Water Testing Misconceptions: fenestrations Product Certification and Forensic Investigation of Building Leakage

WATER TESTING MISCONCEPTIONS:
FENESTRATION PRODUCT CERTIFICATION AND
FORENSIC INVESTIGATION OF BUILDING LEAKAGE
FRANCESCO J. SPAGNA, PE; AND BRANDON S. BUCHBERG, PE
SIMPSON GUMPERTZ AND HEGER, INC.
41 Seyon Street, Building 1, Suite 500, Waltham, MA 02453
Phone: 781-907-9224 • Fax: 781-907-9009 • E-mail: fjspagna@sgh.com and bsbuchberg@sgh.com
2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2 B U C H B E R G A N D S P A G N A • 9 3
ABSTRACT
One of the primary goals of building envelope leakage investigations is to trace the
sources of building leaks. While national testing standards used to certify newly installed
fenestration products for compliance with specified performance criteria can be adapted and
used in building envelope leakage investigations, water testing for the purpose of tracing the
sources of building leakage should be based on physical evidence, experience, and engineering
judgment rather than on guidelines for product certification. Investigators and other
professionals often mistakenly apply guidelines for product certification testing during building
envelope leakage investigations, misunderstand the purpose of accepted test devices
(such as the applicable water spray volume), and apply inappropriate test procedures.
This paper will provide an overview of fenestration product certification. Through a study
of testing standards and case studies of building leakage investigations, the paper will differentiate
between test methods applicable to fenestration product certification and those
applicable to building envelope leakage investigations. It will also illustrate some common
ways that product certification test methods are misapplied during building envelope leakage
investigations.
SPEAKER
BRANDON S. BUCHBERG, PE — SIMPSON GUMPERTZ AND HEGER, INC. WALTHAM,
MA
BRANDON BUCHBERG, PE, is a senior staff I in the Building Technology Division at
Simpson Gumpertz & Heger Inc. His work involves design, investigation, and repair of building
envelopes pertaining to roofing, cladding, windows, curtain walls, and waterproofing
(above-grade, plaza, and below-grade). In his work, he performs water testing to identify
leakage paths and to verify fenestration performance. Mr. Buchberg is a member of ASTM
Committee C18 and a member of the Building Enclosure Council of the Boston Society of
Architects. He has bachelor’s and master’s degrees in civil engineering from the Georgia
Institute of Technology.
COAUTHOR
FRANCESCO J. SPAGNA, PE — SIMPSON GUMPERTZ AND HEGER, INC. WALTHAM,
MA
FRANCESCO J. SPAGNA, PE, is a senior project manager at Simpson Gumpertz & Heger
Inc. (SGH). He is experienced in the performance of building envelope investigations, condition
assessments, construction litigation support, design of building envelope repairs and
subsequent construction administration, and peer reviews of building envelope designs. He
has been involved in the investigation and remedial design of curtain walls, windows, roofing,
EIFS and stucco claddings, brick and stone veneer, load-bearing masonry, plaza waterproofing,
and below-grade waterproofing. He has a bachelor’s degree in architectural (structural)
engineering from Pennsylvania State University and a master’s degree in civil (structural)
engineering from North Carolina State University.
9 4 • B U C H B E R G A N D S P A G N A 2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2
WATER TESTING MISCONCEPTIONS:
FENESTRATION PRODUCT CERTIFICATION AND
FORENSIC INVESTIGATION OF BUILDING LEAKAGE
INTRODUCTION
Water testing is common practice to
evaluate new building components (such as
windows and doors) for resistance to water
penetration or to diagnose reported leakage
paths in buildings. The architecture
/engineering (A/E) industry has developed
methods and procedures for water testing
as a way to standardize testing. Water testing
for certification is the testing of fenestration
products in a laboratory to confirm
compliance with a stated level of performance
and to provide rating to that product
to identify its performance. Certification
testing of fenestration products adheres to
standardized test methods that differ from
those for investigative testing to assess
building leakage.
Water testing standards are often misused
or misquoted during both certification
programs and forensic investigations.
Testing pressures, test durations, and
methods for testing in general are often
debated amongst investigators. Misuse of
standards can result in misleading results,
confusion, delay, and additional costs to a
project; or delays and additional costs in
dispute resolution. Understanding the standards
and their intent is essential for investigators,
whether the purpose is evaluating
a fenestration product or identifying the
cause of leakage in a building.
DEVELOPMENT OF STANDARDS
The development of current water testing
standards dates back to 1950, when the
University of Miami began a testing program
to study the water penetration resistance
of jalousie windows.1 The testing
involved subjecting the window to a water
spray while an aircraft propeller forced air
against the window specimens. This test
method was a predecessor to the current
AAMA 501.1. Standard Test Method for
Water Penetration of Windows, Curtain
Walls, and Doors Using Dynamic Pressure.
Around 1953 in Norway, researchers
created an individual spray rack that moved
vertically and horizontally along the test
specimen face.2 This spray rack, however,
was not adaptable to larger specimens, so
use dwindled. Also, in the early 1950s,
researchers in the United States developed
a spray rack consisting of a perforated horizontal
pipe located at the top of a window
specimen.3 The testing setup wet the window
specimen using gravity flow but failed
to test overhang surfaces such as a meeting
rail on a double-hung window. To increase
the coverage, the spray rack was modified to
include a grid of vertical and horizontal perforated
pipes that covered the framing
members of the window.
Around 1958, the Aluminum Window
Manufacturers Association (AWMA) created
a spray rack with a series of commercially
available nozzles placed in a grid pattern.4
The rack was adaptable to any size specimen
and more economical to construct
than its predecessors.
In 1964, the American Society of Testing
and Materials (ASTM) began work on a static
water pressure test for windows. The
standard was published in 1967 as ASTM
E331-67T, Tentative Method of Test for
Water Resistance of Windows by Uniform
Static Air Pressure Differential. Other standards
followed with the publication of ASTM
E547, Test for Water Penetration of Exterior
Windows, Curtain Walls, and Doors by
Cyclic Static Air Pressure Differential, in
1975; and the publication of ASTM E1105,
Standard Test Method for Field Determination
of Water Penetration of Installed
Exterior Windows, Curtain Walls, and Doors
by Uniform or Cyclic Air Pressure Difference,
in 1986, which adopted Standards E331
and E547 to perform water testing in the
field.
In 1985, the American Architectural
Manufacturers Association (AAMA) produced
AAMA 101-85, Voluntary Specifications
for Aluminum Prime Windows and
Sliding Glass Doors, replacing AAMA 302.9-
1977, Voluntary Specifications for Aluminum
Prime Windows, and AAMA 402.9-1977,
Voluntary Specifications for Aluminum
Sliding Glass Doors. In 1997, AAMA published
AAMA/NWWDA 101/I.S.2-97, a
standard consolidating the I.S.2-93, Wood
Windows by the National Wood Window and
Door Association (NWWDA), which pertained
to wood fenestrations, with AAMA
101-93, Voluntary Specifications for
Aluminum and Poly (Vinyl Chloride) (PVC)
Prime Windows and Glass Doors, which
applied to aluminum and PVC windows.
AAMA published AAMA 502, Voluntary
Specification for Field Testing of Windows
and Sliding Glass Doors in 1990 to provide
guidance for evaluating windows and doors
in the field. This standard evolved into the
current 2008 version of AAMA 502,
Voluntary Specification for Field Testing of
Newly Installed Fenestration Products.
AAMA 502-08 provides guidelines for fieldtesting
of newly installed windows to verify
water penetration resistance and air infiltration
resistance. The standard requires
water testing the windows using the cyclic
or uniform static air-pressure testing methods
stated in ASTM E1105.
Today, ASTM Standards E331, E547,
and E1105 are used in conjunction with
AAMA 101 and AAMA 502 to evaluate the
performance of installed fenestration products
with respect to specified performance
criteria. These standards provide requirements
for the test pressure, procedures,
and duration. These standards, however,
are often misapplied during building leakage
investigations, where the purpose (in
part) is to replicate known building leakage
and not necessarily to evaluate windows
with respect to their originally specified performance
criteria.
In 2001, ASTM published ASTM E2128,
Standard Guide for Evaluating Water Leakage
of Building Walls, to provide guidance in
investigating water leakage through building
walls. This differs from the previous
Standards E331, E547, and E1105, which
were primarily used to evaluate fenestration
products. In 2008, AAMA published AAMA
511, Voluntary Guidelines for Forensic
Water Penetration Testing of Fenestration
Products, which focuses on identifying leak-
2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2 B U C H B E R G A N D S P A G N A • 9 5
Class Y PG NN Size Tested W x H in.
Misconception No. 1: Design pressure
is sufficient to specify the performance
of a window.
Misconception No. 2: Laboratory certification
test reports for a model of window
indicate that the window will perform
adequately for my project and to
the same level of performance once
installed in the field.
age paths of in-service fenestration products
(as opposed to evaluating windows
with respect to originally specified performance
criteria) and follows guidelines set by
ASTM E2128.
PRODUCT RATINGS
AAMA 103, Procedural Guide for Certification
of Window and Door Assemblies,
establishes the guidelines of the AAMA certification
program. Products that are tested
in accordance with AAMA 103 can be AAMAcertified
and identified with a gold AAMA
label, and certifications are valid for four
years. AAMA 103 currently allows manufacturers
to rate windows based on the 1997,
2002, 2005, or 2008 versions of AAMA 101.
However, certification testing to AAMA 101-
97 will end March 1, 2012; therefore, products
certified under this standard will be
valid until March 1, 2016.
AAMA 101-08 provides the performance-
class rating system and testing
requirements for windows, doors, and skylights.
We will focus on this standard as it
relates to water testing of windows.
Windows are designated based on performance
class, performance grade, and
maximum size tested. The performance
class rating associated with a specific window
indicates that the product meets a certain
level of performance. Windows that are
tested in conformance with AAMA 101
should exhibit a label similar to the following:
– –
where Y = Performance class (R, LC,
CW, AW)
NN = Performance grade number
W x H = Maximum size tested, width x
height of window (listed in inches, or metric
with English standard units in parentheses).
Performance class ratings are defined as
follows:
R = Residential
LC = Light commercial
CW = Commercial window
AW = Architectural window
This designation is slightly different
from that included in the previous versions
of AAMA 101 (1997, 2002, and 2005) in that
the previous versions included product type
and alternate performance-class designations
(R, LC, C, HC, and AW).
9 6 • B U C H B E R G A N D S P A G N A
CERTIFICATION TESTING
Water testing developed for certification
of fenestration products stipulates certain
air pressures, water distribution, and water
volume. For example, AAMA 101-08 states
that windows having performance classes
R, LC, and CW are water-tested in conjunction
with an applied air-pressure differential
equal to 15% of the design pressure.
Windows having performance class AW are
tested using air pressure differential equal
to 20% of the design pressure.
Using this standardized rating system
and protocol, manufacturers produce windows
with an expected level of performance
and market the products as such. The
intent is to allow designers, owners, contractors,
and end-users to select window
products with a performance that meets
their project needs. Although this is also the
case for doors, for brevity, we refer to windows
only in this paper. It is the responsibility
of the design professional to determine
the required rating for a particular
project based on various conditions, including
type of building, typical weather, and
building code requirements.
The laboratory certification testing programs
only evaluate the individual products
that are provided by the manufacturer for
the testing. The manufacturer then applies
the performance class rating to every window
of that model that it produces. The rating
assigned to a window product does not
guarantee the reliability that a given production
model will exhibit the same performance
results in the field as the sample
tested in the laboratory. Workmanship,
adherence to material tolerances, material
formulation, etc. can all result in variations
from tested laboratory performance. Field
inspection, critical review of perimeter
details, and field-testing of mock-up window
installations should be utilized to evaluate
the performance of the fenestration
product in the field.
AAMA 101-08 provides
a clear distinction between performance
grade (PG) and design pressure (DP).
These terms often are inappropriately interchanged.
Design pressure specifically
relates to a window product’s ability to perform
adequately when tested at a structural
uniform loading per ASTM E330. A performance
grade is assigned to a window
product when the product performs ade-
2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N
quately when tested in accordance with air
leakage resistance, water penetration resistance,
uniform load deflection, uniform
structural loading, operating force (if applicable),
and force-entry resistance (if applicable).
Designers/specifiers should specify the
required performance grade of a product in
their specifications, rather than just the
design pressure.
Further, designers should specify the
equation used to develop the water testing
pressure such as, “Water testing pressure =
design pressure x 15%,” instead of specifying
a test pressure. If the contractor provides
a window with a performance grade
higher than specified in the specifications,
specifying the equation will ensure that the
window is tested in relation to its corresponding
rated test pressure rather than a
lower test pressure that may be stated in
the project documents.
A certification test
report shows that a sample of a particular
model of window of a particular size met a
specified level of performance during certification
testing. It does not indicate whether
that window is appropriate for the winds,
exposure, and other building-code requirements
that a designer should consider.
AAMA labels showing the product rating
as tested in accordance with AAMA 101
specify the maximum-size window tested.
Prior to installation, the design professional
should verify that the size of the window
used for the project is smaller or equal to
the listed “maximum-size-tested” value
stated on the label for the window. In some
cases, manufacturers will have multiple
design pressures for the same window
model. In general, when dealing with the
same window model, smaller-sized windows
will perform to a higher design pressure for
water penetration, but this is not necessarily
the case for other criteria such as air
infiltration.
Certification testing is also not a quality
control program and does not always guarantee
comparable field performance, nor
does it ensure reliability of manufacture or
workmanship. Shipping, handling, and
installation tolerance can affect the performance
of installed windows on a project.
These effects are not accounted for in labo-
A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2
Misconception No. 3: A certified window
indicates the window passed a level
of performance and will not allow water
penetration.
ratory certification testing. The testing also
does not account for aging effects such as
shrinkage and embrittlement of gaskets and
deterioration of internal seals over time, all
of which can detrimentally affect window
performance.
It must also be noted that the rated performance
of a window product is not indicative
of the overall performance of the wall in
which the window is installed. Window
perimeter conditions can affect the performance
of the wall system as a whole, and
these components are not necessarily tested
during certification testing.
This statement depends on
the definition of leakage established for a
particular project or testing program. The
stated definition of water penetration in
ASTM E331 (certification in laboratory),
ASTM E547 (certification in laboratory), and
ASTM E1105 is:
Penetration of water beyond a plane
parallel to the glazing (vertical plane)
intersecting the innermost projection
of the test specimen, not including
interior trim and hardware,
under the specified conditions of air
pressure difference across the specimen.
For products with nonplanar
glazing surfaces (domes, vaults,
pyramids, etc.) the plane defining
water penetration is the plane
defined by the innermost edges of
the unit frame.
AAMA 502-05 defined water leakage in a
similar manner. AAMA 502-08 makes a distinction
between water leakage through the
window product and water through the
perimeter joint and defines water penetration
as follows:
The penetration of uncontrolled
water beyond a plane parallel to the
innermost edges of the product and
that indisputably originates from
the fenestration product. Water penetration
attributable to the perimeter
joint shall be defined as uncontrolled
water that indisputably originates
at the joint.
AAMA 502-08 states that if the origin of
the leakage is inconclusive from testing,
then AAMA 511 should be used to identify
the cause of leakage.
These definitions require water to break
the innermost plane of the window frame or
product for water leakage to occur. Testing
where water accumulates on interior surfaces
without breaking the innermost plane
such as at meeting rails (Photo 1) and surfaces
of horizontal framing (Photo 2) do not
constitute leakage by these definitions.
These definitions do not coincide with
the typical end-user expectation (i.e., water
will not pass through window), which
should be paramount. Designers can specify
their own definition of water leakage in
the construction documents to clarify the
project objectives; this should be brought to
the contractor’s attention during bid-
Photo 1 – Water accumulation on a window meeting rail.
Photo 2 – Water ponded on the interior of the window against the glazing unit seal.
2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2 B U C H B E R G A N D S P A G N A • 9 7
Misconception No. 4: Specifying
AAMA 502 addresses water testing of
newly installed windows on a project.
Misconception No. 5: Regarding field
testing of windows, windows should
always be field-tested to a pressure that
is two-thirds of the laboratory testing
pressure, in accordance with AAMA 502.
Misconception No. 6: The manufacturer
is providing a warranty so that I am
covered for costs of all repairs.
Misconception No. 8: The spray rack
should not run for more than 15 minutes
during a forensic water test.
Misconception No. 9: The test apparatus
is applying too much water to the
building components.
ding/negotiation so he or she can make the
manufacturer aware.
The
2005 version of AAMA 502 indicates testing
of windows during the construction process
and prior to final acceptance by the owner.
The 2008 version applies only to windows
that were installed prior to building occupancy
and that are six months old or newer.
There is no suggestion of a performance
according to any accepted standard after six
months, only an evaluation of how windows
perform during the first six-month period.
The standard states that windows tested
after six months of installation or after
issuance of the occupancy permit shall be
tested using AAMA 511-08, Voluntary
Guidelines for Forensic Water Penetration
Testing of Fenestration Products—i.e., not a
performance evaluation against levels promoted
by the manufacturer.
As the title states, AAMA 511 is for
forensic testing used when investigating
windows that are known to leak. By referring
to a forensic standard, AAMA implies
that windows tested after six months of
installation shall not be tested to determine
if they perform to their original rating.
Instead, AAMA 511 requires that a forensic
investigator calculate differential pressures
based on actual wind-driven rain events,
which may be lower than pressures
required by AAMA 502. This method does
not adequately address reasonable performance
expectations for a window that is
less than a year old.
For fiel- testing, AAMA 502 specifies testing
the windows at a pressure that is two-thirds
that of the laboratory testing pressure,
reducing the testing pressure by one-third.
The reasons for the reduction include ambient
environmental conditions, installation
variations from laboratory to field, and
shipping/handling by subsequent trades.
While these are valid factors for a reduction,
the one-third reduction in the testing pressure
is without scientific basis; it is
unknown whether this represents the true
reduction in performance by the various
factors. It does allow the window to perform
at a lower level than the original certified/
rated performance of the window and
likely lower than specified for the design.
However, it is a misconception that
designers are bound by this rule. Designers
can specify that field-testing shall not
include the one-third reduction; and if so,
this should be brought to the attention of
both the contractor and window manufacturer.
Alternatively, designers can allow the
one-third reduction and test to the coderequired
design pressure by specifying a
performance grade with a design pressure
that is 150% of the code-required design
pressure.
Most window
manufacturer warranties will allow one
of the following three actions at the manufacturer’s
sole option: 1) provide a refund on
the windows, 2) provide a repair/retrofit of
the existing windows, or 3) provide new window
assemblies. The warranties usually do
not cover labor costs to remove/reinstall the
windows, including any associated trim,
interior wall finishes, or adjacent exterior
wall claddings that may require replacement,
which often are much larger costs
than the window product itself. The warranties
also likely do not cover consequential
damage to framing, finishes, or surrounding
materials.
FORENSIC TESTING
ASTM E2128, issued in 2001, provides
a guide to conduct water testing investigations
to identify leakage origins and paths.
ASTM E2128 identifies a systematic protocol
for investigating building leakage that
goes well beyond the requirements found in
the standards that govern certification testing.
The distinction between forensic testing
and certification testing needs to be understood
prior to performing a forensic investigation;
otherwise, confusion and invalid
conclusions can result.
Forensic testing is a systematic
approach to investigating leakage. ASTM
E2128 suggests the following steps:
• Review of project documents
• Evaluation of design concept
• Determination of service history
• Inspection
• Investigative testing
• Analysis
• Report preparation
It is often the investigative testing phase
where the procedures and methods
employed during certification testing are
confused with the “art” of water testing to
trace building leaks. A key distinction is
that ASTM E2128 applies to exterior walls
inclusive of fenestration products and not
just fenestration products themselves.
ASTM E2128 references ASTM E1105 and
AAMA 502, but these standards are not the
sole basis of a meaningful leakage investigation.
Overreliance on tests based on certification
procedures can lead to points of
contention among investigators, including:
• Test duration
• Testing apparatus volume, flow rate,
and intent
• Methodology
These are discussed below.
Certification
tests generally use 15 minutes of water
application. Duration for water testing is
not specified in ASTM E2128, but the standard
does recognize that testing durations
specified for product certification may not
be sufficient for diagnosing in-service leaks.
Factors such as wall construction, potential
length of leakage paths, water absorption of
materials, and internal storage capacity of
the components being tested all factor into
the required duration of any forensic water
test. Investigators need to realize such factors.
For example, mass masonry walls usually
take much longer to leak than nonabsorptive
systems such as a glass/metal curtain
wall because (in the absence of discrete
cracks or bridging elements) masonry has a
much higher internal moisture-storage
capacity.
Applying a water test duration that
duplicates certification tests of an individual
fenestration product during a forensic
investigation can provide false results and
prevent the investigator from identifying the
leakage path.
ASTM E1105
requires a water spray system (such as a
calibrated spray rack) to deliver water at a
rate of 5.0 gal/ft2 per hour. Investigators
often will identify Note 3 of Paragraph 6.2.4
in ASTM E1105 as a reason to ignore the
significance of leakage when subjected to a
9 8 • B U C H B E R G A N D S P A G N A 2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2
Case No. 1 Field Certification of
Windows Sometimes 15 Minutes Is Not
Enough
a rate of rainfall is misguided
and ignores the
purpose of the test
setup.
ASTM E2128 clearly
states that the intent of
the spray rack is to
“deliver a continuous
water film to the test
area, rather than to
simulate a particular
rain event.” We find that
placing a spray rack calibrated
to a rate of 5.0
gal/ft2 per hour approximately
1 ft. away from
the testing surface produces
a consistent film
for testing. In some
cases, though, site conan
operable casement over a fixed sash
(Photo 3). We followed testing Procedure B of
ASTM E1105, which applies a cyclic static
air-pressure difference across the window
test specimen. This test procedure requires
a minimum total time of pressure application
of 15 minutes, typically achieved by
three five-minute pressure cycles with a
one-minute period of zero air pressure
between cycles.
Much to the initial disagreement of the
observing contractor and window manufacturer’s
representatives, we chose to include
six cycles of pressure (for a total of 30 minutes
of pressure).
It was not until the last cycle of air pressure
that we observed leakage out of the
snap-in interior glazing stop at the sill of the
fixed sash (Photo 4). After conducting isolation
tests and disassembling the window,
Photo 4 – We observed leakage out of the
interior-glazing stop at the sill of the fixed
sash (at the location of the arrow).
Photo 3 – Window assembly consisted in part of
operable casements over fixed sash (arrow points
to casement).
spray rack calibrated in accordance with
ASTM E1105. The note reads:
The National Weather Service Technical
Paper No. 40 records that in
the contiguous 48 United States, the
greatest rainfall for a 1-hour period
is less than 5 inches. The rate of 5.0
gal/ft2 per hour specified in this test
method corresponds to a rainfall of 8
in./h unless otherwise specified.
The argument is that the area being
tested will never see 5.0 gal/ft2 per hour,
and for that reason, the testing volume and
rate are too severe. This note is confusing to
many and should be removed from the
standard. Correlation of the rate of spray to
ditions such as wind or
obstructions may prevent investigators
from placing the spray rack as suggested.
As a result, the spray rack may need to be
placed farther away from the surface, and
the pressure through the spray rack may
therefore need to be increased to achieve a
consistent film.
CASE STUDIES
–
–
A recent field-certification water testing
program on a high-rise residential tower in
southern California illustrated how sometimes
15 minutes of certification testing
simply is not enough.
The window configuration consisted of
we discovered the source of the leakage.
Water leaked through a hole in the frame
corner seal at one of the sill corners of the
operable casement. The water then flowed
to the head of the lower fixed sash, collected
in the hollow chamber of the head glazing
stop, flowed to the jamb glazing stop,
and then flowed down the jamb-glazing stop
to the sill-glazing stop at the bottom of the
fixed sash. Once water filled the chamber of
the sill-glazing stop, it began to flow onto
the interior sill and became visible to all
observers (however, leakage occurred the
moment water bypassed the frame corner
seal of the operable casement).
Had we chosen to test for a shorter
duration, this leakage path very likely
would have gone undetected.
2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2 B U C H B E R G A N D S P A G N A • 9 9
Case No. 2 Forensic Investigation of
Solid Masonry 15 Minutes Is Not
Enough
Case No. 3 Forensic Investigation of a
Curtain Wall/Sloped Glazing Assembly
With a Long Leakage Path
Photo 5 – Plaster damage at interior finishes.
Photo 6 – Water leakage through brick and terra cotta at a location of exfoliated
plaster.
– outer brick wythe over a terra cotta backup
– wall with interior plaster finishes.
We water-tested the wall for one hour
At an existing building, we observed before we replicated leakage through the
exfoliated plaster on the interior, indicative terra cotta (Photo 6) to the exfoliated interiof
water leakage (Photo 5). Exterior openings or plaster. The water soaked into the exteriaround
windows showed the wall construc- or brick through the mortar joints or
tion to be solid masonry consisting of an through cracks in the brick and then flowed
through joints or cracks in the backup terra
cotta, where it wetted the interior plaster.
The absorptive nature of the wall construction
required a relatively longer testing
duration than the 15-minute durations
noted in certification standards. Had we
tested for a shorter duration, we would not
have replicated the leakage.
–
You would think that diagnosing leakage
of nonabsorptive materials such as a
glass/metal curtain wall and sloped glazing
assembly would require relatively short
testing durations. Such was not the case at
an office building in the Midwest on a
recent investigation.
We tested the curtain-wall-to-slopedglazing
assembly for approximately four
hours (Photo 7) but did not observe leakage
to the interior. Upon arriving on site the
next morning, maintenance personnel told
us that approximately 20 minutes after we
finished testing and left the site the previous
day, leakage occurred in the exact location
that they always observe leaks during
rainstorms—40 ft. away from our test location.
We continued testing at the same location
and replicated this leakage within
about ten minutes.
Contrary to what the shop drawings
showed, removal of a metal flashing
between the base of curtain wall and head
of the sloped glazing revealed an unsupported
silicone sheet gutter that sagged
between the sloped glazing rafters (Photo 8).
Each valley of the silicone sheet gutter contained
approximately 3 in. of water after our
testing (Photo 9).
We determined that water bypassed the
metal flashing at a discontinuity (in the
location of our test area) and filled the valley
of the silicone sheet gutter directly
below. The gutter valley filled with water,
which then spilled into the adjacent valley.
This process continued until water reached
a defect in the gutter 40 ft. from our test
location, at which point it drained to the
interior of the building. Although we were
testing nonabsorptive materials, in the end,
it was a long leakage path that required a
prolonged testing duration.
RECOMMENDATIONS
When testing for certification and fieldtesting
of window products, the following
recommendations can help to alleviate con-
1 0 0 • B U C H B E R G A N D S P A G N A 2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2
fusion:
• Specify the required performance grade for
fenestrations rather than just the design
pressure.
• State the equation to be used to calculate the
air-pressure differential (i.e., percentage of
design pressure) to be used for water testing
rather than stating the exact pressure to use
in testing in order to ensure that the
approved window is tested to a water test
pressure corresponding to its certified performance
grade instead of a test pressure that
may be higher or lower.
• Confirm that the test reports provided by the
manufacturer include a test specimen equal
Photo 8 – Removal of base flashing
revealed an unsupported silicone
sheet gutter filled with water.
Photo 9 – Each valley of the silicone sheet
gutter contained approximately 3 in. of water
after our testing.
Photo 7 – Water testing of a curtain wall
and sloped glazing assembly.
2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2 B U C H B E R G A N D S P A G N A • 1 0 1
to or larger than windows used on
the project.
• Specify field-testing of installed windows
to evaluate the installed performance
of the windows rather
than relying solely on laboratory
tests.
• Provide a definition of water penetration
leakage in the specifications
that meets end-user expectations.
• Field-testing standards include an
effective one-third reduction in the
field-testing pressure compared to
the laboratory certification pressure.
Specify to perform field-testing without
taking the one-third reduction,
and make the contractor and manufacturer
aware; or specify windows
with a performance grade having a
design pressure that is 150% of the
code-required design pressure.
• Specify that field-testing be performed
within six months of installation
or that the windows shall perform
as rated until building occupancy.
• Do not rely on warranties as a safeguard
against poor design or poor
installation.
The specifications are part of the contractual
requirements of a project. The
designer can provide the safeguards stated
above in the specifications as a way to clarify
the required performance of the windows.
The designer should specifically identify
these requirements to the contractor to
make sure they are aware of these provisions.
When investigating a building for leakage,
investigators should use ASTM E2128
as the basis for the investigation and consider
the following:
• Evaluate the wall construction,
building materials, possible lengths
of leakage paths, and the history of
leakage to determine the appropriate
duration and test pressures for
testing of building components.
• Use a testing apparatus oriented
and pressurized to provide a consistent
film of water on wall surfaces.
SUMMARY
Understanding the intentions of testing
standards (both for certification and for
forensic testing) and the differences
between standards associated with product
certification and standards associated with
forensic investigation of leakage is essential
in eliminating some common misconceptions
regarding water testing. By realizing
these misconceptions and addressing them
in project documents and during testing
protocols, confusion, frustration, and inaccurate
results can be avoided.
FOOTNOTES
1. H.H. Sheldon, A Study of Glass
Jalousie Windows Under Hurricane
Conditions, University of Miami,
Coral Gables, FL, 1952.
2. S.D. Svenson and R. Wigen, ASTM
STP 251, “Testing Window Assemblies,”
American Society for Testing
and Materials, 1959, pp. 36-38.
3. A.A. Sacknovsky, ASTM STP 552,
“Testing for Water Penetration, Window,
and Wall Testing,” American
Society for Testing and Materials,
1974, pp. 31-35.
4. Aluminum Window Manufacturers
Association, “Specification Performance
Test Program and Procedures,”
Aluminum Window Manufacturers
Association, AWMA Bulletin
100, No. 1959.
1 0 2 • B U C H B E R G A N D S P A G N A 2 7 T H RC I I N T E R N A T I O N A L C O N V E N T I O N A N D T R A D E S H OW • MA R C H 1 5 2
0 , 2 0 1 2