Aim of the study was to create an animal model for the investigation of the role of subchondral bone damage without initial cartilage lesion in the pathogenesis of osteoarthritis, the mechanical properties of the joints as well as its role in cartilage metabolism. Therefore, after cadaver studies an animal model was created to apply a transarticular load to the femoro-patellar joint under reproducible conditions and produce a pure subchondral damage without affecting the articular cartilage. Following the cadaver studies a first group of four dogs was impacted to identify forces to produce isolated subchondral fractures in the femoral condyle. Then a second group of 12 dogs knee joints was impacted under identical conditions with forces of approximately 2100 N to produce similar subchondral fractures without cartilage damage in one joint under MRI control: T1-weighted SE-sequences. T2-weighted TSE, fat suppressed TIRM-sequences and 3D-FLASH fat saturated sequences. FLASH 3D-sequences revealed intact cartilage after impact in all cases and TIRM-sequences showed subchondral fractures representing bleeding, microfractures and fragmented bone trabecules. Turbo spin echo sequences and T1-weighted images revealed other intact intraarticular structures such as ligaments and menisci. The proposed experimental animal model is suitable to investigate the effect of pure subchondral damage on the articular cartilage and on means of treatment of cartilage defects without surgical intervention and without initial cartilage damage.