To investigate the influences of parameters such as the position of post-installed rebar on the mechanical behaviors of a Type Ⅱ grouted sleeve lapping connector named APC connector (all vertical members precasted in concrete structures), a unidirectional tensile test was conducted on 36 specimens. The failure mode, ultimate bearing capacity, ductility and sleeve strain of the specimens were studied, and ABAQUS was used for numerical simulation. Test results showed that the use of anti-deflection measures effectively improved the ultimate bearing capacity and initial stiffness of the specimens. The bearing capacity and ductility of the connector were mainly influenced by three factors, namely the spacing of the steel bars, the eccentricity of the post-installed rebar, and whether the post-installed rebars were arranged tightly against the sleeve. When the post-installed rebars were arranged in the middle of the sleeve with a larger spacing between the rebars, the restraint provided by the longer side of the sleeve was weaker, resulting in a decrease in the bearing capacity and ductility of the splice. When the post-installed rebars was eccentrically positioned, it was prone to bending deformation, resulting in reduced connector bearing capacity and ductility. When the post-installed rebars was placed in close proximity to the sleeve, it was susceptible to grouting defects, resulting in reduced connector bearing capacity and ductility. When the rebars were symmetrically and tightly arranged on the long axis, the specimens exhibit the maximum bearing capacity and ductility. At the ultimate load, the longitudinal strain on the short side of the sleeve was tensile strain, but there was a compression trend, while the longitudinal strain on the long side of the sleeve was compressive strain. The absolute value of the circumferential strain on the long side of the sleeve was basically greater than that on the short side but smaller than the yield strain of the sleeve. The failure modes and ultimate bearing capacity of the simulated specimens were in good agreement with the experimental results, validating the reliability of the model. With different positions of post-installed rebars, the minimum value of the ultimate load was about 80% of the maximum value.