ETL 1110-2-536
31 Dec 94
(4) Calibrating the longitudinal model may be
would be more appropriate in this case. If the foun-
done by adjusting the thermal conductivity of differ-
dation conditions are modeled by the use of springs,
ent element sets, allowing for higher conductivity in
the degree of restraint may be adjusted by softening
the earlier time steps and reducing the conductivity in
or stiffening the spring constants. In the case of
the later time steps where the convective surface
Zintel Canyon Dam, the foundation rock was highly
plays a more significant role in cooling.
fractured. Therefore, the degree of restraint provided
by the concrete to rock interface may be less than
(5) Examination of the stress history plots indi-
fully restrained. However, unless sufficient data
cates stresses are still increasing as a result of
support softening the spring constants, the results of
decreasing ambient air temperatures. For both mod-
the stress analysis would be unconservative in this
els the cooling period should, at a minimum, be
case. Secondly, the model can be modified by insert-
applied until thermal stresses begin to decrease and
ing a transverse joint at the mid-point of the spillway
attain a steady state. For this analysis the cooling
or at the corners of the spillway/non-overflow inter-
period was 9 months for the transverse model and
section. Additional measures could include reducing
3 months for the longitudinal model. Admittedly, this
the RCC placement temperature and mandating place-
analysis fell short of predicting maximum stresses.
ment schedules to avoid hot seasons.
However, review of the principle stress contours from
the longitudinal model reveals high stress areas where
(2) For Zintel Canyon Dam, considering the
cracking is likely to occur. One area is at the center
frequency of reservoir impoundment and the function
of the spillway, the second, at the intersection of the
of the structure, the observed cracking is well within
spillway and non-overflow section, and the third, in
acceptable levels for the project. There would be no
the upper reaches of the abutments. With the version
benefit for implementing these measures.
of UMAT used for this study, it is difficult to predict
where cracking will initiate first. However, where
(3) The results of this NISA analysis were, in
locations of high stresses occur, the model may be
general, consistent with the results of the approximate
modified to depict the location of transverse joints if
thermal analysis performed during design of the pro-
cracking is undesirable in those regions. Other condi-
ject. Both indicate that cracking may occur in three
tions that affect the stress in the dam are the assumed
areas. Based on that comparison and the observed
foundation restraints. Both models include fully
performance of Zintel Canyon Dam and other RCC
restrained boundary conditions at the RCC/foundation
dams, the NISA for Zintel Canyon Dam provided no
interface. By visual observations of the rock and
additional information to attain the desired objectives
postconstruction coring of the foundation, the
stated in the ETL. This statement is based on the
assumed restraint conditions used in the model could
fact that:
be modified to provide a more flexible restraint con-
dition or an adjusted foundation modulus. Before
Some cracking would be acceptable as long
proceeding with any further analysis, the most recent
as structural stability is not compromised.
version of UMAT should be used to include the
redistribution of stress that occurs after cracking. For
Joints for control of cracking is not necessary
this study, no further calibration of the longitudinal
for serviceability conditions since there is no
thermal model was completed. This is mainly due to
permanent reservoir.
limited scope of the study and the fact that cracking
is not of great concern for Zintel Canyon Dam.
Because of the combinations of the above,
real cost savings have been achieved by
maximizing production rates for placement of
e. Conclusions.
RCC.
(1) Performance of the analysis leads to several
Unusual loadings, extreme loadings, or
conclusions and recommendations for subsequent
severe operational conditions do not exist.
steps to proceed with further evaluation of cracking.
The first step would be to make the required adjust-
(4) This is not to say that a NISA should not be
ments in the foundation modulus and/or the restraint
performed for RCC structures. Each structure is
conditions. The model for Zintel Canyon Dam
subject to unique conditions and loadings resulting in
included the foundation in the stress analysis.
unique structural features. These factors are
Adjusting the foundation modulus to tested values
A-6
Previous Page
