Assessment of Deflection-Based Acceptance Criteria for Load Testing

The Canadian Standards Association’s CSA A23.3:19 Design of Concrete Structures allows for evaluating the safety of existing structures based on analytical assessment or load testing. The latter is often favored by practitioners because the underlying concept is intuitive, so results are conclusive. The load testing procedures have the following shortcomings: (1) They do not provide insight on the actual load-carrying capacity, or how close the proof load was to the ultimate capacity of a structural component; (2) the proof load may cause irreversible structural damage, or even structural failure; and (3) many practitioners believe that a structure loaded to its nominal ultimate capacity typically engages alternate load paths to avoid failure, and that a different test result may occur under different environmental conditions. The goal of the present research is to model analytically the standard monotonic load test to assess the effectiveness of the acceptance criteria based on deflection recovery. Deficient members are represented using an effective mechanical reinforcement ratio, ω, less than the “expected” ω. Then, the total deflection of load-tested deficient members is evaluated based on a modified moment–curvature relationship that accounts for the effect of strain hardening of the reinforcing steel, and secondary and tertiary creep of concrete under high sustained loads. Residual deflections are obtained assuming the unloading is elastic. The results indicate that a limited range of deficient members can sustain the test load for 24 h without collapse, achieve a residual/total deflection ratio less than 40%, and so pass the load test. Thus, the link between structural capacity and deflection is tenuous at best.