In this paper we present the comparison of error estimators using in 3-D magnetoharmonic FE model dedicated to eddy-current testing. The first error estimator is based on the nonverification of the behaviour law. This estimator is very accurate but requires the solution of both complementary formulations. The second estimator is based on an energetic approach using the Poynting vector in conductive regions and using the variation of the magnetic energy in nonconductive regions. Both estimators are tested within an adaptive meshing procedure and the results are compared with experimental results.