ETL 1110-2-365
31 Aug 94
construction joints. Construction lift joints must be
c
specific heat (Btu/lb F)
treated the same way as any other face of a structure.
∆L
A construction lift joint or vertical construction joint
length between adjacent nodes
cannot be contained within an element; it must be at
of an element (in.)
the interface of a row of elements. Therefore, if
analyses with different lift joint locations are antici-
While every effort should be made to adhere to the
pated, the FE mesh should be developed so that the
equation given above, it may become necessary,
revised locations can be accommodated. Accommo-
particularly in 3-D analyses, to exceed this criteria to
dating such changes should also account for inclusion
enable a problem to be solved. Truman, Petruska,
of at least two elements in each lift. If the FE mesh
and Ferhi (1992) have shown that exceeding the
can accommodate the new locations, then the only
above criteria in the direction perpendicular to heat
change needed for the ABAQUS input is to redefine
flow is acceptable. This information can be utilized
the sets of elements associated with each lift.
in long slabs or thin walls where the direction of heat
flow is generally in one direction.
(4) Preprocessing. There are various programs
available that may be used to provide preprocessing
d.
Heat transfer analysis.
capabilities in developing a mesh. If a decision is
made to use a preprocessor, users should select a
(1) Ambient temperature. Once the data for the
preprocessor with which they are familiar or feel they
ambient temperature have been developed as
can learn easily. The user must determine the prepro-
described in paragraph A-2c, this appendix, it can be
cessor's compatibility with ABAQUS. One possible
put into the ABAQUS input file using the
choice for a preprocessor would be the preprocessing
*AMPLITUDE card. This will provide an average
portion of the ANACAP-U software (ANAGEN,
daily temperature curve. The data can be input in
described in the ANACAP-U User's Manual
5-to 10-day increments, and the analysis will linearly
(ANATECH Research Corp 1992)) which allows the
interpolate between the values provided for times in
user to define various key nodes in the geometry of
the analysis when a value is not specifically given.
the structure and then requires specification of the
mesh density, thereby generating the mesh. In a
(2) Cold fronts. The passage of cold fronts
similar manner, node sets and element sets may be
where the temperature drops significantly over a short
defined.
period of time can be a critical factor when evaluat-
ing the cracking taking place during mass concrete
(5) Element size limitation. Due to the formula-
construction. Explicit modeling of cold fronts is not
tion of the heat transfer algorithm used in ABAQUS,
required in a NISA, but the effects cold fronts can
there is a criterion which relates the length of the
have on the behavior of the structure should be con-
timestep to the size of the elements being used.
sidered as described in paragraph A-6d(3), this
Violation of the criterion could lead to numerical
appendix. The fact that initial conditions are conser-
inaccuracies in the solution which could introduce
vatively selected by the design team and that extreme
errors into the temperature values computed. Due to
ambient conditions are used as well as the reduced
concerns about capturing the heat generated at early
evaluation criteria discussed in paragraph A-6d(3)
times and the length of the timesteps needed to cap-
provide the justification for not requiring a specific
ture this behavior, the length of the timestep is the
cold-front analysis. If the criteria in para-
controlling factor and must be used to compute the
graph A-6d(3) is exceeded, then changes to the insu-
element size. The size of the element is therefore
lation may be required. The film coefficient value
limited by the following equation:
for insulation can be used for a longer period of time
in the analysis, and an increased amount of insulation
6k∆t
∆L 2 <
can be modeled by decreasing the value of the film
(A-1)
ρc
coefficient as described in the following
paragraph A-5d(3).
∆t
time increment (days)
(3) Film coefficients. An essential part of the
ρ
density (lb/in.3)
heat transfer analysis is to model convection which is
A-18
Previous Page
