ABSTRACT
In the last decades, several structures affected by Delayed Ettringite Formation (DEF) have been discovered and studied, impulsing development of numerical models to evaluate its structural consequences. Based on our experience with real structures, some reflexions on the use of numerical models for assessing DEF-affected structures are presented.
First, the real cases are briefly described. Then we consider three issues:
The role of numerical models for managing affected structures.
The assumptions chosen for the model: which phenomena shall be reproduced by the numerical model, which can be neglected?
Correctly fitting the model: which informations can be used, and how? For real cases, crucial parameters such as temperature at early-age are difficult to obtain, especially with old structures, and might be evaluated by indirect methods.
Finally, this communication proposes some suggestions to improve the usefulness of numerical tools when applied to reassessment of real structures.
