the butt joint configuration, there can be prema-
characterizing a seal's response to load sometimes
ture failure of a seal because the load distribution
adopt the movement capability concept without
results in increased stresses in the seal (e.g., ACI
concern of the load. These techniques are intended
1993, ASTM 1991f).
to provide performance data so that a seal designer
A pavement seal can be subjected to many types
can ensure that the working movement capability
of loadings. These include loadings caused by sea-
of a seal is greater than the movement that the joint
sonal, daily and traffic-induced movements of the
will experience.
joint or crack opening; the direct loading of the
Tests of model seals are generally conducted
seal itself by tires or hard objects such as small
using block-shaped seal specimens constrained be-
rocks under the weight of tires; the forces of con-
tween parallel surfaces of a given substrate. Per-
straint caused by the volume change response of
formance-based standard tests of these model seals
the seal material that results from the curing pro-
(Beech 1985), for both the pavement and building
cess, temperature variations or the interaction of
seal industries, have been developed primarily to
the seal material with a foreign substance such as
provide a basis for quantifying and indexing move-
jet fuel; and the forces of constraint caused by
ment capability in the context of the movements
changes in the stressstrain behavior of the seal
associated with seasonal joint opening and clos-
material induced by temperature variations or the
ing. As suggested by the standards, these tests are
aging of the material. The distribution of loads on
typically conducted at a specific low temperature
pavement joint seals as a result of long- and short-
in recognition of the severity of the low tempera-
duration movements of the joints is not a directly
ture extreme. None of the standard methods in-
measurable quantity, whereas joint movements can
corporate test series for measuring the possible stiff-
be measured directly. As a consequence the "move-
ening of the response as the temperature is lowered,
ment capability" (see e.g., Panek and Cook 1984,
however. Many other laboratory and field tests that
ASTM 1991f) of a model seal in terms of either a
are not standard have been performed to measure
safe-working elongation or an ultimate, failure
the movement capability of model seals. These in-
elongation, rather than a load capability in terms
clude laboratory tests of a model seal's response
of the sealant's strength, is commonly used to
to transverse movements (e.g., Shisler and Klosow-
comparatively describe sealants. Examples of hori-
ski 1990), and long-duration field tests using ther-
zontal joint movement measurements are the data
mally designed loading fixtures that allow a model
of Minkarah et al. (1982), which are from portland
seal to be subjected to environmental weathering
cement concrete highway test sections in Ohio with
and temperature variation-induced joint move-
regular joint spacings. Minkarah et al. mentioned
ments (e.g., Karpati et al. 1977). Regarding the lat-
that short-term measurements of 25% were typi-
ter, although an extensive amount of research on
cal for joints formed by 6.4- and 12.2-m highway
sealant movement capability has been performed
sections, and were greater than longer cyclic move-
using outdoor strain cycling exposure racks (e.g.,
ments. Such movements can be roughly estimated
Karpati 1980, 1989), continuous load and defor-
with manageable structural analysis calculations.
mation measurements with strain cycling have not
For example, the American Concrete Institute (ACI
been made, and thus quantitative structural re-
1993) has suggested an analysis technique for pre-
sponse information has not been obtained during
dicting horizontal movements of joints in portland
the exposure. Such measurements could provide
cement concrete pavements due to thermal con-
significant quantitative data regarding tempera-
traction and expansion of the concrete material.
ture-induced hardening and its effect upon the load
and deformation response of seal specimens. How-
ever, typically only the joint movement and tem-
Conventional performance testing for
perature have been measured while the seals are
studying the load and deformation
on the rack, and changes of the model seals have
response of joint and crack seals
As mentioned above, it is typically a "move-
been noted visually (Karpati 1980).
ment capability" of a model seal, in terms of ei-
Beech (1985) has suggested that tests of model
ther a safe or ultimate extension, compression or
seals yield superior results for measuring move-
other deformation, that is used to summarize the
ment capability, compared to tests of actual seals
load and deformation performance of the corre-
in field service, because the results from in-ser-
sponding sealant product. In general, although an
vice tests reflect a specific set of circumstances and
apparent modulus is often reported for a model
do not lend themselves to statistical analysis. In-
seal, conventional experimental techniques for
deed, when the goal of the test is to determine a
10
Previous Page
