Stone-Faced Aluminum Honeycomb Composites – Quality Control and Testing

Stone-Faced Aluminum Honeycomb Composites —
Quality Control and Testing
Daren S. Kneezel, RA; and Michael J. Scheffler, PE
Wiss, Janney, Elstner Associates, Inc.
330 Pfingsten Road, Northbrook, IL 60062
Phone: 847-272-7400 • Fax: 847-480-9534 • E-mail: dkneezel@wje.com
2 9 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 rc h 2 0 – 2 5 , 2 0 1 4 K n e e z e l a n d S c h e ff l e r • 1 5 5
Abstract
Stone-faced aluminum honeycomb composite panels are engineered building products
that provide an alternative to traditional stone cladding and are increasingly being used as a
lightweight exterior building cladding. Stone-faced composites use only a thin stone veneer,
which minimizes the influence of the stone on mechanical properties and variability of the
composite. Based on independent laboratory testing of certain strength and durability characteristics
of stone-faced composite panels from different manufacturers, several variables
that can affect the performance of these composites will be provided, along with guidance
for evaluating and specifying stone-faced composite panels.
Speakers
Daren S. Kneezel, RA – Wiss, Janney, Elstner Associates, Inc.
Since joining WJE in 2003, Daren Kneezel has been involved in projects ranging from
investigation of high-rise building façades to laboratory testing of stone and other construction
materials. He has also performed structural and finite element analyses and developed
repairs for thin-stone cladding and its anchorage systems. Kneezel has performed laboratory
testing of various materials with a concentration on stone (marble, limestone, granite,
slate, travertine, and composite stone panels) to evaluate structural performance and to
determine compliance with project specifications and industry standards. This includes
testing of materials in accordance with the standards of ASTM International (ASTM).
Michael J. Scheffler, PE – Wiss, Janney, Elstner Associates, Inc.
Since joining WJE in 1980, Mic hael Scheffler has been involved in more than 1,000
investigations of deterioration and distress in buildings and other structures. The investigations
have included high-rise exterior façades of limestone, marble, granite, masonry, and
terra cotta. He has extensive experience in the investigation and testing of distressed thinstone
façades. Scheffler has performed laboratory and in-situ testing of building assemblies
related to stone material performance and structural performance and has performed longterm
monitoring and instrumentation of distress on several types of structures. He has also
given professional seminars on building envelope maintenance and repair.
1 5 6 • K n e e z e l a n d S c h e ff l e r 2 9 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 rc h 2 0 – 2 5 , 2 0 1 4
INTRODUCTION
Stone-faced aluminum honeycomb
composite panels are engineered building
products that provide an alternative to traditional
stone cladding and are increasingly
being used for exterior building cladding.
Stone-faced composite panels provide a
very similar aesthetic to traditional thinstone
cladding, with the advantage of being
lighter weight, which can result in a lower
installed cost.
While the mechanical properties of
many stone types are intrinsically highly
variable, stone-faced composites use only
a thin stone veneer, which minimizes the
influence of the stone on composite panel
variability. Understanding and controlling
the manufacturing variables of these composites
related to adhesive, aluminum honeycomb,
and honeycomb facing in order
to produce a uniform composite with the
desired behavior and performance requires
significant research and development by
the composite manufacturer during product
development, as well as quality control during
production.
Based on evaluation of stone-faced composite
panels from different manufacturers,
this paper will describe several variables
that can affect the performance of stonefaced
aluminum honeycomb panels and will
summarize independent laboratory testing
that has been performed by the authors
to measure certain strength and durability
characteristics of stone-faced aluminum
honeycomb products. Based on the test
results and findings, guidance is provided
for testing and evaluating specific physical,
mechanical, and fabrication aspects
of stone-faced composite panels prior to
their procurement, as well as during their
fabrication.
PRODUCT
DESCRIPTION
Aluminum Honeycomb
Aluminum honeycomb
is typically fabricated using
the expansion method, in which sheets of
aluminum are stacked and adhered with
staggered lines of adhesive. After curing, the
stack of adhered aluminum sheets is sliced
to the desired thickness and expanded
to form a honeycomb. Facings are then
adhered to the expanded honeycomb to
form a laminated honeycomb.
Stone-Faced Aluminum Honeycomb
Stone-faced aluminum honeycomb panels
are a relatively recent engineered building
product that provides a lightweight alternative
to traditional stone cladding. These
panels weigh approximately 3.5 pounds per
sq. ft., compared to 15 pounds per sq. ft. for
3-cm-thick, 150-pcf stone. First developed
in the early 1970s, these lightweight stonefaced
panels have been used on buildings
in the U.S. for over 30 years. Stone-faced
honeycomb is produced by adhering laminated
aluminum honeycomb to each side of
a single stone panel, as shown in Figure 1.
The stone is then wire-cut at midthickness
to produce two stone-faced panels. A typical
stone-faced honeycomb panel is shown
in Figure 2.
Stone-Faced Aluminum Honeycomb Composites —
Quality Control and Testing
2 9 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 rc h 2 0 – 2 5 , 2 0 1 4 K n e e z e l a n d S c h e ff l e r • 1 5 7
Figure 1 – Laminated honeycomb adhered to
both sides of a stone panel.
Figure 2 – Stone-faced aluminum
honeycomb panel. From top to bottom,
the layers are stone facing (marble),
adhesive, honeycomb facing (aluminum),
honeycomb core, and honeycomb facing.
For new construction projects, the lighter
weight of stone-faced honeycomb panels
can simplify installation and minimize cladding
dead loads. For existing projects with
deteriorated stone that would otherwise
require cladding replacement with a thicker
or denser stone, the use of stone-faced honeycomb
can eliminate the need for structural
retrofit. In addition, since composite panels
utilize only a thin-stone veneer, a stone
with limited availability can be optimized.
In some cases, reliance upon the strength
of the honeycomb may allow the use of a
stone with marginal mechanical properties
or the use of stone in applications or sizes
that would otherwise be unsuitable for traditional
stone cladding. The size of these
panels is typically limited by the size of the
stone slab.
While stone-faced honeycomb offers several
advantages over traditional stone cladding,
there are a number of material and
manufacturing variables that can greatly
influence the performance and durability
of the finished product. Understanding and
controlling these variables, in combination
with periodic quality control testing, is
important in ensuring a product with uniform
properties and expected performance.
MATERIAL AND MANU FACTURING
VARIABLES
The mechanical properties of many
stone types are inherently highly variable.
By using only a thin-stone veneer, stonefaced
honeycomb minimizes the influence of
the natural material (stone) on the variability
of the composite. The remainder of the
materials (adhesive, aluminum honeycomb,
and honeycomb facing) are synthetic, allowing
the manufacturer to control several
material and manufacturing variables to
produce a composite that is relatively uniform.
Understanding and controlling these
variables to produce a uniform composite
with the desired behavior requires significant
research and development by the
composite manufacturer during product
development and a high degree of quality
control during production.
Material variables include stone type,
adhesive type, honeycomb facing type, and
aluminum honeycomb gauge, cell size,
thickness, and orientation. Manufacturing
variables include degree of honeycomb core
expansion, surface preparation prior to
adhering stone to the honeycomb, adhesive
application, lamination procedures, and
cutting methods. The method of attachment
is also a material (type and composition
of anchor) and manufacturing (spacing
and adhesive-curing) variable. Considering
these variables, the manufacturer makes
decisions during product development to
balance cost with desired performance.
For instance, a manufacturer may determine
that it is more cost-effective to space
anchors more closely than to optimize the
strength/span of a
panel. The product
specifier should be
aware of the variables
that can affect
the mechanical
properties, durability, uniformity, and cost
of stone-faced honeycomb so that they can
anticipate an appropriate level of preconstruction
testing and monitoring during
production.
As with any product, the variables associated
with the manufacture of stone-faced
honeycomb panels should be reviewed during
preconstruction testing and periodically
throughout production to confirm that the
honeycomb product remains consistent.
This will help to ensure that tested performance
is representative of the supplied
product. Several of the material and production
variables are described below in
greater detail
MATERIAL VARIABLES
Stone Type
The durability of the stone is an important
consideration, because a nondurable
stone may crumble or delaminate within
itself, regardless of how well it is bonded to
the honeycomb. Prior to selection, it should
be confirmed that the stone facing has been
used successfully in an environment similar
to the intended environment, or physical/
mechanical testing should be performed
to review facing properties. This should
include absorption, compressive strength,
and flexural strength testing; and pass/
fail criteria should be established based on
the intended application. For material to be
used in severe weathering environments,
simulated weathering testing may be advisable,
depending on the size and significance
of the project as well as availability of information
regarding the material’s weathering
durability.
In addition, a stone with significant
1 5 8 • K n e e z e l a n d S c h e ff l e r 2 9 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 rc h 2 0 – 2 5 , 2 0 1 4
Figure 3 – Honeycomb core in strong (ribbon) orientation
relative to the length of the specimen.
Figure 4 – Honeycomb core in weak (transverse)
orientation relative to the length of the specimen.
voids, such as travertine—
or a more absorptive
stone, such as
limestone—will provide
less resistance to water
penetration than a more
impervious stone such
as polished granite. If
water is able to bypass
the stone facing, it
may degrade the bond
between the stone and
honeycomb facing and
may also affect the longterm
durability of honeycomb.
The durability
and water resistance of
the stone should be
considered when selecting
a stone facing, and
the potential effect of moisture on the adhesive
and honeycomb should be considered
when developing a testing program.
While the stone facing contributes to
the strength of the composite, it may be
possible to ignore the contribution of a lowstrength
or highly variable stone if the honeycomb
alone provides sufficient strength.
Aluminum Honeycomb Core
Variables in the aluminum honeycomb
core include cell shape and size, aluminum
alloy and gauge, adhesive used to
adhere aluminum sheets prior to expansion,
and honeycomb thickness. Changing
any of these variables can have a significant
impact on the mechanical properties
and durability of the composite. In order to
ensure uniform properties of the laminated
honeycomb product, it is desirable to obtain
periodic quality control testing of the product
without stone facing.
Honeycomb Orientation
Expanded honeycomb has a weak and
strong orientation relative to the direction
in which it is expanded. The strong (ribbon)
orientation consists of continuous strips of
metal, while the weak (transverse) orientation
is interrupted by adhesive. Figures 3
and 4 show the strong and weak orientations
of honeycomb core specimens. If the
honeycomb is distorted, as shown in Figure
5, it may not be possible to distinguish a
strong and weak orientation. The strength
of each orientation should be well understood
so that the panel span can be oriented
to optimize the strength of the panel.
Adhesive and Honeycomb Facing Type
The facings applied to the aluminum
honeycomb core behave similar to the flanges
of a structural steel beam, while the
honeycomb core behaves similar to the
web. A stronger facing will provide greater
resistance to tensile and compressive forces
imposed on the composite. Common honeycomb
facings include fiber-reinforced epoxy
(typically woven carbon or woven glass
coated in epoxy), aluminum sheet metal,
and plywood.
The strength of the laminated honeycomb
is primarily derived from the composite
behavior of the facings and core. As
such, the strength and durability of the
adhesive between the core and facing is
critical to panel performance.
MANU FACTURING VARIABLES
Degree of Honeycomb Expansion
The degree of honeycomb expansion
has a direct impact on the strength of the
honeycomb core. If the honeycomb is overexpanded,
panels that span in the strong
(ribbon) orientation will have reduced
strength, and panels that span in the weak
(transverse) orientation will have increased
strength. Conversely, if the honeycomb
is under-expanded, panels that span in
the strong orientation will have increased
strength, and panels that span in the weak
orientation will have reduced strength.
Surface Preparation
The bond between the honeycomb and
stone is reliant on proper surface cleaning.
Variables in surface cleaning include type of
cloth, type of cleaner, number
of cleaning passes, frequency
of cloth replacement,
and surface protection measures
between cleaning and
adhesive application. Quality
control during production
should include procedures
for reviewing dirt/contaminant
pickup just prior to
application of adhesive and
additional cleaning if dirt/
contaminant is observed.
Surface preparation
may also include measures
such as surface roughening.
Grinding of the stone and/
or honeycomb facing can
increase surface roughness
and enhance bond. Poor surface
cleaning or inadequate roughening
can result in significantly reduced bond
strength between the stone and honeycomb.
Adhesive Application
Another manufacturing variable is the
uniformity and coverage of adhesive on the
stone prior to application of the honeycomb.
A uniform thickness can be achieved by
applying a consistent amount of adhesive
per unit surface area and always spreading
the adhesive with the same trowel type.
Trowel types include smooth or notched and
a variety of notch depths. Quality control
during adhesive application should ensure
that 100% adhesive coverage is achieved,
that the honeycomb is applied within the
adhesive working time, and that adhesive
mix proportions and mixing procedures are
verified regularly. Special attention to adhesive
application at panel corners is important
to ensure complete coverage.
Lamination Procedures
Variables in lamination procedures
include duration and magnitude of applied
pressure, ambient conditions during adhesive
curing, and panel stacking parameters
(e.g., maximum stacking height). Quality
control should include methods to ensure
that pressure is applied equally to all panels
and that temperature and humidity are
consistent (or carefully monitored) throughout
pressing and curing of panels.
Wire-Cutting Methods
Variables in wire-cutting methods for
cutting the stone at midthickness include
2 9 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 rc h 2 0 – 2 5 , 2 0 1 4 K n e e z e l a n d S c h e ff l e r • 1 5 9
Figure 5 – Distorted honeycomb core.
the cutting wire type, cutting speed (rpm),
and panel feed rate. Modifications to cutting
methods may have an effect on production
efficiency but are not likely to have a significant
impact on composite panel strength
and durability.
Due to the many material variables
associated with the various layers of stonefaced
honeycomb panels, preconstruction
testing may be important for certain projects
to establish strength and durability
characteristics of the composite. In addition,
periodic testing of representative specimens
may be warranted during production
to confirm uniformity and to monitor the
many manufacturing variables associated
with panel production.
TESTING PROCEDURES
In some cases, such as projects with
limited use of the material, reliance upon
manufacturer-provided test results may
be sufficient. Where use of the material is
more extensive or requires a greater degree
of quality control, preconstruction and periodic
testing during panel production is
likely appropriate.
Preconstruction testing of stone-faced
aluminum honeycomb typically includes
flexural strength, tensile bond strength,
and anchor strength testing of representative
specimens cut from completed panels
in order to establish the basic mechanical
properties of the stone-faced composite.
For stone-faced honeycomb to be used in
exterior applications, consideration should
also be given to testing the composite
while exposed to anticipated temperature
extremes and after simulated weathering
exposure. Additional testing to provide
information on the performance characteristics
of the composite panels may include
full-scale flexural strength, impact load
testing, full-scale heating and cooling, and
flame spread.
Periodic testing during production
should include, at a minimum, tensile
bond strength testing to confirm full adhesive
coverage and uniform bond strength.
Periodic testing may also include flexural
strength and anchor strength testing. The
frequency of periodic testing should be
based on project surface area, and it may
be appropriate to decrease the frequency
of testing if the product consistently meets
performance requirements. For instance,
testing could be performed on ten specimens
for the first 100 sq. ft. of material
produced, then ten specimens for the next
1,000 sq. ft., then ten specimens from every
5,000 sq. ft. thereafter, assuming it continues
to meet requirements.
As with any testing, the test results are
only representative of the actual materials
and production methods used to fabricate
the composite. Additional testing should
be performed each time there is a change
in raw materials (adhesive, stone facing,
honeycomb backing) or production methods.
The basic tests for establishing and
confirming the mechanical properties of the
stone are discussed in greater detail below.
Flexural Strength
Flexural strength testing provides the
basis for establishing maximum spans and
deflection limits for the composite panels.
Testing of the honeycomb with and without
stone facing provides information on the
contribution of the stone to the strength of
the composite. For some stone-faced honeycomb
products, it may be determined that
the majority of the strength is derived from
the honeycomb core, in which case flexural
strength testing of multiple types of stone
facings may not be warranted.
Flexural strength testing of stone-faced
honeycomb panels was performed by the
authors in general accordance with ASTM
C393, Standard Test Method for Core Shear
Properties of Sandwich Constructions by
Beam Flexure. This test method is used
to determine the flexural strength of test
specimens cut from the composite panels
through quarter-span loading of simple
beam specimens. During testing, composite
specimens were tested with the stone face
in tension, where it contributes least to the
composite’s strength. Deflection was measured
during the flexural strength testing
to evaluate the stiffness and ductility of the
composite.
Offsets in the load-deflection graphs
also clearly showed the load at which the
stone facing initially cracked relative to
honeycomb failure. This can be an important
consideration to panel performance, as
cracking of the stone facing at loads that
are within the anticipated service load range
may affect panel aesthetics and weathering
durability. This was evaluated for one of the
products tested, and cracking of the stone
facing did not occur within the anticipated
service load range for this product. Flexural
strength testing of a stone-faced honeycomb
specimen is shown in Figure 6.
Flexural strength testing was performed
by the authors in both the wet (submerged
in water for 48 hours) and dry (dried in an
oven at 140ºF for 48 hours) condition and
in the strong (ribbon) and weak (transverse)
honeycomb orientation. Wet and dry test
results were compared to evaluate the effect
of moisture on the composite. Testing in
1 6 0 • K n e e z e l a n d S c h e ff l e r 2 9 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 rc h 2 0 – 2 5 , 2 0 1 4
Figure 6 – Typical view of flexural strength test setup.
both orientations is performed because
the strength of some composite systems is
predominantly derived from the orientation
of the aluminum honeycomb core material,
distinguishing it from typical dimension
stone, which is governed by its vein, bedding,
or rift orientation. Attention should
be given to the stone orientation relative to
the honeycomb orientation for fabrication of
the test specimens, particularly where the
stone’s contribution to the strength of the
composite is being considered.
Tensile Bond Strength
Tensile bond testing provides
information on the
adhesive strength between
laminations of the composite
panels. Based on past experience
of the authors, the
bond between the stone and
honeycomb core can vary
greatly between stone types,
so tensile bond testing is recommended
for each type of
stone facing to be used.
Tensile bond testing of
stone-faced honeycomb panels
was performed by the
authors in general accordance
with ASTM C297,
Standard Test Method for
Flatwise Tensile Strength of
Sandwich Constructions. The
tensile bond test is conducted
by bonding a metal plate
to the front and back face of
a specimen and applying a
tension load to a rod that is
attached to the center of each
plate until failure to determine
the ultimate tensile bond strength of
the weakest layer of the composite. This
testing should include specimens removed
from the edges of panels so that areas
that are least likely to have full adhesive
coverage are tested.
Tensile bond
strength testing
of a stone-faced
honeycomb specimen
is shown in
Figure 7.
Anchor Strength
Anchor strength testing provides information
on the methods of panel attachment.
An appropriate test approach is
to test in general accordance with ASTM
C1354, Standard Test Method for Strength
of Individual Stone Anchorages in Dimension
Stone. For this test, individual small-scale
anchor specimens are subjected to a test
load until failure to determine anchor
strength. Anchor strength testing is important
for evaluating a proposed panel attachment
system; however, it is not the subject
of this paper.
Temperature Extremes
Strength testing at hot (170ºF) and
cold (10ºF) temperatures is performed to
evaluate the strength characteristics and
adhesive bond of the panels at the high
and low ends of the typical outdoor service
temperature range. Specimens tested
at hot temperatures are preheated in an
oven for 24 hours and transferred to the
test machine in an insulated box, where
temperatures are maintained through the
use of electrical strip heaters. Specimens
tested at cold temperatures are prechilled
in a freezer for 24 hours and transferred to
the test machine in an insulated box, where
temperatures are maintained using dry ice.
Figure 8 and Figure 9 are views of the hot
and cold tests, respectively.
2 9 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 rc h 2 0 – 2 5 , 2 0 1 4 K n e e z e l r a n d S c h e ff l e r • 1 6 1
Figure 7 – Tensile bond test setup.
Figure 8 – Overall view of hot testing.
Figure 9 – Overall view of cold testing.
Simulated Weathering
Strength testing is performed after simulated
weathering exposure to evaluate the
effect of temperature cycling on material
durability. Simulated weathering is typically
performed in accordance with the “accelerated
weathering” conditioning procedures
summarized below.
Accelerated weathering conditioning has
been used for more than 25 years to evaluate
stone durability and employs procedures
similar to those outlined in ASTM
C666, Standard Test Method for Resistance
of Concrete to Rapid Freezing and Thawing.
The procedure consists of exposing the test
specimens to 100 or 300 cycles between
-10ºF and +170ºF, while the specimens are
partially submerged (1/8 to 1/4 in.) in, typically,
a 4pH sulfurous acid solution, which
simulates exposure to acid rain. For stonefaced
honeycomb, the stone facing is oriented
downward so that the bond between the
stone and honeycomb is submerged. These
accelerated weathering test conditions are
intentionally made to be harsh to demonstrate
the suitability of a stone for exposure
to weathering in a harsh environment such
as that experienced in the northern Midwest
of the United States.
In order to evaluate strength loss associated
with accelerated weathering exposure,
the flexural strength and/or tensile bond
strength of specimens tested after exposure
to accelerated weathering is compared to
the strength of specimens not exposed.
A visual examination of the specimens is
also performed before and after accelerated
weathering exposure, and deterioration and
staining are photographed and recorded. As
a way to evaluate the rate of deterioration,
testing can be performed at periodic intervals
throughout the simulated weathering
exposure.
TESTING PERFORMED
Testing was performed on three different
stone-faced aluminum honeycomb products
to evaluate mechanical properties and
durability. The following is a brief description
of each product and a summary of testing
performed on each product.
“Product A” consisted of 3/16- to ¼-in.-
thick stone adhered to ¾-in.-thick aluminum
honeycomb faced with a fiberreinforced
epoxy skin. The honeycomb cells
were approximately 0.30 in. wide (between
flat edges) by 0.35 in. long (between vertices).
Testing included flexural strength and
tensile bond strength of specimens cut from
the strong (ribbon) and weak (transverse)
orientations of full-sized panels. Specimens
were tested in the wet and dry conditions,
at hot and cold temperatures, and without
exposure and after exposure to 50, 75, and
100 cycles of accelerated weathering conditioning.
Testing was performed on specimens
removed from newly fabricated panels
faced with granite, limestone, and travertine
and on specimens removed from travertinefaced
panels that had been in service for
more than 15 years in the Chicago environment.
Flexural strength testing was also
performed on laminated core material without
a stone facing. Anchor strength testing
was performed on two anchorage types.
“Product B” consisted of 3/16-in.-thick
stone adhered to ¾-in.-thick aluminum
honeycomb with aluminum facing. The
honeycomb cells were approximately 0.30
in. wide (between flat edges) by 0.35 in. long
(between vertices).
Testing included flexural strength and
tensile bond strength of specimens cut
from the strong and weak orientations
of a newly fabricated, full-sized, marblefaced
panel. Specimens were tested in the
wet and dry conditions, without exposure,
and after exposure to 60 and 100 cycles of
accelerated weathering conditioning. Due to
concerns regarding compatibility between
the expansion of the stone facing and the
aluminum-faced honeycomb at high temperatures,
additional testing was performed
to evaluate the effect of temperature cycling
on a full-sized panel and included bow measurement
and tensile bond testing after 130
cycles from 65ºF to 140ºF. No measurable
bowing occurred during this temperature
cycling, and there was no apparent effect
on the tensile bond strength of specimens
removed from the panel after temperature
cycling. Testing of this product did not
include any unfaced specimens or specimens
at hot and cold temperatures.
“Product C” consisted of ¼-in.-thick
stone adhered to ¾-in.-thick aluminum
honeycomb faced with a glass-fiber reinforced
epoxy skin. The honeycomb cells were
approximately 0.35 in. wide (between flat
edges) by 0.50 in. long (between vertices).
Testing included flexural strength and
tensile bond strength of specimens cut from
the strong and weak orientations of newly
fabricated full-sized panels. Specimens
were tested in the wet and dry conditions,
at hot and cold temperatures, and without
exposure and after exposure to 25, 50, 90,
and 100 cycles of accelerated weathering
conditioning. Testing was performed on
specimens with marble, limestone, and
granite facings, and flexural strength testing
was also performed on laminated core
material without a stone facing.
TEST RESULTS
The following provides a summary of
flexural strength and tensile bond strength
test results for Products A, B, and C. The
effect of stone facing (unfaced vs. faced with
various stone types), honeycomb orientation
(ribbon vs. transverse), conditioning (wet
vs. dry), temperature (ambient vs. hot and
cold), and accelerated weathering (initial vs.
cycled) on the average specimen strength is
provided. Comparisons between products
are also provided where applicable.
Flex ural Stre ngth
Unfaced vs. Stone-Faced
For Product A, the average strength of
limestone-faced, new travertine-faced, old
travertine-faced, and granite-faced specimens
was approximately 5% greater, 2%
less, 10% less, and 22% greater, respectively
than the honeycomb core material
with no stone facing. With the exception of
the granite-faced panels, there was not a
significant difference between the flexural
strengths of stone-faced and core specimens
for this product, indicating that its
flexural strength is dominated by the manufactured
honeycomb material.
The average flexural strength of Product
A core specimens with no stone facing
was approximately 6 times greater than
Product C core specimens. For Product C,
the average strength of limestone-faced,
marble-faced, and granite-faced specimens
was approximately 59%, 83%, and 77%
greater respectively than the honeycomb
core material with no stone facing. Because
Product C core material has significantly
lower average tested strength than Product
A material, the relative contribution of the
stone facing to the panel strength is fairly
significant. No unfaced specimens were
tested for Product B.
Based on the testing performed, the
stone facing may or may not have a significant
effect on the flexural strength of a
stone-faced composite, depending on the
1 6 2 • K n e e z e l a n d S c h e ff l e r 2 9 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 rc h 2 0 – 2 5 , 2 0 1 4
manufacturer. The lower the strength of the
honeycomb core material, the greater the
influence of the stone facing on the composite
strength.
Variability
For Products A and B, the flexural
strength variability of new (unweathered)
test groups with no facing and faced with
limestone, travertine, and marble was less
than 10%, with most under 5%. The maximum
variability of granite-faced specimens
was approximately 12%. While the granite
appears to be contributing to higher average
strengths (as discussed above), the associated
higher degree of strength variability mitigates
the significance of this contribution.
The flexural strength variability tested
for Product A and B material is substantially
less than that commonly found in
panels made only of stone, likely because
the flexural strength of these materials is
dominated by the strength of the honeycomb.
The significance of this is that design
“safety factors” (when “allowable stress”
principles are used) or “strength reduction
factors” (when “load and resistance factor”
principles are used) are, or at least should
be, proportional to variability. In other
words, as variability decreases, safety factors
and strength reduction factors may
be decreased. Significant additional testing
beyond the limited testing performed would
be required to establish appropriate safety
factors for each product.
Product C material has much greater
variability than the Product A and B material,
and more comparable variability to
the associated stone facings. This is likely
because the stone provides a significant
contribution to the panel strength. Product
C core material also has slightly greater
variability than the Product A core material,
which also likely contributes to the greater
variability of Product C.
The variability in test results for the
stone-faced honeycomb can be very low
compared to traditional stone cladding,
depending on the contribution of the stone
facing to the strength of the composite. A
high-strength honeycomb core with low
variability will likely result in a stone-faced
composite with low variability.
Orientation
Product A core specimens tested in the
weak honeycomb orientation had approximately
7% lower average strength than
specimens in the strong orientation. The
strength of Product A stone-faced specimens
did not necessarily correlate with the
orientation of the honeycomb core. Some
stone facings resulted in a higher average
specimen strength in the strong honeycomb
orientation, and some resulted in higher
strength in the weak honeycomb orientation.
This is likely due to the orientation of
the stone relative to the orientation of the
honeycomb (e.g., strong limestone facing
orientation in combination with weak honeycomb
orientation).
Marble-faced Product B specimens tested
in the weak honeycomb orientation had
approximately 30% lower average strength
than specimens tested in the strong orientation.
Because no testing was performed on
Product B core specimens, it is unknown
whether this difference is primarily due to
the orientation of the honeycomb core, or is
also affected by the orientation of the marble
facing. Regardless, the orientation of panels
fabricated with Product B is more critical
than panels fabricated with Product A.
For Product C, core and stone-faced
specimens tested in the weak honeycomb
orientation had approximately 20% to 25%
lower average tested strength than specimens
with honeycomb in the strong orientation.
Although the stone facing provides
a significant contribution to the panel
strength of this product, the orientation of
the honeycomb had a similar effect on the
strength of core specimens and stone-faced
specimens. This indicates that the strong
and weak stone orientation likely aligns
with the orientation of the honeycomb core.
Based on the testing performed, the honeycomb
core materials tested had a definite
strong and weak orientation. Depending on
the orientation of the stone facing relative
to the orientation of the honeycomb, stonefaced
specimens may not exhibit a significant
difference in strength between strong
and weak honeycomb orientations.
Conditioning
The average flexural strength of specimens
tested in the wet condition is approximately
equivalent to specimens tested in
the dry condition for Products A and B,
indicating that moisture had little effect
on flexural strength of these products. For
Product A, this is not surprising, given that
the honeycomb dominates the strength of
the composite and is likely unaffected by
short-term moisture exposure. For Product
B, this suggests that either the stone facing
is not significantly affected by moisture, or
the honeycomb core dominates the strength
of the composite.
For Product C, the flexural strength of
specimens tested in the wet condition was
approximately 29%, 13%, and 3% lower
than the dry condition for the specimens
faced with limestone, marble, and granite,
respectively. This indicates that moisture
had a significant effect on the strength of
material faced with limestone, minor effect
on material faced with marble, and little
to no effect on material faced with granite.
Because stone provides a significant contribution
to the strength of this product, the
reduced strength likely corresponds to the
effect of moisture on the stone. Only dry
testing was performed on core material with
no stone facing, so no comparisons could be
made for the core material.
Based on the minimal difference in wet
and dry conditioned specimens for Products
A and B (where honeycomb dominated the
strength of the composite), short-term moisture
exposure did not have a significant
effect on the strength of the honeycomb
core materials. Moisture often does have an
effect on stone strength, so the strength of
a composite with significant contribution
from the stone facing can be affected by
moisture.
Temperature Effects
The flexural strength of Product A and
C specimens tested at hot temperature was
approximately 31% and 53% lower, on average,
than panels tested at ambient temperature.
Flexural strength of Product A and C
specimens increased by approximately 7%
and 14% on average at cold temperatures.
No specimens were tested in the hot and
cold conditions for Product B.
While the granite facing contributed to
greater specimen strength of Product A at
ambient temperature, and the marble and
granite facings provided a greater contribution
than the limestone facing for the
Product C material at ambient temperature,
the magnitude of flexural strength at the
higher temperature is similar, regardless of
the facing material. This suggests that the
flexural strength contributions from stone
facing are lost at the elevated temperatures.
This loss of the contribution of the stone
facing is consistent with a softening of the
adhesive between the stone and honeycomb.
2 9 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 rc h 2 0 – 2 5 , 2 0 1 4 K n e e z e l a n d S c h e ff l e r • 1 6 3
The test data indicate that raising the temperature decreases
panel strength, and that the effects are significant. Honeycomb
panel strength loss is most likely due to softening of the facing
material and/or its bond to the aluminum honeycomb core.
Accelerated Weathering
For Product A, results show that unfaced test specimens
experienced approximately 12% strength reduction from 100
cycles of accelerated weathering exposure. Limestone-, granite-,
and marble-faced specimens for Products A and B had an apparent
strength loss of 5% to 10% from 100 cycles of accelerated
weathering exposure, while travertine-faced specimens had a
greater strength reduction of approximately 13% to 18%. The
greater strength reduction in travertine-faced specimens is likely
due to direct exposure of the honeycomb core to freezing and
thawing water from voids in the travertine facing that do not
provide a barrier against water. The flexural strength reduction
of travertine specimens after each interval of accelerated weathering
exposure also varied considerably more than specimens
faced with limestone or granite. The greater variation in strength
reduction at periodic testing intervals for travertine-faced specimens
is likely due to the significant variation in the size and
amount of void area in the stone, resulting in significant variations
in bond contact area from specimen to specimen.
The fiber-reinforced epoxy facing on nearly every honeycomb
core specimen for Product C had partially or nearly fully delaminated
without load application after exposure to 100 cycles of
accelerated weathering, as shown in Figure 10. The average
core specimen strength after accelerated weathering exposure
was approximately 72% less than companion specimens not
exposed. This is significantly greater strength loss than the
honeycomb core of Product A and is
likely due to the delamination of the
epoxy facing from the core.
The stone facing on several marblefaced
Product C specimens fully delaminated
from the honeycomb core
during accelerated weathering exposure
without application of load, as
shown in Figure 11. The core for
each of these specimens was tested
to failure and included in the average
test results. While none of the limestone-
faced or granite-faced specimens
delaminated during accelerated
weathering exposure, the limestone
facing exhibited slight deterioration
after 25 cycles and more
heavy deterioration after 50 cycles of
accelerated weathering exposure, as
shown in Figure 12. Product C specimens
faced with limestone, marble,
and granite had approximately 57%,
66%, and 28% lower average tested
strength, respectively, after accelerated
weathering exposure. Each
group exhibited less strength loss
than the honeycomb core with no
1 6 4 • K n e e z e l a n d S c h e ff l e r 2 9 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 rc h 2 0 – 2 5 , 2 0 1 4
Figure 10 – Full delamination of facing on honeycomb core after
100 cycles of accelerated weathering and prior to applied load.
Figure 11 – Full delamination of marble facing from honeycomb core during accelerated
weathering exposure prior to application of load.
stone facing, likely because the facing provides
protection to the backup. Because the
stone provides a significant contribution to
the strength of Product C specimens, the
limestone-faced specimens (which exhibited
deterioration) and marble-faced specimens
(several of which delaminated from the honeycomb
core) exhibited greater strength loss
than the granite-faced specimens (which
did not exhibit deterioration or delamination).
The test data related to exposure clearly
indicate that there is an associated loss
of flexural strength over time due to exterior
exposure in a Midwestern climate for
at least Product A. The magnitude of this
effect is very difficult to estimate based on
the laboratory testing performed, primarily
because there is only one small set of data
representing actual aged panels (Product
A, travertine-faced panels). While the data
from the accelerated weathering specimens
are useful in a generally qualitative sense,
the number of cycles cannot be translated
into an equivalent number of years of realtime
exposure for the stone-faced composite,
based on the limited testing performed.
Structural design of stone-faced composite
panels to be used in exterior applications
should be based on consideration of
the durability of the facing on the honeycomb
core and its bond to the aluminum
core. For stone-faced composites that rely
to some degree on the strength of the stone,
additional loss of panel strength caused by
exposure can be due to degradation of the
stone or the bond between the stone facing
and honeycomb facing. Structural design of
panels that rely in part on stone strength
should also consider the durability of the
stone facing and its bond to the honeycomb
core.
Panel structural design should take
into consideration the reductions in panel
strength due to the exposure factors
described above, as appropriate, as well as
the variability in the strength of the panels.
Tensile Bond
Product A
Tensile bond strength testing of Product
A was performed at ambient laboratory
conditions on specimens not exposed and
after exposure to 100 cycles of accelerated
weathering. The tensile bond strength of
unweathered Product A specimens faced
with limestone, new travertine, and old
travertine were similar, while specimens
faced with granite had approximately 57%
greater average strength than the other
three facings. Failure in all of the specimens
faced with limestone and granite was
due to internal fracturing of the
stone, while half of the travertine
specimens failed within the
stone and half at its bond to
the honeycomb. Since the bond
strength of Product A often
exceeded stone strength, the
test results only provided a low
estimate of the bond strength.
The bond strength of the travertine
specimens varied more than
the other facings, likely because
there is significant variation in
the void area in the travertine,
resulting in significant variations
in bond contact area. The
greater failure loads of granitefaced
specimens only indicate
that the granite has a greater
internal fracture strength than
the limestone.
The average tensile bond
strength of the Product A limestone-
faced, travertine-faced
(new), and granite-faced specimens
after 100 cycles of accelerated
weathering exposure was
approximately 7%, 23%, and 22% lower
respectively than companion specimens
that were not exposed. The average tensile
bond strength of old travertine-faced specimens
after exposure was approximately 4%
lower than companion specimens. While
this indicates that the bond strength in new
panels is reduced by accelerated weathering
exposure, the test group with the lowest
average tensile bond strength even after
exposure had a strength of approximately
150 psi, which is considered to be more
than adequate for virtually all potential
applications.
Product B
Tensile bond strength testing of Product
B (marble-faced specimens) was performed
in the wet and dry conditions on specimens
not exposed, and in the dry condition on
specimens after exposure to 100 cycles of
accelerated weathering. All of the wet specimens
failed due to internal fracturing of
the stone, while all of the unweathered dry
specimens failed at the bond between the
marble and honeycomb. Dry specimens had
greater variability and approximately 18%
lower average strength than wet specimens
and similar average strength to Product
A specimens that were not exposed. The
variability in dry test results is likely in
2 9 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 rc h 2 0 – 2 5 , 2 0 1 4 K n e e z e l a n d S c h e ff l e r • 1 6 5
Figure 12 – Severe deterioration of limestone facing after 50 accelerated weathering cycles.
part due to variability in preparation of the
aluminum substrate prior to application of
the adhesive. Figure 13 shows the evenly
roughened aluminum substrate at the failure
plane for a dry test specimen that failed
at approximately 180 psi (approximately
average), and Figure 14 shows the unevenly
roughened aluminum substrate for a dry
specimen that had been removed from a
panel corner and failed at approximately
80 psi.
The reduction in strength of Product
B specimens after exposure to accelerated
weathering was approximately 24%. While
the reduction in strength was likely in part
due to exposure, there were also voids present
in several exposed specimens that were
not present in the initial test specimens, as
typically shown in Figure 15. The reduced
bond area due to this manufacturing variable
(uneven application
of adhesive
at areas
where certain test
specimens were
removed from the
panel) likely contributed to the lower tested
strength of exposed specimens.
Product C
Tensile bond strength testing of Product
C was performed at ambient laboratory
conditions on specimens not exposed and
after exposure to 100 cycles of accelerated
weathering. The tensile bond strengths of
Product C specimens faced with limestone,
marble, and granite were similar but had
approximately 60% lower average bond
strength than Product A and B specimens
that were not exposed. All of the limestone
and granite-faced specimens failed at the
bond between the honeycomb facing and
honeycomb core, and all but one of the
marble-faced specimens failed at the bond
between the stone and the honeycomb.
Based on the failure modes, it appears
that the adhesive was less effective in
adhering to the marble, and that additional
surface roughening of the marble would be
appropriate prior to adhesion. While the
lower tested strength and adhesive failure
between the facing and the core indicate a
honeycomb material with reduced physical
properties, the unweathered test group with
the lowest average tensile bond strength
was approximately 63 psi, which is still
likely to be more than adequate for virtually
all potential applications.
The average tensile bond strength of
the Product C marble-faced, limestonefaced,
and granite-faced specimens after
100 cycles of accelerated weathering exposure
was approximately 91%, 53%, and
36% lower, respectively, than companion
specimens that were not exposed. The
stone facing on nearly half of the marblefaced
specimens had delaminated due to
exposure, and the remainder failed at the
marble-to-honeycomb interface during testing,
further indicating that marble surface
preparation should be reconsidered.
1 6 6 • K n e e z e l a n d S c h e ff l e r 2 9 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 rc h 2 0 – 2 5 , 2 0 1 4
Figure 13 – Specimen with evenly
roughened aluminum substrate.
Figure 15 – Void within the adhesive between the marble and aluminum facing.
Figure 14 – Specimen with unevenly
roughened aluminum substrate.
The limestone facing exhibited deterioration
after exposure, and all but one test specimen
failed within the limestone. The failure
mode for granite specimens remained
the honeycomb facing-to-core bond. The
reduced strength of these specimens indicates
that the adhesive strength between
the core and facing is reduced by accelerated
weathering exposure.
In general, the tensile bond testing of
the three products showed that the bond
strength of panels from different manufacturers
varies significantly. Bond strength
can be affected by material variables such
as strength of adhesive and manufacturing
variables such as surface preparation and
adhesive coverage. Accelerated weathering
testing showed that specimens lose some
strength due to exposure, although the
degree of strength loss also varies among
manufacturers. For Products A and B, it
should be noted that in no case, even after
weathering exposure, was the tensile bond
strength significantly low that it resulted in
concern that insufficient bond was present.
SUMMARY
The test results for stone-faced honeycomb
from three different manufacturers
clearly show that physical properties and
durability can vary significantly among
manufacturers. This is due to differences
in materials and manufacturing processes.
For Product B, the flexural strength reduction
due to accelerated weathering and the
tensile bond strengths did not meet the
manufacturer’s reported values.
The test results also show that certain
manufacturers’ stone-faced honeycomb
panels have physical properties and durability
that can vary significantly for a specific
product. The specifier or purchaser of
these types of panels should be aware of
the variables that can affect the mechanical
properties, durability, uniformity, and, ultimately,
the cost of stone-faced honeycomb,
so that they can anticipate an appropriate
level of preconstruction testing and monitoring/
testing during production. Material
variables include stone type, adhesive type,
honeycomb facing type, and aluminum
honeycomb gauge, cell size, thickness,
and orientation. Manufacturing variables
include degree of honeycomb core expansion,
surface preparation prior to adhering
stone to the honeycomb, adhesive application,
lamination procedures, and cutting
methods.
Based on the results of the testing performed
on specimens selected from stonefaced
honeycomb panels manufactured by
three different companies, several recommendations
are provided for testing and
evaluating these types of panels for suitability
for exterior use. These include recommendations
for testing and evaluation
during the design process prior to construction,
as well as during construction. These
recommendations are most appropriate
for projects where significant quantities of
stone-faced aluminum honeycomb panels
are to be used and where material performance
and durability are important.
Recommended Testing and Evaluation
Prior to Construction
1) Review physical and mechanical test
data for the stone facing being considered.
If available, review projects
with the same stone type in the
same environment as proposed for
use to evaluate the stone’s durability.
2) Review available test data for the
composite product being considered.
Confirm that the adhesives and
materials used for the reviewed panels
and for manufacturer-provided
test results are the same as those to
be used for the project for which the
panel product is being considered.
3) Perform flexural strength testing:
a) Of composite specimens with
each stone facing as well as core
material to determine the contribution
of the core to the strength
of the composite.
b) In the wet condition to provide
the lower bond strength for the
composite.
c) In both honeycomb orientations
to provide information on how to
optimize panel spans.
d) At temperature extremes to evaluate
their effect on the strength
of the composite (for material to
be used in the exterior environment).
e) After accelerated weathering
exposure to evaluate its durability
(for material to be used in the
exterior environment).
4) Perform tensile bond strength testing:
a) Of composite specimens with
each stone facing.
b) After accelerated weathering
exposure to evaluate its durability
(for material to be used in the
exterior environment).
5) Perform anchor strength testing of
each anchor type.
6) Document stone-faced honeycomb
material and manufacturing variables
to confirm consistency during
production.
7) Panel structural design should take
into consideration the reductions in
panel strength due to the exposure
factors, as appropriate, as well as
the variability in the strength of the
panels.
Recommended Testing and Evaluation
During Production
1) Review and document stone-faced
honeycomb periodically to confirm
that the honeycomb material and
manufacturing procedures remain
consistent.
2) Perform periodic tensile bond testing
of representative composite specimens
to confirm uniformity during
production.
3) Consider performing periodic flexural
strength and anchor strength
testing, particularly if tensile bond
strength testing indicates a lack of
uniformity.
Some or all of the above recommended
testing and review may be performed as
part of the manufacturer’s own internal
quality control procedures, and a review
of their test reports may provide sufficient
information regarding the listed material
properties and product variability.
2 9 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 rc h 2 0 – 2 5 , 2 0 1 4 K n e e z e l a n d S c h e ff l e r • 1 6 7