par Espion, Bernard 
;Halleux, Pierre ;Provost, Michel
Référence Materials and structures, 18, 105, page (185-191)
Publication Publié, 1985-05
;Halleux, Pierre ;Provost, Michel Référence Materials and structures, 18, 105, page (185-191)
Publication Publié, 1985-05
Article révisé par les pairs
| Résumé : | Non linear methods for computing deflections of reinforced concrete linear structures in cracked stage are now spreading rapidly. In some latest reinforced concrete codes and manuals, the basic assumption regarding such calculations lies in the relationship between the mean steel (or concrete) strain and the steel strain in a fully cracked section. The authors have established a state-of-the-art for such relationships. All proposed formulas can practically be written in the following form: {Mathematical expression} where ε m stands for the mean steel and concrete strain, ε s the steel strain in a cracked section and where Δθ represents the tension stiffening effect of concrete lying between the cracks. Various propositions have been made for the relationship between Δθ and the steel stress (e.g. constant, linear, hyperbolic,...); other important factors too are the strength of concrete in tension and the reinforcement ratio. Considering that little experimental evidence supports the various propositions, an experimental program on thirteen reinforced concrete prisms (ties) subjected to uniaxial tension was carried out. The influence of the reinforcement ratio is especially studied. Careful assessment of the validity of previous formulas is made. A new bilinear relationship between the tension stiffening effect and the steel stress is proposed for analytical calculations. © 1985 Bordas-Gauthier-Villars. |
