Horizontal Above-Grade Waterproofing

Horizontal Above-Grade
Waterproofing
Kar im Allana, RRC, RWC, PE
Allana Buick & Bers, Inc .
990 Commercial Street, Palo Alto, California 94303
Phone: 650-543-5600 • Fax: 650-543-5625 • E-mail: bd@abbae.com
S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5 A l l a n a • 6 9
7 0 • A l l a n a S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5
ABSTRACT
Podium deck waterproofing assemblies present significant design and liability challenges
to building owners and designers. This paper will review Allana Buick & Bers Architectural
and Engineering’s (ABBAE’s) waterproofing assembly design considerations, examine widely
used assemblies, and provide lessons learned through real-world case studies. The presentation
will cover podium deck waterproofing and highlight design considerations through a
series of forensic investigation case studies. Each study will illustrate a failed waterproofing
system and identify the causation and appropriate remedial solutions. Through each forensic
case study, we will review alternate podium waterproofing systems, their anticipated
longevity, typical installation costs, and risk factors for designers.
SPEAKER
Karim Allana, RR C, RWC, PE — Allana Buick & Bers, Inc .
Karim P. Alla na is the CEO and senior principal of Allana Buick & Bers, Inc., an
architectural-engineering firm specializing in the building envelope, sustainable construction,
and construction management services. Allana has a BS in civil engineering and is a
licensed professional engineer in four states. He has been in the A/E field for 30+ years,
specializing in forensic analysis and sustainable construction of roofing, waterproofing, and
the building envelope. He is a frequent speaker and presenter at professional forums.
S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5 A l l a n a • 7 1
INTRODUCTION
This paper is limited to protected-type
waterproofing membrane assemblies such
as plaza decks, split slabs, and concrete over
wood deck assemblies. Trafficable waterproofing
membrane systems like deck coatings
and traditional protected roof assemblies
such as inverted roof membrane assemblies
are excluded from this discussion.
Plaza decks are defined as exposed
pedestrian surfaces constructed over parking
or other occupied space, often built
over structural concrete. The International
Building Code (IBC) defines garages as a
type of occupancy, so technically, plazas
built over garages are over occupied space.
Plaza decks can have many features such
as driveways, pools, lawns, large mature
trees, terrazzo and concrete toppings, glass
screens, pavers over
gravel or pedestals,
planters, etc.
Also discussed
are protected waterproofing
membranes
over wood decks surfaced
with many of
the same types of
permanent surfaces,
such as concrete topping
slab, tile planters,
etc. Typical Type
V Podium assemblies
are four to five stories
of wood-framed
construction over
a concrete podium,
commonly used in
California and other
states for multifamily
construction.
These elevated wood
assemblies also have
exposed, horizontal
waterproofing with
hard-wearing surfaces,
like concrete,
which are integrated
to exterior wall
claddings such as
masonry, stucco, siding,
dimension stone, etc. Because these
assemblies are surfaced with permanent
landscape and hardscape features and integrated
with doors, ramps, and cladding
elements, the wood-framed, waterproofed
assemblies are also built with a life expectancy
equivalent to the life of the building.
While properly designed and constructed
plaza deck assemblies such as these
can and do last the lifetime of the building
(50+ years), poor-quality construction often
causes premature failures within 10 years
and requires very expensive repairs, costing
well over $100/SF. The focus of this presentation
is to explore the actual performance
of these types of assemblies, explore causes
of premature failures, and learn what seems
to work and what to avoid. Explored are
failures related to sloping issues, drainage
issues, uses of galvanized versus stainless
flashings, issues with integration to sliding
doors and man doors, and issues related
to degradation and failure of waterproofing
membranes under certain conditions.
COMPONENTS OF PROTECTED
ASSEMBLIES
There are several base components of
any protected system, as demonstrated
in Figures 1 and 2: the wearing or traffic
surface, a protection layer that sometimes
doubles as isolation and drainage layer, a
waterproofing membrane, and the structural
deck or substrate. Additional systems
may be necessary, and the order of the systems
can vary, depending on the purpose
and design of the deck, but the base components
remain consistent across various
deck types.
STRUCTURAL SYSTEMS
Concrete
Concrete slabs are the most common
structural substrate for podiums.
Slabs are either reinforced with mild
steel or post-tensioned steel (Figures
3 and 4). The advantage of posttensioned
steel is reduced thickness—
consequently, less concrete and less
cost. However, the labor associated
with tensioning cables is higher, and
overall, the cost of post-tensioned
slab can be competitive, depending on
subcontractors.
Horizontal Above-Grade
Waterproofing
Figure 2 – Typical podium under
construction with waterproofing
complete inside of planters. Due
to the heavy construction traffic,
waterproofing membrane is
phased and installed last, before
the wearing course.
Figure 1 – Protected
membrane components.
7 2 • A l l a n a S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5
Often, Type V structures
are built with wood
framing over concrete
podium. Commonly,
elevated exterior wood
structures are waterproofed
and may have
permanent features
such as protective concrete
topping slab,
public walkways, stair
landings, exercise equipment,
private decks,
bicycle racks, planters,
and other features similar
to podium decks.
DECK SLOPE AND
PROPER DRAINAGE
One of the keys to
lifetime performance of
horizontal, protected
waterproofing assemblies is proper design
of slope and drainage. Because these types
of assemblies have a substrate and topping
surfaces, both layers of surfaces require
proper design and construction of slope.
Also, since these assemblies not only drain
from the surface but also from within
the protected assembly, it is important to
understand the mechanism of drainage,
especially at the membrane level. Proper
construction of both slope and drainage
are critical to the long-term performance of
these assemblies.
The IBC requires slopes under the waterproofing/
roofing membrane to be a minimum
of ¼ in. per foot. In some geographic areas,
we commonly see that the common practice
is to not slope the structural slab, and to
provide slope only in the topping material,
which may not meet the intent of the code.
In my opinion, the slope requirement is
primarily for the waterproofing substrate.
IBC Section 1402.3 states, “Balconies, landings,
exterior stairways, occupied roofs and
similar surfaces exposed to the weather and
sealed underneath shall be waterproofed and
sloped a minimum of ¼ in. unit vertical in
12 units horizontal (2% slope) for drainage.”
Slope is best achieved by being built
into a structural deck because it becomes
a permanent building feature. Building the
structural level and adding slope thereafter
adds additional cost and introduces potential
for mistakes and premature failures.
Sometimes, porous topping surfaces
such as pavers are built with no slope to
achieve a level, usable floor. When a level
surface is desired, pavers, natural stone,
gravel, and other porous concrete surfaces
are used to allow water to completely drain
through the topping material and travel on
the membrane level. In those instances, it
is important to install primary and overflow
drains at the membrane level to accommodate
the extra flow of water.
On less-porous surfaces such as concrete
topping slab, the topping surfaces
have to be sloped 2% to drain, in addition
to the substrate being sloped to drain as
well. Such surfaces require main drains to
be located at the topping level and secondary
weeps to be located at the membrane
level. Proper functioning of both drainage
methods is very important to the success of
plaza and elevated outdoor type assemblies.
It is easiest to achieve slope via castin-
place concrete because it can be directly
cast into the deck surface, and is easier to
resolve slopes for decks with complex shapes
and geometry. Slopes can also be achieved in
plywood decks by tilting the framing members
to slope and installing plywood crickets
for cross slopes. Sometimes, wood joists are
ripped on the top to achieve slope, and the
bottom of the joist can remain level. Splitslab
systems with steel pan decks are sloped
similarly, sloping the steel framing and steel
deck and/or sloping the concrete slab cast
on the steel pan. Using tapered insulation
to create gradient is not recommended for
split-slab assemblies because it limits the
types of waterproofing membranes that can
be used, it can be more readily damaged,
and if punctured, it can allow water to travel
within the assembly.
With new construction, slope is often
indicated on architectural drawings but
sometimes neglected in structural drawings.
Sometimes, podium slopes are indicated
on civil drawings instead of structural
or architectural drawings. In new
construction projects, proper review and
coordination of slope on civil, structural,
architectural, and landscaping drawings is
important. It is important to check for conflicts
among the various disciplines and to
make sure proper slopes are clearly shown
and the contractor(s) are aware of the
requirements. On new construction projects,
the best place to show slopes is on the
structural drawings (in addition to architectural)
because the slab/structure needs to
be engineered by the structural engineer.
Also, framing and structural subcontractors
often build from information conveyed
on the structural drawings and mistakenly
ignore slopes shown on architectural draw-
Figure 3 – Podium slab
with tensioning cables
and reinforcing steel.
Figure 4 – Post-tensioned reinforced concrete slab under
construction. Also common in steel structures is steel
pan deck, with mild steel-reinforced concrete. Prior to
the adoption of reinforced concrete, the two-way flat
plate and one-way slab-over-beam systems were common;
however, they are not used much anymore.
S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5 A l l a n a • 7 3
ings, so it’s important that the structural
and architectural drawings are in sync. It’s
also important to review the structural substrate,
as some lightweight concrete substrates
and sloping courses are not suitable
for waterproofing membranes.
DRAINS
A typical plaza deck drain, as pictured
in Figure 5, consists of a top-mounted
inlet drain for surface drainage in addition
to a secondary inlet
located at the membrane
level for membrane
level drainage.
The podium drain
body typically has an
adjustable height and
threads for proper
installation and function.
The location of
the drain is important
in preventing ponded
water—both on the
wearing surface and
at the waterproofing
membrane level.
Standing water at
the membrane level
can damage certain
types of waterproofing
membranes and cause the system to fail.
Certain waterproofing membranes, such as
asphalt-modified polyurethane membranes,
are very susceptible to damage from standing
water.
Drains must have weep ports at the
membrane level to allow the water to drain
off the surface of the waterproofing. A typical
plaza drain will handle the majority of
the drainage and water runoff from the top
inlet. However, podium assemblies that
have pavers on
gravel/sand bed
or pedestal pavers
should have
their primary
drains at the
structural slab
level because
the majority of
the water will be travelling on the membrane
level. If primary drains are installed
at the membrane level, the concealed drains
should be identified on top with pavers that
are marked to indicate where the drains
are located. Overflow drains should also be
provided to prevent flooding, and subdrains
should be maintained regularly to prevent
clogging.
SUB-SLAB DRAINAGE
A drainage board or air gap between
the surface and waterproofing layers serves
as drainage and occasionally as additional
protection for the overburden during construction.
Its main purpose is to conduct
subsurface water to the drain, gutter, or
edge, and provide mechanical protection
from the overburden.
Manufacturers of the waterproofing
materials will often allow the designer to
omit the drainage layer. However, drainage
layers work best when the substrate
is built with proper drainage, and when
pathways towards the drains and gutters
are not obstructed or blocked. Lack of slope
or reverse slope leads to long-term standing
water on the waterproofing membrane.
The drain board cores are made from polypropylene,
polystyrene, and polyethylene,
which are vapor barriers or vapor retarders.
The vapor barrier nature of the drain
board can trap standing water, prevent it
from evaporating and drying, and can lead
to damaging some types of waterproofing
membranes due to being immersed
in water. The same is true for some types
of protection boards and XEPS insulation
boards, as both can restrict drying if proper
slope and drainage are not present.
Figure 5 – Typical plaza drain configuration.
Figure 6 – Elevated public walking surfaces on wood
deck with colored concrete and porcelain tile wearing
surfaces, integrated with handrails, doors, exterior
walls, elevators, etc.
Figure 7 – Typical podium with concrete topping and
pavers, shade structures, fences, planters, trees, pool,
spa, light pole, free-standing walls, etc. integrated with
building walls, fences, doors, pools, and planters.
7 4 • A l l a n a S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5
INSULATION LAYERS
Insulation is usually applied over the
waterproofing membrane or under the
structural deck, as it should be independent
of the two. In cold or high-humidity
climates, it is best to locate insulation above
the waterproofing membrane to prevent
freeze/thaw damage.
For plaza surfaces that receive live loads
such as foot and vehicular traffic, it is
important for the insulation layer to have a
high compressive strength. Due to immersion
in water, it is important that insulation
materials such as closed-cell extruded
materials be used as opposed to expanded
polystyrene, to prevent water absorption.
The most typical extruded polystyrene foam
has a compressive strength that ranges
from 20 to 100 psi and an R-value of 5.0/
inch. The required compression strength
of the insulation board will depend on the
dead load as well as the type of live load
(foot traffic versus vehicular traffic). The
board needs to be protected from the sun
during construction to prevent permanent
damage. Foamed glass is another insulating
material that is not as commonly used, but
it does have a high compressive strength
(90 psi), R-value (3.44/inch), and melting
point (1,800ºF), making it very dimensionally
stable.
WEARING SURFACE
The upper-most layer (also referred to as
the wearing, pedestrian, or traffic surface) is
usually composed of concrete or pavers that
are subject to traffic by pedestrians, vehicles,
or both (public walkways – Figures 6
and 7). This is the first line of defense against
damage to the membrane and assembly
and can be the
primary means
of drainage for
the assembly.
Concrete topping
is often sloped to
shed the majority
of surface water, but significant amounts of
water can get under the concrete.
WATERPROOFING MEMBRANES
The waterproofing membrane serves as
the main water barrier, and the selection
of the system is critically important to the
success of the project. Our forensic experience
has shown that standing water is often
present on the membrane for months or—in
some cases—all the time. Therefore, it is
important that the selected membrane can
handle continued immersion in water. While
drainage mats may help move the water
towards the drains, when proper slope is
not present, drainage mats can serve as an
unwanted reservoir of water standing on the
membrane. The
standing water
does not dry for
a very long time,
and in freezing
climates, can
freeze, causing
freeze/
thaw damage.
Generally, even
without freezing
climate, standing
water accelerates
the decay
and degradation
of waterproofing
membranes, and some membranes are
more affected by it than others.
Some of the most common fully adhered
membranes in use today are hot-fluidapplied
rubberized asphalt (HRA), modifiedbituminous
membranes (Figure 8), modifiedbitumen
sheets, and PVC (Figure 9). Many of
these systems have been in service for over
40 years and are still performing well today.
Asphalt-modified polyurethane coldfluid-
applied membranes (Figure 10) tend
to be some of the least expensive products,
very simple to install, and requiring little or
no equipment, making them very popular.
However, asphalt-modified polyurethanes
have a very poor track record and tend to
suffer from premature failure due to higher
Figure 8 – Combination self-adhered modified-bitumen
membrane with fluid-applied waterproofing.
Figure 10 – Asphalt-modified polyurethane.
Figure 9 – PVC membrane.
S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5 A l l a n a • 7 5
permeability under standing water, which
causes premature failure, swelling, and
blistering (Figure 16).
Another common membrane selection is
modified bitumen, which can be adhered,
torched, or fluid-applied. Some self-adhesive
SBS membranes use a combination of cold,
asphalt-modified polyurethane, deck prep,
and self-adhered sheets (Figure 8).
Hot-Applied Fluid Systems –
Rubberized Asphalt (HRA)
Based on my forensic review of many
different types of waterproofing systems,
HRA is one of the most reliable waterproofing
membranes in service (Figure 11
and 12). Due to its reliability
and relatively
low cost, HRA systems
have become very popular
in new construction
and are often used
on large podium-type
projects. HRAs require
a specialized doublejacketed
kettle to heat
the asphalt; and for
remedial construction,
the smell of asphalt is
sometimes a deterrent
for occupied buildings.
HRAs have the ability
to be phased and joined
seamlessly.
Due to its
recent surge
in popularity
and low failure
rate, new
material manu
f a c t u r e r s
have jumped into the market. Many waterproofing
contractors that historically did
not have the required hot-jacketed kettles
have bought new equipment and are now
offering this system. Overall, more competition
has driven the cost of the system lower,
bringing the price on
par with cold-applied
asphalt-modified polyurethanes.
Unreinforced HRA
membranes typically
have had more failures
and installation defects
such as pinholing and
leaks. These membranes
are unable to bridge cracks as
effectively and can fail as the underlying
concrete or substrate shrinks and cracks or
moves over time; whereas, polyester, fabricreinforced
HRA systems are more forgiving
and redundant and have fewer construction
issues and, if properly constructed, can last
for the life of the building. Therefore, it is
recommended to provide reinforced systems
for long-term performance.
Nevertheless, there are multiple challenges
to successfully installing HRA, the
four most common being: adhesion failure
to concrete, material and flashing build-up
Figure 11 – Adhesion of membrane with sheet neoprene.
Figure 12 – Typical HRA application
Figure 14 – Waterproofing around
reinforcing steel.
Figure 13 – HRA challenge –
flashing buildup.
7 6 • A l l a n a S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5
at base flashings, flashing around rebar,
and substrate acceptance.
HRA is typically about ¼ in. (215 mils)
thick, with the embedded neoprene at
around 60 mils. The protective flashings
used at seams and joints are about 110-180
mils, so when overlap occurs (as depicted in
Figure 13), the membranes can build up to
about ¾ of an inch, resulting in problems at
the base flashing of doors and storefronts.
Waterproofing around rebars is always
challenging because the membrane prevents
concrete from bonding directly to the
reinforcing steel (Figure 14). Proper care
needs to be taken to detail the reinforcing
steel to both waterproof the area and allow
the rebar to serve its primary
function.
And finally, HRA does not
always properly adhere to the
substrate, so regularly performing
peel tests for testing
adhesion is important. It is
also important to check for
moisture escaping the structural
deck, using a mat test
(ASTM C4263) to check for
capillary moisture in concrete
and by mopping some hot
asphalt to see if it fizzes or
pinholes.
Manufacturers of HRA
don’t offer specific guidelines
for adhesion testing, but the
most common method is
using a reinforcing fabric or
neoprene to perform a pull
test (Figure 15). Pass/fail is
best defined with cohesive/
adhesive failure. Waterbased
primers are always
tricky and not as reliable
as solvent-based primers.
Water-based primers
are sensitive to cold
temperature and can reemulsify
with high alkalinity
and water. Solventbased
primers are also
sensitive to application
rates, curing time, and
temperature.
Cold-Applied Fluid
Systems
A common coldapplied
fluid system
material is asphalt-modified
polyurethane. Due to its ease of installation
and relatively low cost, it has been
a very popular option for protected waterproofing
membrane assemblies. However,
our forensic experience has shown that
most fluid-applied polyurethane-type coats
don’t work very well in plaza-type applications;
and where there is constant immersion,
water may be present. Typical failure
modes are membranes absorbing large
amounts of water, and swelling and waterfilled
blisters.
Most waterproofing manufacturers list
very low water-absorption rates after testing
for just three days of immersion. However,
water may be present on the membrane for
months at a time—sometimes for the life of a
system. Impact of long-term standing water
can be very different than a standard threeday
test. Some membranes absorb more
than 100% of water per weight of the material,
excessively swell, and look like the brain,
often called the “brainiac” effect (Figure 16).
Most polyurethane membranes have
some level of permeability, which ultimately
causes them to transmit water through the
membrane and result in failure. The most
common and dramatic types of permeability
failures are large water-filled blisters over
concrete decks and damage to sheet metal
flashings though permeability. It is best to
select a membrane that has a perm rating
of 0. But the permeability of the membrane
must always be less than the permeability
of the substrate. Most concrete slabs have a
permeability of 3.2 perms per inch. A 10-in.
slab would have an approximate permeance
of 0.32.
It is also important to check the “wet
permeability” any type of fluid-applied membrane
being used in a protected assembly.
Manufacturers of fluid-applied membranes
often publish permeability rates that only
measure permeance with vapor transmission,
and that test is not valid. The membrane
is subjected to standing water; therefore,
the proper test methods are those that
measure permeance through standing water
that closely mimics the actual conditions.
If the wet permeability of the waterproofing
membrane is higher than the permeability
of the slab, water will
start to accumulate on the
concrete surface and not dry
through the slab fast enough.
This leads to moisture-filled
blisters. Moisture-filled blisters
mix with salts that are
naturally occurring in concrete,
and the saline solution
creates an osmotic action and
draws fresh water through
the membrane. In reinforced
concrete slabs, we have seen
these blisters grow to the size
of baseballs and footballs.
Unreinforced membrane blisters
are generally smaller (the
size of a dime or silver dollar)
before they burst or break.
Brian Hubb, PE, conducted
permeability tests of
polyurethane membranes
and published a paper in the
Figure 15 – Mat test (right) and various samples for pull
test about to be performed.
Figure 16 – Modified polyurethane membrane is swollen and
looks like the texture of a brain and is referred to as “brainiac.”
Prolonged immersion of the membrane results in swelling and
water absorption far beyond the manufacturer’s published results.
S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5 A l l a n a • 7 7
Proceedings of the RCI Building Envelope
Technology Symposium of October 2009.
Hubbs reported permeability of modified
polyurethane membranes with the inverted
wet cup method to range from 0.20 to 0.50
perms with membrane thicknesses ranging
from 30 to 150 mils. In comparison, most
podium slabs are 18 in. or thicker with
permeability of 0.18. This disparity allows
water to accumulate
on the slab and
start the osmotic
cycle. As water
slowly permeates
through the membrane
and doesn’t
dry through the slab
fast enough, it starts
to accumulate on
the surface of the
slab. Salts from the
concrete mix with
the water to form an
ionic solution. Due
to the ionic potential
across fresh water
(standing on top
of the membrane)
and salty water (on
the surface of the
concrete), the membrane
allows the
fresh water to be
drawn through and
create water-filled
blisters (Figure 17).
While permeability
can allow for
water to dry through
the membrane, it
can be detrimental to roofing and horizontal
waterproofing applications.
Due to high permeability of these membranes,
we have also seen high levels of moisture
permeation and
damage in plywood
substrates and sheet
metal flashings.
Standing water on
modified polyurethane membranes can raise
the moisture levels in plywood and cause
corrosion of galvanized steel flashings (Figure
18). We have seen moisture content above
19% in many plywood substrates due to
standing water and corrosion on sheet metal
through 90+ mils of membrane thickness.
WATER TESTING AND WARRANTIES
Warranties do not prevent roofs and
plaza decks from leaking. The biggest difference
between roofing and waterproofing
warranties is that membrane manufacturers
exclude overburden replacement. While
most membranes cost about $10/SF or
less, podium replacements can range from
$60/SF to $200/SF, depending on the overburden.
Most warranties exclude removal
and reinstallation of the overburden. In
a plaza deck, 90% or more of the cost of
replacement is the overburden and removal
of doors, window walls, and façades to
perform the proper tie-in. Most membrane
manufacturers only warrant the material
cost, which could be less than 2% of horizontal
waterproofing replacement cost.
Water testing only tells part of the picture.
Just because a horizontal waterproofing
assembly is shown to be watertight
from an industry standard ASTM D5957
test (depicted in Figure 19), doesn’t mean a
system is built watertight for the life expectancy
of the system. Most systems may pass
the ASTM 5957 test even if the system had
a slow leak. Electronic field vector mapping
Figure 17 – Osmotic water-filled blisters.
Figure 19 – ASTM D5957 standing water testing.
Figure 18 – Corrosion of galvanized sheet metal flashings
under 90 mils of fluid-applied membrane.
7 8 • A l l a n a S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5
(EFVM) is a better way to test for watertightness
because often the penetrated water
doesn’t leak and drip through the ceiling
in a regular flood test. Conversely, some
leaks that can be detected through a flood
test cannot be detected through an EFVM
test under conditions that require elevated
water, such as expansion or seismic joints.
Despite both methods of testing, premature
failures can occur years after the system
is in place due to inherent
construction defects or
membrane failure. Water
tests are no substitute for
proper design and quality
assurance.
Electronic Field Vector
Mapping (EFVM)
EFVM is a nondestructive
testing method that
uses electricity to precisely
test the waterproofing
membrane for leaks. EFVM
is preferred over other traditional
testing methods
because it is nondestructive
and the membrane can
be tested without removing
the overburden or ballast.
It can be used on both hotand
cold-liquid-applied
membranes, coatings, and
other sheet membranes in
addition to metal, concrete,
and wood substrates, provided
a grounding grid is used with the
wooden deck.
Certain types of membranes are incompatible
and can’t be tested with EFVM.
Leaks around drains and penetrations can
also be difficult to detect.
The test works by measuring the difference
in electric potential between charged
water, the nonconductive membrane, and
the grounded, positively charged substrate.
First a very thin layer of water, which serves
as a conducting medium, is applied to the
surface of the membrane, and a low-voltage
pulse (40V for one second every three seconds)
is introduced into the water using
a wire loop that encircles the perimeter of
the tested area. The current flow is measured
using a potentiometer (Figure 20). If
the membrane is watertight, the electricity
is isolated to the water; however, if there
are leaks, the electricity will be pulled
towards the positively charged substrate.
The resulting electrical contact and exact
point of entry can be determined by using
the potentiometer to follow the electrical
pull. Because EFVM is so sensitive, leaks as
small as a pinhole can be detected (Figure
21). In some circumstances, the leads and
wire loops can remain below the overburden
to allow for easier future testing.
One of the drawbacks of EFVM is that
not all membranes can be tested due to
their conductive nature. Membranes that
contain carbon black or aluminized coatings
with modified-bituminous membranes do
not resist enough electricity and thus are
not testable using EFVM.
FORENSIC CASE STUDIES
The best way to learn is by making
mistakes or by studying failures caused by
others. ABBAE has been very fortunate to
have had the opportunity to study hundreds
of failed podium decks due to numerous
types of design, construction, and material
failures. There are several types of failures
and forensic studies presented, including:
1. Drainage or slope issues resulting in
unpredictable leaks to the interior
2. Failures due to high permeability
rates in waterproofing membranes in:
a. Concrete substrates
b. Wood substrates
c. Sheet metal flashings
3. Issues related to sub-slab drainage
such as weep holes being clogged
and discontinuous drain board
4. Impact from root damage due to
Figure 20 – Using
potentiometer to discover
source of leaks.
Figure 21 – Pinhole
leak discovered
using EFVM.
S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5 A l l a n a • 7 9
improper root protection
5. Impact from construction activities
due to lack of protection
6. Use of galvanized sheet metal versus
stainless steel
7. Failure due to improper surface
preparation
8. Failure due to water-based primers
used under fully adhered membranes.
CASE STUDY 1 – IMPORTANCE OF
PROPER SUBSTRATE SLOPE
We were studying an ongoing leak in
a concrete podium-level living area of a
large (over 250-door) apartment structure.
The sliding doors to the private decks were
covered with 10-ft. overhangs. Since the
doors were not typically subject to winddriven
rain, we wondered why there was considerable
damage inside at the base of the doors.
The water testing consisted of placing
water on the pavers 20 ft. away from the
sliding door, in the weather-exposed area.
Although the water was never placed anywhere
near the sliding glass doors, and the
surface water never made it to the door or
the building walls, we observed leaks inside.
We observed that the water began to emerge
from the doorjamb, both outside and inside.
The leak mechanism and mode of travel of
water are depicted in Figures 22 and 23.
The weeps at the waterproofing membrane
level were small relative to the drain,
and the weep holes were clogged, allowing
water to mostly drain off the topping
slab and build a hydrostatic head at the
surface. The control joints in the concrete
and perimeter joints served as a conduit for
water, filling up and creating hydrostatic
head at the level of the topping slab.
Lesson learned: It is important to allow
the sub-slab to properly slope to drains and
that the weep holes be clear to allow for
proper drainage. It is also important to not
have any topping slab surfaces higher than
the interior height. In the absence of robust
sub-slab drainage, water can build a hydrostatic
head equal to the highest surface of
the topping slab.
CASE STUDY 2 – IMPORTANCE OF
PROPERLY INSTALLED DRAINAGE
MAT
This case study depicts damage and failure
caused by standing water on a modified
polyurethene membrane (Figure 24). The
hospitality project featured private balcony
decks, built with plywood substrate sloped
to drain in one direction with an internal
stainless-steel gutter. The waterproofing
membrane has a drainage board/mat with 2
in. of mortar bed and travertine tile.
Generally, the deck was uniformly
sloped towards the gutter. However, due to
the drainage mat being discontinous across
the middle of the deck, water ponded on the
membrane in pockets of some areas. At the
standing water areas, water stood on top
of the membrane for long periods of time.
While the drain mat can facilitate the flow
of water, in a standing water condition, it
can act as a vapor barrier and not allow
the water to dry through the surface of
the slab. Water standing on the membrane
for extended periods of time caused the
asphalt-modified polyurethene membrane
to swell “brainiac” and fail (Figure 16).
CASE STUDY 3 – ADHESION
TESTING AND FAILURE MODES
On a new-construction, podium-style
apartment project with over 1,500 doors
and several acres of podium deck, we
observed sporadic failures in HRA membrane
adhesion. The failures modes were
multiple and complicated.
One of the failure modes was the
improper cure of the structural slab surface,
which resulted in a very soft surface,
Figure 22 – Water flows from the door
even though it never flows to it.
Figure 23 – Water flows under the
topping slab and appears at door jamb.
8 0 • A l l a n a S y m p o s i u m o n B u i l d i n g E n v e l o p e T e c h n o l o g y • No v e m be r 2 0 1 5
making it easy to fail cohesively (Figure 25).
The other type of adhesion failure was due
to a water-based primer that didn’t properly
cure due to low temperature or had emulsified
in the presence of water (Figures 26 and
27). Even though membranes passed adhesion
tests before application, a month or two
later, they failed to adhere.
The standard method that we use for
testing for adhesion is similar to the ASTM
D6862 laboratory test, but modified for use
in the field. A flexible fabric is embedded in
HRA, and an attempt is made to peel the
membrane by a 90-degree pull test. The
qualitative information
that we are
looking for is primarily
an adhesion
(not cohesion) mode
of failure, rather
than the actual peel
strength.
CONCLUSION
Unlike traditional roofs, podium waterproofed
assemblies are built to last the
life of the building because it is extraordinarily
expensive to fix them. Replacing
a failed waterproofing
system can
require removal
and reconstruction
of many exposed
structures, such as
stairs, free-standing
walls, planters, topping slab pavers, exercise
equipment, trails, driveways, walking
paths, etc.
Successful podium waterproofing
assembly requires proper drainage, both
at the surface level and at the membrane
level; proper selection of materials that can
handle long-term exposure to water and
moisture; proper construction and design;
and testing of the assemblies.
Figure 24 – On this wood-framed balcony deck, the drain
mat was discontinuous down the center of the deck.
This resulted in water not properly flowing and ponding
on the upslope side of the discontinuity.
Figure 26 – This sample exhibits failure in adhesion
due to a water-based primer that was used in
the application that was not curing due to cold
temperature.
Figure 27 – Adhesion failure of cold-fluidapplied
membrane to stainless steel flashing
due to lack of proper priming and prep work.
Figure 25 – Interestingly, the failure mode is in
cohesion of the concrete, as opposed to adhesion
of the membrane. The concrete slab surface had
improperly cured and was soft and chalky.