intended to simplify the construction process
that might have questionable levels of
demonstrated performance. When products
get a foothold in the marketplace, they
can become so ubiquitous that higherperforming
systems have a difficult time
competing. This can result in a situation
where performance and life span of individual
materials within the same assembly
are mismatched. In a related vein, pressure
to use materials with the lowest first
cost rather than considering total cost of
ownership can either reduce the durability
and long-term performance of systems or
potentially result in a higher life cycle cost
of the building.
While these negative forces are not
expressly intended to work for the detriment
of continuous improvement in
the design and construction of building
enclosures, they nevertheless increase the
likelihood of various process omissions or
poor decisions that will negatively impact
the outcome of the project, including the
enclosure.
Some challenges and areas of concern
with the building enclosure receive an
elevated level of exposure and discussion
during the design and construction process,
and consequently are less likely to be
neglected—even in the face of the opposing
forces described previously. Examples of
this include review and confirmation of
material compatibilities or field testing of
fenestration products in accordance with
AAMA 502.2 Other aspects surrounding
behavior of materials or systems within the
building enclosure are more likely to be
neglected as the project team responds to
negative pressures.
The position of this paper is that one
of the more significant functional aspects
that regularly receives under-consideration
in the design and construction of building
enclosures is structural effects on the
building enclosure itself. These structural
effects, which are a subset of the larger
category of enclosure loadings, may be
experienced by the enclosure as directly
applied loads or as secondary effects from
loadings which are not applied directly to
the enclosure elements. This underconsideration
in turn increases the likelihood
that components within the building
enclosure will either experience an
impaired level of function or fail prematurely.
This paper will describe a range of
structural effects that are commonly experienced
by building enclosures, and how
they contribute to the behavior of the
enclosure. Additionally, examples will be
shared of projects where various structural
effects were not adequately considered in
the design and construction of building
enclosures, circumstances around how
the under-consideration occurred, and
the resulting effects on the progress of the
project or the performance of the enclosure.
Finally, this paper will offer several
recommendations that project teams
should consider in order to avoid these
issues.
BASIC PRINCIPLES OF
STRUCTURAL LOADINGS
Enclosure loadings experienced by
buildings can be broadly categorized as
exterior environmental loadings, interior
environmental loadings, or loadings from
the enclosure itself.3 Within these categories,
the loadings can further be classified
as perceptual, environmental, or structural.
Perceptual loadings are primarily
physical effects that can be detected with
the human senses. For example, sound
is a perceptual loading that is a common
design consideration for enclosures that
house residential and performance art
uses. Environmental loadings are considered
physical effects that are a result of or
in response to changing interior or exterior
conditions or usage. For example, the
wear and eventual deterioration of a floor
covering due to foot and equipment traffic
across it is an environmental loading.
Interior environmental loadings, such as
those that are a result of occupant behavior
and indoor ventilation affecting the
enclosure, are critical enough to warrant
their own discussion. However, they are
In the context of structural analysis
and design, serviceability refers to
the performance of structures under
normal service loads and is concerned
with the uses and/or occupancy of
structures, including items such
as excessive deflection, slipping,
vibrations, and cracking.
3 4 | P o r t e r 2 0 2 0 I I BE C B u i l d in g E nc l o s u r e S y m p o s i u m | No v e m b e r 9 , 1 0 , 1 2 , 1 6 , 1 7 , 1 9 , 2 0 2 0
beyond the scope of this paper. Structural loadings are unique in that while they share characteristics with perceptual and environmental loadings, they also feature physical effects that require consideration for stability and the prevention of rapid and unanticipated failure for the protection or life safety of occupants. Within the discipline of structural engineering, there are two primary considerations common to most analysis and design efforts—strength and serviceability.4 Within the structural engineering profession, serviceability is a universally defined concept featured in both model codes and material-specific codes. In the context of structural analysis and design, serviceability refers to the performance of structures under normal service loads and is concerned with the uses and/or occupancy of structures, including items such as excessive deflection, slipping, vibrations, and cracking. While strength more directly focuses on enclosure stability and occupant life safety, serviceability addresses the structure’s ability to adequately perform to its intended function given its use. In general, serviceability issues are different from strength issues in that they involve the response of people or objects to the behavior of the structure under load. There are many conditions where a structure can achieve a satisfactory level of stability and life safety, while failing to enable use of the structure for its intended purpose. The Leaning Tower of Pisa and the Millennium Footbridge of London are two famous examples of this (Figure 2).5
The Leaning Tower of Pisa, due to unfavorable below-grade soil conditions, settled and shifted to its side. While it has maintained its overall stability and not experienced any abrupt material failures (strength), its floors are too pitched to be used for human occupancy (serviceability). The Millennium Footbridge of London as designed and originally built had its fundamental mode of vibration very close to that set up by natural footfall frequency of humans walking. As a result, it would generate sympathetic vibrations that rendered the bridge too bouncy for pedestrians to comfortably walk across (serviceability), even though those vibrations did not result in the bridge becoming structurally unstable or in elements becoming overstressed (strength). The footbridge was consequently reinforced to increase its stiffness such that its fundamental mode of vibration was sufficiently dissimilar to that of humans walking, and unable to resonate under pedestrian usage.
Material stresses and deformations are created by structural loadings. Stresses and deformations experienced are proportionate to the magnitude of force, or load, that is placed on the element and the physical and material properties of that element. While deformations often are mostly considered on an incremental level, they become the source of movement exhibited on a larger scale. These movement types can include deflection, displacement, shortening/elongation, and sway.
Structural loadings can be experienced as forces applied externally to elements by wind, gravity, imposed displacements, or forces developed within those elements, such as change in volume. An example of imposed displacement is excessive settlement of a column footing, which in turn generates forces in the structural frame it is supporting that otherwise would not have occurred. Common examples of gravity loading can include material self-weight and live loads. Common examples of change-in-volume loadings can include expansion or contraction due to thermal or moisture content changes. It is common for moisture content fluctuations in beams and slabs of a cast-in-place concrete parking ramp to induce shear and moment in the columns.
In addition to the means in which structural loads can be applied, the effects of structural loads can be experienced on elements either as directly or indirectly applied loads. Directly applied loads are those in which the element under consideration is directly experiencing the load and the resulting effects of that load. An example of a directly applied load is when an element experiences heat gain or loss and subsequently undergoes volume expansion or contraction. If that element is constrained by adjacent materials from changing its volume, it will experience an increase in internal compressive stress if prevented from expanding, or internal tensile stress if prevented from contracting.
Alternatively, indirectly applied loads are those in which an element is experiencing the consequences of a structural loading applied to another element. Indirectly applied loads are often converted from one load type on a secondary element to another load type on the element in focus. An example of an indirectly applied load would be a masonry control joint experiencing tension across its width due to concrete masonry shrinkage. Directly applied loads are often understood on a more rudimentary level because their application and their effect can be more easily visualized. Indirectly applied loads, while understood in concept, can often be more difficult to anticipate on an intuitive level, due to degree of separation between the load that needs to be acknowledged and the element that will ultimately be impacted by that load. This degree of separation can often result in indirectly applied loads being overlooked, and thus their effects on the component might not be fully assessed during design and construction.
As noted above, a primary consideration common to most analysis and design efforts within the structural context is serviceability. Serviceability directly speaks to the degree of movement resulting from specific loading, and whether that degree is considered acceptable for the use of the
Figure 2 – The Leaning Tower of Pisa (left), and the Millennium Footbridge of London (right).
November 9, 10, 12, 16, 17, 19, 2020 | 2020 IIBEbeC BuildINg ENClosure Symposium P Porter | 35
element, assembly, or system in focus. With structural performance, if the degree of movement exceeds the range considered acceptable, the performance of the structure is considered inadequate. An example of this is when a structural lintel supporting a masonry wall allows so much deflection that cracks form in the head mortar joints of the masonry. While the structure has not collapsed in that it has met its strength and stability obligations, it has failed to perform because it has not met its serviceability obligations.
In enclosure design and construction, performance is primarily evaluated on whether the control layers for water, air, vapor, and heat function as intended. This holds true whether the enclosure is low performing, such as a carport, or high performing, such as a surgical suite in a medical facility. The intersection of structural analysis and design with enclosure analysis and design is one where both strength and serviceability must be employed for the benefit of enclosure performance. Excessive movement imposed on elements or assemblies in building enclosures can result in unacceptable performance of the enclosure. This can be exhibited in a number of ways, including a compromised enclosure that is leaking, an element or assembly that features a reduced life span due to premature wear and tear, or an element or assembly that cannot function at a minimum threshold of continuity because it prevents an interior environment required for use of the space.
Materials used in building enclosures all have recommended service ranges that are typically based on their physical and mechanical properties. These properties are the basis-of-use criteria, which are typically provided through material standards such as ASTM, by trade organizations such as BIA, SMACNA, AAMA, or by material manufacturers. To ensure anticipated performance of materials, it is critical to design, install, and maintain them within recommended ranges. Recommended material operating ranges, coupled with structural induced movement, is what ultimately defines whether that movement will result in effective enclosure performance or impaired enclosure performance.
GAPS IN THE DESIGN AND CONSTRUCTION PROCESS
If goals are adequately defined and aligned among all project stakeholders at the inception of the design and construction process, expectations of enclosure performance can be well established. This is the premise behind establishing owners’ project requirements (OPR) and validating the basis of design for proposed enclosure systems.6 A shared understanding of the intent for enclosure performance can guide design and construction professionals engaged in the project. To meet the larger goal of the OPR, it is critical to identify and define material performance criteria such that materials are selected appropriately and those materials and assemblies operate in an intended fashion. This definition ultimately informs the performance and associated design criteria related to elements of the structure that directly or indirectly apply loads to the enclosure.
One area where the building design and construction industry has recurring challenges with appropriate definition of structural performance is in situations where the performance criteria are not adequately specified and communicated. This is where project teams expect an outcome while neglecting to describe how to achieve it with a minimum baseline of definition. This can exhibit itself in a variety of ways. The performance criteria might not be fully understood by the party specifying the criteria. The specifying party might not possess experience with the material being considered for use, or they might not understand the limits of the material. A material might be novel or new enough such that its operating ranges are not yet fully understood by the building industry. The reinforcing work needed on Frank Lloyd Wright’s Fallingwater house is a prominent example of a designer not fully understanding an acceptable range of structural performance in the original design.7 In this situation, the designer of the house did not understand the limits of reinforced concrete, resulting in not enough reinforcing being specified for the cantilevered beams. Even with the contractor surreptitiously doubling the amount of reinforcing that Wright had specified, the house required extensive structural retrofitting years later to bring it to a point of sufficient strength and serviceability.
Another way this may occur is when a performance criterion is shared, but it is too vague in its presentation to ensure that the resulting performance is appropriate. Performance specifications for enclosure materials and systems that state “…system shall withstand movements of supporting structure…” without defining those movements are a notorious example of this type of omission. A slightly modified version of the omission is when the performance specification states “…provide system capable of withstanding movement of supporting structure indicated on drawings…” but neglects to include the movement amounts in the drawings that form part of the contract documents.
Conversely, there can be circumstances where performance specifications are properly prepared and provide specific information to accommodate movement, but the supplier or installer does not faithfully reflect those requirements on their fabrication or erection drawings.
A third means of inadequate specification is when a performance criterion is defined, but the value defined is not appropriate to the situation. This can often be the result of standard, or “boilerplate,” criterion values in a performance specification that have not been modified to appropriately reflect project-specific conditions. Boilerplate values are often placeholders that might accommodate a wide range of conditions and therefore possibly might be adequate for a building structure and enclosure with a conventional layout or configuration. When complication or irregularity is introduced into a structural framing plan or an exterior enclosure, boilerplate values for movement accommodation quickly become inappropriate.
Another area where building design and construction professionals struggle with appropriate application of structural performance is in situations where there is inadequate consideration to movement behavior, independent of specifying and communicating performance. This is commonly manifested as a misunderstanding of the limits or nature of a material, or the lack of holistic thinking when multiple materials are combined. All materials have the ability to change shape, regardless of their relative stiffness and stability characteristics.
This misapplication might occur when movement behavior is not fully understood. A long-standing example where this misapplication has been overcome can be seen in exterior steel railings embedded in concrete walls or slabs. While steel and concrete have coefficients of thermal expansion similar to each other and thus can be considered to exhibit monolithic behavior when cast together as reinforced concrete, steel
36 | Porter 2020 IIBEbeC BuildINg ENClosure Symposium | November 9, 10, 12, 16, 17, 19, 2020
railings require movement joints within their assembly because their opportunity for heat gain or loss far outpaces the concrete they are embedded into below.
Additionally, this misapplication might occur when movement behavior is not fully anticipated. An example of this is when a building exterior expansion joint turns from a vertical plane to a horizontal plane and the specifier defines contraction and expansion but neglects to define shear performance for the joint.
The following section shares specific examples of the issues discussed above, and details how they have resulted in building enclosure issues of varying difficulty.
SELECTED EXAMPLES – STRUCTURAL FRAME DEFLECTION AFFECTING ENCLOSURE CONTROL LAYERS
Case Study 1 – Compounded Deflections and Curtainwall
With curtainwall assemblies, a specifier must define the amount of movement the curtainwall system should accommodate at its interface with the abutting rough opening or supporting structure if that movement originates from outside of the curtainwall system. This is true regardless of the specific material or construction that abuts the curtainwall. Furthermore, if the design of the rough opening or supporting structure assumes curtainwall connections back to it are fixed for a combination dead and wind load anchorage at certain framing locations and sliding for wind load only anchorages at other locations, the specifier must indicate the locations at which those conditions occur.
The primary seal for the curtainwall provides continuity of water and air control at the interface between the curtainwall and the weather-resistive barrier turning from the plane of the wall into the rough opening. The primary seal needs to be sized to accommodate anticipated movement at that interface to perform adequately.
The project in Case Study 1 featured a curtainwall elevation that was supported at its head condition with a wind load connection back to soffit framing. Contract documents instructed the curtainwall supplier to accommodate movements of the supporting structure while neglecting to define what those movements were. The head condition detail in the contract documents showed a sealant joint at the head with no indication of joint size, but it was drawn proportionately to infer that the joint was about ½-in. wide. A ½-in.-wide sealant joint using sealant with class 100/50 movement capability would be able to accommodate ½ in. of expansion movement and ¼ in. of compression movement.
A request for information was sent to the structural engineer of record to provide the movement accommodation needed at this condition. The response received was an instruction to consider the movement equal to a ratio of the edge beam span, expressed as L/360 for building live load deflection.
However, a review of structural framing occurring directly above the curtainwall head showed that additive multiple member deflections would dictate the required joint size. The curtainwall head connection was to framing that was supported by extended joist ends that cantilever beyond the end of the joist’s supporting beam. In this situation, the true structure deflection would be the combination of joist extension deflection at the point of curtainwall connection and the supporting beam deflection. This deflection could then be combined with the anticipated thermal movement of the curtainwall system to establish an appropriate value for movement accommodation at the curtainwall and rough-opening interface.
A second request for information was sent to the structural engineer of record to provide the movement accommodation that considers the compounding effects of both the joist-end deflection and the beam deflection, and to provide a numerical value rather than a ratio value. The second response received was a numerical value of 21/8 in. that included all components of structural movement.
As a result of the new information, the ½-in.-wide primary seal was modified to a 25/16-in. pre-cured silicone sheet, as a sealant joint would have needed to be in excess of 4 in. wide to adequately accommodate the required expansion and contraction (Figure 3).
In this instance, while the project team avoided an enclosure failure, several other issues required navigation. There was a six-week delay in arriving at a clear direction on what the joint configuration should be due to multiple rounds of RFI communication, coordination phone calls, and discussion on how to best modify the joint detail. There was a change order associated with the modification of materials and labor to achieve the new detail. The detail modification itself altered the aesthetic
Figure 3 – Head joint detail as shown in contract documents (left) and as shown after movement requirements were determined (right).
November 9, 10, 12, 16, 17, 19, 2020 | 2020 IIBEbeC BuildINg ENClosure Symposium P Porter | 37
appearance of the joint—there was now a more than 2-in.-tall band of sheet material where there was once a ½-in.-tall line of sealant. While a larger joint may have always been necessary, the true requirement was not fully communicated until well after the completion of design.
To avoid these issues and to prevent premature enclosure failure, the specifier must correctly establish the movement accommodation required. Ideally, this is defined in the construction documents prior to the work being bid. While it can still be resolved later in the project with engagement from both the designer and the curtainwall supplier, the sooner it is identified, the easier it is to facilitate a solution.
Case Study 2 – Roof Ponding Following an Adaptive Reuse Project
Adaptive reuse projects offer many benefits. They are a common strategy to extend the useful life of a built asset while offering the potential to reduce total construction costs in comparison to a new ground-up project. However, a comprehensive analysis of the existing structure is critical to establishing the true feasibility of the reuse. A thorough analysis will highlight project scope requirements, especially any measures needed to retrofit the existing structure so that it will comply with appropriate sections of the current building code. In model codes, the amount of retrofit typically required in an existing structure is based on the degree of planned modification to the original structure.
Retrofit analysis is particularly important when considering structural elements that were designed to older building codes that often prescribed lower serviceability requirements. For example, older building codes often did not address minimum roof-slope requirements, or mandate measures to be considered if the roof was less than a minimum slope. Current building codes require the design to assess the impact of roof slope and deflection on the likelihood of the surface’s ability to drain and remain stable. Specifically, Section 1611 of the International Building Code, in conjunction with Chapter 8 of ASCE 7, requires that if a roof is designed to a slope of less than ¼ in. per foot, that the roof structure needs to be analyzed for progressive deflection due to ponding instability.8,9 Standard practice in new commercial construction is to avoid designing roofs with less than ¼ in. per foot of slope. A few reasons for this are to avoid any potential issues with ponding instability, to comply with a roofing manufacturer’s warranty terms, and to avoid the need to apply for a variance in jurisdictions that specifically prohibit the use of shallower slopes.
It can be difficult to analyze existing structures in adaptive reuse projects due to the challenge of acquiring good information on existing structural materials and configurations. If comprehensive information is not available on structural elements, an analysis will often not be exhaustive, but rather made on a comparative or qualitative basis.
This case study was a one-story big-box retail building constructed in 1979. Based on a review of available existing drawings, it appeared that the roof structure was marginally designed with no structural provision beyond code minimums at that time. The existing built-up roof assembly was substantially dilapidated and saturated. The existing roof structure was a metal roof deck over open-web steel joists, which is one of the least stiff types of non-combustible construction commonly used.
The project converted the existing retail building into a new medical clinic. The project team was not required by code to convert the roof slope from its existing profile set by structural framing elevations to ¼ in. per foot. As the roof structure was not being modified to a degree where the code requirement was triggered, the project team did not perform an exhaustive roof ponding analysis. As a result, the existing 1/8-in.-per-foot roof slope, coupled with the existing framing deflection, resulted in ponded water at the mid spans of framing bays (Figure 4).
While the structure was confirmed to be stable under the ponded condition and the roof assembly was certified under accelerated weathering testing, nevertheless, the standing water presents opportunities for future leaks and could have been circumvented if a ponding analysis had been performed or if additional surface slope had been provided through a tapered insulation system. In this specific example, the condition did not result in a critical life safety issue. However, cumulative roof ponding as a result of not considering building movement is a very serious issue that has resulted in many roof collapses. This is a condition that warrants careful consideration, regardless of code requirements.
Figure 4 – Ponded water condition on the roof of an adaptive reuse project.
38 | Porter 2020 IIBEbeC BuildINg ENClosure Symposium | November 9, 10, 12, 16, 17, 19, 2020
Case Study 3 – Compounded Deflections and Finished
Insulated Metal Panels
Insulated metal panel (IMP) is an exterior wall system that typically features a thin metal surface skin encapsulating a core of foam plastic insulation. IMPs are available either with an exterior surface finish or for use as a backup wall that will be overclad with a finish material. In both cases, the panels come in standard widths at stock lengths. The panels interlock with each other in a tongue-and-groove fashion, where a bead of factory-applied non-curing sealant in the groove provides continuity of control layers from panel to panel. The stock lengths are cut to dimension, and continuity of control layers at the raw exposed edges of foam plastic is achieved through a combination of closure plates, self-adhered sheet membranes, and non-curing sealant.
The water control layers for IMPs are at their exterior surfaces, so finished IMPs are effectively a barrier wall system. The water control layer “dives” down into the panel-
to-panel joints, across the non-curing sealant, and then comes back out to the outer surface. Both the interlocking joints and the cut-to-length joints are intended to be static, with no differential movement occurring in the joints between panels. If such movement is anticipated, a true movement joint at the exterior surface is typically required to maintain the integrity of the water control layer.
The project in Case Study 3 featured finished IMPs oriented vertically. These panels ran past the slab edges in a three-story area of the building. In this area, the end of the building structure cantilevered off the last line of columns, with the cantilevers increasing in length at each subsequent story.
Multiple floors of cantilevers meant that compounded deflections would be experienced at the slab edge perpendicular to the cantilevers. If one floor was loaded on its cantilever, the cantilever’s supporting backspan would be deflected upward. At the same time, if an adjacent floor level was loaded on its supporting backspan, its backspan would be deflected downward while its cantilever would deflect upwards. These compounded deflections, which were additive in opposing directions, would result in the need for amplified movement accommodation.
While the contract documents required delegated design for the IMP system, they did not indicate the need to consider movement accommodation between the system and its supporting substrate, including movement in the IMP system at the floor lines. The slab edge details indicated a reveal in the IMPs but did not specify a true movement joint. After clarification was requested of the design team, it was determined that a movement joint was needed at the floor lines (Figure 5).
In addition to the joint added at the floor lines, the deflection amounts in the cantilevered edge beams and their backspans were large enough that there was a risk of binding or gapping at the panel interlock joints, which were not considered able to accommodate any movement as a component of the enclosure’s water control layer. The supplier of the IMP system determined and evaluated incremental values of slab edge deflection caused by the cantilever loading, based on panel widths, and determined that the amounts of movement in the panel interlock joints would be acceptable.
The loads in this case study were indirectly applied loads; that is, the loads originated on other components (the building’s structural framing). As the floor and roof framing loaded the cantilevers, the load was transferred over to and experienced by the IMPs. The applied loads included gravity and wind. The gravity loads of concern were live load from occupants and snow load from the roof. The wind load was applied as uplift at the roof level. Similar to Case Study 1, there was no enclosure failure because the inadequate deflection consideration was identified and addressed before installation. However, the issues described herein could have been avoided if the proper movement accommodation had been understood prior to bidding.
Case Study 4 – Stacked Ribbon Windows Interspersed with
Steel Stud Infill
Multistory construction with ribbon windows has been successfully used in building enclosures for many years. However, proper definition of where and
Figure 5 – Configuration of cantilevered framing (left) and IMP cladding with movement joints at the floor lines (right).
November 9, 10, 12, 16, 17, 19, 2020 | 2020 IIBEbeC BuildINg ENClosure Symposium P Porter | 39
how floor movement accommodation should occur is critical to the enclosure’s performance. Steel-stud and aluminum-
framed window systems that stack and bypass slab edges will require consistency of intent for movement accommodation from the time of design through the time of installation, including any fabrication and installation submittals created by installers whose trades comprise part of the wall enclosure. Any point along the process where design intent for movement accommodation is not reflected consistently is a place where omissions could occur that might impair the performance of the enclosure or result in premature enclosure failure.
The project in Case Study 4 featured a two-story elementary school. The exterior wall bypassed the slab edge at the second floor and the deck edge at the roof level, rather than being interrupted by the slab and deck edge framing. The wall composition included ribbon windows at the lower and upper levels, bounded vertically by stud backup walls at the floor-level bypass locations. The sequence of vertical stacking for the opaque wall sections and fenestration from grade was: steel-stud knee wall, ribbon window, steel stud running past the second floor line, ribbon window, steel stud running past the roof level.
Between the contract documents and the window supplier’s installation drawings, there was agreement on the loading configuration of the full-height wall. The concept was for both the windows and the stud wall to stack up from grade with provisions for a deflection connection back to the structure where it passed the second-
floor level and the roof level. While this presents a significant question as to whether the windows could truly support the dead load of the wall system above them, the window supplier did not take any exception to this approach. However, through the submittal process the steel stud supplier indicated that there should be fixed connections from their stud system back to the slab and deck edges, while requesting in their approval drawings that the window supplier provide a deflection connection at the heads of their frames (Figure 6).
While the architect responded with approval in their shop drawing review comments, the change was not communicated to the window supplier, and no formal documentation was issued. Subsequently, the stud wall was built with a fixed connection at the bypass locations. As a result, the second floor and, in turn, the roof deflection, were now able to impose their loads on the stacked wall system. Consequently, window head conditions would leak in early spring after the seasonal cycle of winter snow load deflection and then spring snow melt relaxation (Figure 7).
This case illustrates the importance of using design and shop drawing coordination and review as opportunities to validate continuity of design intent throughout the process of design, documentation, and implementation for enclosure systems. Additionally, it highlights the value of ensuring appropriate levels of communication among all affected parties.
Figure 6 – Static head joint consistent with contract documents (left) and steel stud supplier’s request to change the connection back to structure from slipped to fixed (right).
Figure 7 – Stacked wall assembly showing location of water intrusion.
40 | Porter 2020 IIBEbeC BuildINg ENClosure Symposium | November 9, 10, 12, 16, 17, 19, 2020
Case Study 5 – Parking Deck Surface Slopes and Water Infiltration
The American Concrete Institute Committee for Parking Structures states that “the slope of slabs should be designed in such a manner that water flows in the desired direction without ponding.”10 Industry best practice is to design deck surfaces to slope within the range of 1½–2%. This range is intended to allow for standard construction tolerances and anticipated deflections experienced by the slab. In essence, the goal is to ensure that positive surface slope is maintained to actively drain any water runoff to intended points of collection despite slope reduction from the as-built profile of the surface and from long-term deflections. In addition to this, slab camber and any unusual deflection conditions should also be factored into how surface slopes are established.
The primary intent of establishing adequate surface slopes is to minimize accelerated structural slab deterioration. This deterioration is due to water percolating through the slab thickness as a result of prolonged standing water on deck surfaces. Usually, parking deck surface slope is not considered an enclosure integrity issue. However, water infiltration through slab cracks will accumulate alkalis as it migrates through the depth of the concrete. Water that reaches the underside of the slab will have attained a high pH level. When this water drips onto cars below, it can be caustic enough to damage the finish of the vehicles. In this way, parking deck slope surfaces can be considered an enclosure issue because of a lack of water control into an enclosed environment.
The project in Case Study 5 featured a multi-level parking structure below grade. The floor system used cast-in-place post-tensioned concrete flat slab construction. The project team decided to reduce slab surface slopes from the recommended ranges with a goal of minimizing excavation depths for the project. After one seasonal cycle of in-service use, when the formation of micro cracks had been established and vehicles had brought in snow, ice, water, and de-icing materials, car finishes began to suffer damage if vehicles were parked on the lower levels (Figure 8).
Consequently, the owner of the parking structure was obligated to pay for auto finish repair work, and a vehicular traffic coating was required to prevent continued issues with damage to vehicles due to water migration of corrosives. This case was an example of deflection in a structure where the surface changed from a condition of adequate to inadequate slope. With the rate of surface water runoff slowed, the rate of water percolation through the slab increased to a point where the enclosure no longer adequately protected its interior volume. Additionally, though the moisture was not a structural load itself, it is significant to note that the effect of water percolating through the slab converted a liquid solution from a less harmful environmental load (snow with road salt) to one aggressive enough to cause damage (water with soluble alkali salts).
SELECTED EXAMPLES – MATERIAL VOLUME CHANGE AFFECTING ENCLOSURE CONTROL LAYERS
Case Study 6 – Solar Heat Gain on a Self-Adhered Membrane Flashing
Self-adhered membrane flashing is a universal component within exterior wall assemblies when utilized as an accessory to a primary control layer component. It is used for continuity of control layers that might be interrupted due to the inherent nature of the materials or the construction methods used. In most cases, the flashing is concealed behind finished cladding, and, thus, opportunities for visual observation of the flashing’s behavior and performance are typically limited to the time when the project is under construction.
Flashing facer material comes in a wide variety of colors. These colors can be
Figure 8 – Alkaline solution created from runoff percolating through slab (left), and subsequent damage to auto finish (right).
November 9, 10, 12, 16, 17, 19, 2020 | 2020 IIBEbeC BuildINg ENClosure Symposium P Porter | 41
a reflection of the manufacturer’s branding objectives, or a desire to differentiate their product from competitors’ products. However, it is known that color affects a material’s potential to absorb heat; this fact is what largely drives an ongoing debate between white versus dark roof coverings and whether heat should be absorbed through the roof assembly or reflected by the outer surface. In the case of flashing, however, thermal expansion and contraction of the material itself can become more critical than the degree to which it is reflecting heat or allowing heat through.
In the project for Case Study 6, an integrated sheathing system that required field treatment of joints, corners, openings, or other locations of discontinuity within the field of the water control layer was being used on an exterior non-load-bearing steel-stud wall. The treatment selected for the project was a medium-bodied flashing material that was black. Portions of the exterior enclosure were installed throughout the winter months. The project is located in climate zone 6a, which is defined as “cold-humid” and extends across portions of the northern Midwest states and upper areas of the U.S. Northeast.
During installation at the end of December, it was discovered that the flashing material was gapping or “fishmouthing” at its outer edges on sunny days. After the sun set in the afternoon, the open edges would disappear. The project team investigated further and found that on days when outside air temperatures would remain around 30° F, the surface temperature of the flashing when exposed to direct sunlight could reach over 130° F, (Figure 9). It was determined that the flashing’s solar heat gain resulted in the material expanding at a rate greater than the underlying substrate. Consequently, channels would form behind the flashing.
These channels represented an infiltration risk, where incidental water behind the exterior cladding could possibly infiltrate the wall assembly. Various remedial actions were discussed. One option considered was to add a termination bead of sealant at the top edge of any flashing that was aligned horizontally. Another option discussed was to install an additional “cap” layer of light-colored flashing in shingle fashion above any horizontal run of the dark flashing. The second option was chosen and implemented on the project (Figure 10).
While the solution did not appear elegant, it was determined to be effective through subsequent water testing on the system. It is worth noting that this remediation step, when repeated at all instances on a building with a large enclosure surface area, can result in a significant increase in labor and material cost. The job was only shut down for one week; however, the schedule for installation was prolonged and an additional cost of $30,000 was taken from contingency.
Future recommendations to the flashing manufacturer included a request that they produce flashing material in a lighter color, in an attempt to achieve greater alignment of heat gain potential between the flashing material and the substrate. In addition to considering heat gain potential, it is also important that self-adhered flashings be properly installed with compression through the use of a J-roller or a plastic spreader to improve adhesion between surfaces, and that the flashing not be exposed for longer than the manufacturer’s recommended duration.
The load demonstrated in this case study was directly applied, while volume change was a consequence.
Case Study 7 – Shrinkage of
Site-Manufactured Spray
Polyurethane Foam
In the past seven to ten years, site-
manufactured spray polyurethane foam (SPF) has become increasingly common on jobsites, ranging in application from limited areas to broad expanses of entire wall or roof planes. Site-manufactured SPF is a two-
component system in which a manufacturer-
produced part “A” and part “B” are combined onsite to create foam plastic. Because of its promoted use as a component of thermal, air, vapor, and even water
Figure 9 – “Fishmouthing” of flashing (left) and surface temperature recording (right).
Figure 10 – Remediation options investigated: flashing cap layer (left) and termination bead of sealant (right).
42 | Porter 2020 IIBEbeC BuildINg ENClosure Symposium | November 9, 10, 12, 16, 17, 19, 2020
control, it is often perceived as a material solution that can address a wide range of enclosure problems despite the challenges that can be associated with it.
As a site-manufactured material, it requires various quality control measures before and during installation to ensure a consistent product result. Material shipping, handling, and storage; installation equipment maintenance and calibration; and processing controls are all critical to ensure a successful installation.11 This paper does not discuss how and why SPF experiences shrinkage failures, but rather focuses on consequential risks and possible remediation actions.
One of the more common installation failures for SPF is excessive shrinkage after the product has been placed and is set up in its hardened state. Depending on the failure, shrinkage can initiate anywhere from several days to several months after placement. Depending on the intended function of the SPF and the degree to which it shrinks, the shrinkage can be either a minor or significant concern. In cases where the shrinkage is so great that the intended control function of the SPF is breached, the undesired bypass might represent a risk. In the case of SPF being used only as a thermal control layer, shrinkage-
induced bypasses might be viewed as a relatively benign issue. In the case of SPF being also used as an air-control layer, shrinkage-induced bypasses can become a considerably larger concern if the air bypass now enables moisture-laden air to condense in lower-temperature cavities or regions. At the far end of the spectrum, if the SPF shrinkage is so great that it damages other control layer components in the exterior enclosure, then the risk is likely to be the most significant.
Two representative projects are discussed for this case study. The first project featured exterior steel-stud wall construction with a fluid-
applied air and water-
resistive barrier placed on exterior sheathing. The stud wall construction bypassed the second-floor slab and roof deck edges. At the roof bypass, SPF “plugs” were placed in the stud cavity at the roof deck insulation elevation to prevent movement of vapor-laden air up into the unheated stud parapet cavity. Due to distress visually observable on the outside face of exterior sheathing at the roof level, it was determined that the SPF had shrunk excessively. After creating inspection openings in the rear side of the parapet wall, shrinkage was confirmed to have pulled the SPF away from the studs and slab edge face, creating an unacceptable air bypass condition.
To remediate the shrinkage condition, all suspected bypass areas were identified with infrared (IR) imaging. First a relief cut was made in the SPF plug, then a new cap layer of SPF was reapplied over top of the cut plug with a renewed focus on processing and placement controls for the install of the fix. Finally, the repair was confirmed with IR imaging to verify that all bypass conditions had been eliminated (Figure 11).
The second project for this case study featured SPF applied along the full height of the exterior walls to the interior side of exterior gypsum sheathing within a steel stud cavity. A fluid-applied air/vapor/water barrier (WRB) was placed on the outer face of the sheathing. Joints in the sheathing were taped with a self-adhered membrane flashing. Due to distress visually observable on the outside face of exterior sheathing, it was determined that the SPF had shrunk excessively. The shrinkage of SPF occurred to such a degree that it bowed the exterior sheathing inward, cracking both the sheathing facer and the WRB (Figure 12).
Figure 11 – SPF shrinkage (left) and use of an IR camera to identify air bypass locations (right).
Figure 12 – Cracks in exterior sheathing and WRB as a result of SPF shrinkage; inspection openings are visible at the bottom third of the wall.
November 9, 10, 12, 16, 17, 19, 2020 | 2020 IIBEbeC BuildINg ENClosure Symposium P Porter | 43
To investigate the condition, test cuts were made on the exterior side of the wall, and representatives from both the SPF manufacturer and the WRB manufacturer performed a site observation. The results of the observation indicated that the SPF acting only as a component of the thermal control layer was not compromised to an unacceptable degree. The WRB manufacturer recommended the cracks be taped over with additional flashing material as an acceptable fix for the compromised WRB.
In both of these situations, repairs were accomplished and the projects were able to proceed. However, several months of additional work and tens of thousands of dollars of added costs could have been avoided if the shrinkage had not occurred. These examples highlight the importance of proper controls on the use of site-
manufactured SPF, to reduce project delays and additional costs. An additional focus on quality control should include validation that the two base components are handled and stored properly within manufacturer-recommended temperature ranges, that spraying equipment is calibrated to the correct mixing ratio, that pressures and line temperatures are within acceptable ranges, and that temperature and relative humidity of the air and the receiving substrate are within manufacturer-
recommended ranges.
Case Study 8 – Sealant Bite on an Aluminum-Framed Opening
This case study does not feature an observed enclosure failure, but rather illustrates the importance of detailed consideration for volume- change-induced
movement. Aluminum-framed openings—whether they are storefront, curtainwall, or AW-class windows — are all fabricated using extruded shapes. The cross section of these shapes is compound and often not symmetrical about any axis. Where framing members intersect normal to each other, conventional fabrication techniques most often rely on butted corners rather than mitered corners.
The bond line surface on the aluminum for the opening’s primary seal is typically a narrow band or lip of the extrusion that is parallel to the plane of the rough opening, usually in the range of 5/16 in. to ½ in. wide. In this way, a sealant joint can be made that will have a proper profile with opposing adhering surfaces. However, in aluminum-framed construction that uses butted corners on asymmetrical extrusions, the return of the corner through the width of the frame will result in the bond line
Figure 13 – Yellow arrow indicates typical bond line surface for adhesion of primary seal along the jamb condition. The purple arrow indicates bond line surface at butted corner joint where the frame transitions to the sill condition (the frame is tipped onto its face in the photo).
Figure 14 – Increasing the adhesion surface area with a fillet reinforcement. In this case, the profile of the sealant needed to be increased, or reinforced to allow for additional adhesion surface area.
44 | Porter 2020 IIBEbeC BuildINg ENClosure Symposium | November 9, 10, 12, 16, 17, 19, 2020
surface being limited to the wall thickness of the front face of the extrusion, which can often be in the range of only 1/16 in. to 5/64 in. thick (Figure 13).
When this occurs, the ability of the primary seal to perform adequately is severely reduced. This impairment is created by the bond line surface defining the critical section through the sealant. For example, if the typical joint width is 5/8 in., the width-to-depth ratio will be somewhere between 8:1 to 10:1; however, the narrowest section of the sealant will not be a mid-section of the joint, but at the bond line with the aluminum frame. Compare this to a reasonable amount of thermal expansion or contraction for an aluminum-frame side that is 72-in. long. Assuming a small temperature difference of 70° F: [24 x 10^(-6)] x 72 x 70 = roughly 1/8 in. This 1/8 in. represents potential compression or expansion for a bond interface that is 5/8 in. by 1/16 in. to 5/64 in., where for the sake of water control layer integrity, the 1/16-in.- to 5/64-in.-wide adhesion plane must resist all of the expansion and contraction movement without any failure.
When this condition is encountered in detailing situations, it is critical that detailers recognize the “weak link in the chain” and that they modify the detail appropriately. One possible solution would be to add end dams at the sill condition. Another possible solution would be to use a sill receptor that can provide a backdam independent of the window frame. If neither option is viable, a suggested best practice is to provide an additional filleted reinforcement to the sealant profile for the width of the reduced section so that additional adhesion surface will be provided through the critical area (Figure 14).
Coupled with this reinforcing should be the addition of a width of bond breaker on the rough opening surface, reducing the total bond line width on that surface to better match that on the aluminum-framed side. While the sealant joint rolling up onto the front surface of the frame might be aesthetically objectionable, it greatly increases the competence of the primary seal, and accordingly reduces the risk of water intrusion in the corner of the opening as a result.
Case Study 9 – Slab Jacking in
a Plaza Deck Assembly
Plaza deck assemblies are a common design function when an exterior space is planned directly above an interior space. They are also referred to as inverted roof or protected roof membrane assemblies.12 Their basic composition, from bottom to top, is as follows: a supporting structural deck, a waterproofing layer, a drainage layer, a rigid insulation layer, and a wear surface exposed at the top side. Many different types of surface finish materials are employed for plaza decks, such as roofing ballast, pavers, or slabs. For the purposes of this case study, we will focus on cast-in-place concrete wear slabs for the use of parking on the top surface.
Plaza deck wear slabs are subjected to thermal and moisture volume change movements, as they are located outside of the thermal, water, and vapor control layers. Detailing of wear slabs for horizontal movement accommodation is critical to the protection of the waterproofing layer if that waterproofing layer turns up vertical surfaces and/or terminates within the depth of the wear slab thickness.
For this case study, two projects will be considered. Both projects featured typical assembly construction. The primary difference between the two was that one used 215-mil hot-rubberized asphalt for its waterproofing layer, while the other employed 60-mil EPDM. Both projects experienced failure within seven years of their original construction, which was not covered under warranty. The failure exhibited water intrusion into the space below, substantial exterior wall cracking and movement at building corners, light pole bases tilting, and accelerated wear slab deterioration (Figures 15 and 16). The primary cause for failure was lack of sufficient expansion joints and sealed control joints in the wear slab, coupled with inadequate deck maintenance procedures.
The mechanism to failure was progressive. The wear slab would shrink in volume over winter months due to reduced temperature, opening untreated control joints and creating new tensile cracks. These gaps would then fill with smaller debris such as sand and roadway grit. At the seasonal change to warmer weather,
Figure 15 – Exterior wall cracking and movement at the building corners. Note the kickers installed to provide temporary stabilization.
Figure 16 – Tilted light poles as a result of base rotation.
November 9, 10, 12, 16, 17, 19, 2020 | 2020 IIBEbeC BuildINg ENClosure Symposium P Porter | 45
the slab would grow in volume and push against the debris-filled gaps that could no longer expand to their closed-gap condition. Each successive seasonal cycle would set a new baseline of minimum horizontal dimensions based on the maximum extent of thermal and moisture expansion plus debris-laden gaps, and then progressively jack against itself to new baseline dimensions with each new season (Figure 17). Eventually the force of jacking distress on upturned surfaces of waterproofing reached the point of breaching the waterproofing.
This evaluation was somewhat controversial on one project, because the prevailing belief was that the cause of failure was snowplowing activities and the resulting force of impact on walls and light pole bases. However, the wall areas of maximum distress were consistently in locations that could not be reached by snowplow blades or by accumulated snow storage piles. They consistently occurred most often at outer corners of the parking plate where the largest expansion displacements would typically be expected.
Repairs were performed at both projects, removing the assembly components down to the waterproofing layer and repairing or replacing that layer, followed by a complete rebuilding of the plaza assembly. In both projects, an additional drainage layer was added above the insulation layer to improve the overall drainage efficiency of the assembly and to reduce the occurrence of moisture accumulation within the rigid insulation layer.
The new wear course slabs featured tighter patterns of fully sealed control joints, with no joints left open. Additional bellows or gland-type expansion joints were placed strategically to influence the center-
of-mass location for separate slab plate areas. The design and construction was then followed by strong recommendations to consistently follow a parking deck maintenance program that would include regular surface sweeping throughout the year, and wash-down of the decks in mid-fall and early spring. Additionally, the program would include regular inspection of all construction and control joints to confirm that sealant remained intact. If sealant or control joints were compromised, they should be replaced promptly.
This case study shows the effect of an indirect applied load creating a different load that leads to failure, as well as the importance of coupling exterior enclosure maintenance with good design and construction practices to maximize the useful service life of the enclosure.
CONCLUSION
In the design and construction of effective building enclosures, one of the first necessary steps is to consider the interaction between the enclosure and the loads that it experiences.13 Unsurprisingly, loads that receive the bulk of attention are liquid water, air, water vapor, and heat. This is completely appropriate, because these loads impact enclosure behavior the most. However, structural loadings are also critical, because continued effective control of heat, air, and moisture is not possible if potential effects of movement are not considered.
This paper has discussed various types of structural enclosure loads and how those loads affect the design and construction of the enclosure. In addition, specific examples have been shared where the enclosure has faced the prospect of impaired or significantly compromised integrity because of the project team neglecting to consider the impact of structural loads on the enclosure. For building enclosures to perform effectively, design and construction activities must include thoughtful and deliberate consideration of structural load. The following are a few recommendations to increase project teams’ success in this area.
Technical Design Improvement. The current state of the design industry is one where the “look and feel” of design receives a significant amount of attention before the built project has experienced any substantial service use. This often results in a deficit of attention paid to buildings’ durability and how design can ensure that enclosures will perform to the expectations of owners and end users. It is critical that the parties responsible for assembling contract documents clearly provide instruction such that the completed project will not just look as expected, but also perform as expected.
Training. Design professionals must have a minimum level of understanding of how materials behave when subjected to various loadings. While this baseline level does not need to include a higher-order study of material science or engineering mechanics, design professionals must know that materials grow, shrink, move, and take on different shapes in response to the environment in which they are placed. This can involve course work and classroom training, but hands-on exposure is essential and is a critical component for attaining this knowledge.
Performance Specifications Need to Define Performance. When one party assigns a portion of their responsibility to another party, expectations must be clearly communicated and information must be provided that will establish the desired outcome. Trade contractors involved in exterior enclosure construction have no
Figure 17 – Progressive growth of unsealed control joints.
46 | Porter 2020 IIBEbeC BuildINg ENClosure Symposium | November 9, 10, 12, 16, 17, 19, 2020
means of anticipating the magnitude of movement outside of their systems other than the parties responsible for the design of systems that create the movement. This information needs to be shared at a time when all parties are most able to efficiently execute against the information—that is, prior to signing of contracts with the installer, but certainly no later than the shop drawing phase.
Include Movement Information on Field-Use Drawings. Movement-induced failures in the exterior enclosure most often appear at interfaces between work performed by different trades. However, installers’ drawings often will not include significant information about components or elements that abut their work if those elements are outside of their contract to furnish or install. Field-use drawings need to bridge this gap by clearly identifying joints or transitions between trades where movement is anticipated and explaining specifically how that movement will be accounted for.
Use Steps in the Documentation and Approval Process as Tollgates. Each time any single party conveys their impression of a plan, an opportunity is created for all stakeholders to confirm a shared understanding of that plan. From documenting owners’ project requirements, to the creation of construction documents, to the submittal of fabrication and installation drawings, to the documentation of pre-installation meetings, to numerous other quality assurance/quality control measures that are easily adopted, each step along the way is a chance to validate the direction for the project. Project teams must be committed to taking advantage of these check-ins along the way, so that movement-related enclosure problems will be prevented before they become painful to deal with.
All of these recommendations, especially the final one, lead to a larger aspect—quality assurance and quality control. Many of the case studies shared included issues that were caught by oversight that would have been much greater problems if left undetected. This points to the importance of a thorough peer review process during design, through building enclosure commissioning by enclosure consultants or through in-house technical specialists on both the design side and the construction side.
With a commitment from both the design and construction professionals at the onset of a project, and the use of the suggestions outlined above, building enclosures can be successfully delivered and perform under the structural loadings they experience.
REFERENCES
1. Tague, N. R. 2005. The Quality Toolbox. ASQ Quality Press. pp. 390-396
2. AAMA 502-12, Voluntary Specification for Field Testing of Newly Installed Fenestration Products.
3. Straube, J. F. and E.F.P. Burnett. 2005. Building Science for Building Enclosures. Building Science Press. pp. 32–35.
4. Tartaglione, L. C. 1991. Structural Analysis. McGraw-Hill, Inc. pp. 2-5.
5. Lawson, H. June 12, 2014. Buildings – Plugging the Performance Gap. https://blogs.bsria.co.uk/tag/millennium-bridge/.
6. ASTM E2813, Standard Practice for Building Enclosure Commissioning. ASTM International. Conshohocken, PA.
7. Wald, M. L. September 2, 2001. “Rescuing a World-Famous but Fragile House.” The New York Times. https://www.nytimes.com/2001/09/02/us/rescuing-a-world-famous-but-fragile-house.html.
8. International Code Council. 2015. International Building Code.
9. American Society of Civil Engineers. (2010). ASCE Standard ASCE/SEI 7-10, Minimum Design Loads for Buildings and Other Structures. Reston, VA.
10. ACI 362.1R, Guide for the Design and Construction of Durable Concrete Parking Structures. Farmington Hills, MI.
11. Fennell, H. 2012. “Avoiding Problems With Spray Foam.” The Journal of Light Construction. https://www.jlconline.com/how-to/insulation/avoiding-problems-with-spray-foam_o.
12. DOW Building Solutions. Protected Membrane Roof Installation Guidelines.
13. Kesik, T. J. 2016. “Building Enclosure Design Principles and Strategies.” Whole Building Design Guide. https://www.wbdg.org/resources/building-enclosure-
design-principles-and-strategies.
November 9, 10, 12, 16, 17, 19, 2020 | 2020 IIBEbeC BuildINg ENClosure Symposium P Porter | 47