ETL 1110-2-365
31 Aug 94
incremental construction analysis procedure. Stresses
benefits of the reinforcing prior to evaluating any
across a vertical construction joint should be
other measures to eliminate the cracking. Modeling
examined closely for determination of any special
of reinforcement should be done using the ABAQUS
measures that should be taken during the design and
*REBAR option and not with discrete truss elements.
construction of the joint (e.g., placement of reinforce-
For information on the modeling of reinforcement,
ment bars across the joint face). In addition, perfor-
refer to the WES report, "Use of Reinforcement in a
mance of a 3-D or a 2-D analysis in the longitudinal
Nonlinear, Incremental Structural Analysis" (Fehl and
plane should be considered for monoliths with verti-
Merrill in preparation).
cal construction joints to confirm results obtained in
the 2-D analysis of the transverse plane. The reason
A-5. Structural Modeling and Analysis
for this is that the joint itself may be located in the
out-of-plane direction. Changes in vertical construc-
tion joints may require changes to an existing mesh
a.
Finite element code.
but, as a minimum, will require changes in the input
with respect to the element set definitions.
(1) Software requirements. The FE code
ABAQUS (Hibbitt, Karlsson, and Sorensen 1989) in
conjunction with the concrete constitutive model
g. Embedded cooling coils. Cooling coils to
contained in ANACAP-U (ANATECH Research
reduce heat within an MCS have been used in some
Corp. 1992) should be used for performing a NISA as
large gravity and arch dam projects but have typically
described within this document. The modeling tech-
not been needed on navigation type structures. If
niques for performing a NISA have all been estab-
placing temperatures have been reduced to their lower
lished using ABAQUS and the ANACAP-U software.
limit, lift heights have been reduced to a practical
Should extenuating circumstances arise which pre-
minimum, and temperatures within the structure
clude the use of ABAQUS or ANACAP-U, a full
remain excessive, then the addition of cooling coils
evaluation of the code selected must be made includ-
should be considered. This can be accomplished in
ing comparisons to results obtained using ABAQUS
ABAQUS by using the CFLUX command and speci-
and ANACAP-U on full-scale problems. In addition,
fying temperatures at nodes at or near locations of
use of an FE code other than ABAQUS and a consti-
cooling coils.
tutive model other than ANACAP-U must have the
approval of CECW-ED.
h. Geometry. The geometry of the structure is
of course a major contributing factor to the behavior
(2) Heat generation subroutines. The subroutines
of the structure. Therefore, a NISA should not be
DFLUX and HETVAL are used by ABAQUS to
performed until the structural geometry is at a stage
define the heat generation for input into the heat
where only minor changes to the geometry are
transfer analysis and either of these subroutines may
expected. While this parameter may be more difficult
be used. The adiabatic temperature rise curve
than other parameters to alter, there may be instances
obtained from testing is used in the subroutine, and
where it will be necessary to make this type of
the volumetric heat generation rate is calculated from
change. If a change is made to the geometry of the
this curve to provide time-dependent heat generation
structure, then coordination between all disciplines is
to ABAQUS in performing the heat transfer analysis.
a necessity to ensure the change does not have an
Annotated examples of the DFLUX and HETVAL
adverse effect on some other function of the structure.
subroutines are given in Annex 3, Appendix A. Both
A change in the geometry will generally require some
will accomplish the task of providing the internal heat
type of revision to the mesh of the model.
generation due to the heat of hydration. The differ-
ence in the two subroutines is that DFLUX is acti-
i. Reinforcing. Reinforcing is an integral part
vated through control of the element numbers while
of many of the MCS's used within the Corps of
HETVAL is activated through the use of an assigned
Engineers, but to date there has been limited use of
material name for each concrete lift. Both DFLUX
reinforcing in NISA's. This was because many of the
and HETVAL are user supplied subroutines which
structures analyzed to date had no cracking problems
must be developed for each project.
and adding reinforcing in a model when cracking is
not occurring has little effect on results. If an analy-
(3) Material model subroutine. UMAT is a user
sis predicts cracking in a structure, then reinforcement
subroutine in the stress analysis portion of NISA that
should be included in the model to determine the
A-16
Previous Page
