Introduction Surgical training and biomechanical testing require models that realistically represent the in vivo injury condition. The aim of this work was to develop and test a method for the generation of distal humerus fractures and olecranon fractures in human specimens, while preserving the soft tissue envelope. Methods Twenty-one cadaveric upper extremity specimens (7 female, 14 male) were used. Two different experimental setups were developed, one to generate distal humerus fractures and one to generate olecranon fractures. Specimens were placed in a material testing machine and fractured with a predefined displacement. The force required for fracturing and the corresponding displacement were recorded and the induced energy was derived of the force–displacement graphs. After fracturing, CT imaging was performed and fractures were classified according to the AO classification. Results Eleven distal humerus fractures and 10 olecranon fractures with intact soft tissue envelope could be created. Distal humerus fractures were classified as AO type C (n = 9) and as type B (n = 2), all olecranon fractures were classified as AO type B (n = 10). Distal humerus fractures required significantly more load than olecranon fractures (6077 N ± 1583 vs 4136 N ± 2368, p = 0.038) and absorbed more energy until fracture than olecranon fractures (17.8 J ± 9.1 vs 11.7 J ± 7.6, p = 0.11), while the displacement at fracture was similar (5.8 mm ± 1.6 vs 5.9 mm ± 3.1, p = 0.89). Conclusion The experimental setups are suitable for generating olecranon fractures and distal humerus fractures with intact soft tissue mantle for surgical training and biomechanical testing.