Background Anterior cruciate ligament reconstruction (ACLR) has developed dramatically in the last century. Now, ACLR has become a reliable and productive procedure. Patients feel satisfied in >90% cases. The aim of this study was to evaluate the feasibility of allogenetic cortical bone cross-pin (ACBCP) used as a clinical fixation method in anterior cruciate ligament reconstruction on the femoral side based on biomechanical tests in vitro. Methods The specimens were provided by the bone banks of the First Affiliated Hospital of People's Liberation Army of General Hospital from September 2011 to June 2012. Fresh deep frozen human allogenetic cortical bone was machined into cross-pins which is 4.0 mm in diameter and 75.0 mm in length. Biomechanical parameters compared with Rigidfix were collected while cross-pins were tested in double-shear test. The load-to-failure test and cycling test were carried out in a goat model to reconstruct anterior cruciate ligament with Achilles tendon autograft on the femoral side fixed by human 4.0 mm ACBCP and 3.3 mm Rigidfix served as control. Maximum failure load, yield load, and stiffness of fixation in single load-to-failure test were compared between the two groups. Cycle-specific stiffness and displacement at cycles 1, 30, 200, 400, and 1 000 were also compared in between. Results In double-shear test both maximum failed load and yield load of 4.0 mm human ACBCP were (1 236.998 +/- 201.940) N. Maximum failed load and yield load of Rigidfix were (807.929 +/- 110.511) N and (592.483 +/- 58.821) N. The differences of maximum failed load and yield load were significant between ACBCP and Rigidfix, P<0.05. The shear strength of ACBCP and Rigidfix were (49.243 +/- 8.039) MPa and (34.637 +/- 3.439) MPa, respectively, P<0.05. In the load-to-failure test ex vivo, yield load and maximum failed load of ACBCP fixation complexity ((867.104 +/- 132.856)N, (1 032.243 +/- 196.281) N) were higher than those of Rigidfix ((640.935 +/- 42.836) N, (800.568 +/- 64.890) N, P<0.05). However, stiffness did not differ significantly between ACBCP group ((247.116 +/- 31.897)N/mm) and Rigidfix group ((220.413 +/- 51.332) N/mm, P>0.05). In the cycling test, the cycle-specific stiffness and displacement at cycles 1, 30, 200, 400, and 1 000 did not differ significantly between the ACBCP group and Rigidfix group, P>0.05. Conclusions Allogenetic cortical bone cross-pin possesses satisfactory biomechanical profile which is safe for ACLR and suitable for an aggressive rehabilitation program. Animal and clinical tests should be recommended before clinical use to secure the ACBCP could successfully substituted by host new bone in vivo.