The objective of this study was to determine how effectively the Weaver-Dunn repair (both unaugmented and augmented with a suture and suture anchor) restores joint translation in response to applied loads to normal. Translation of a reference point on the clavicle relative to a reference point on the acromion was assessed in five cadaver shoulders by applying anterior, posterior and superior loads of 50 N to the clavicle using a specially designed test rig while measuring movement of the acromion and clavicle with an optical measurement system. Translation was determined for the intact joint, after simulated injury and Weaver-Dunn repair, and after augmentation of the Weaver-Dunn repair with a suture fixed to a suture anchor in the coracoid process. Joints were significantly more mobile after Weaver-Dunn repair (16.1 mm anterior, 15.7 mm posterior, 11.1 mm superior) than when intact (4.1 mm anterior, 3.2 mm posterior, 4.0 mm superior) (p < 0.005). Augmentation with a suture and suture anchor reduced separation of the Weaver-Dunn reconstruction significantly (to 5.3 mm anterior, 4.1 mm posterior, 2.0 mm superior) (p < 0.005). Joints reconstructed using an augmented Weaver-Dunn repair were not significantly more (or less) mobile than normal joints (p > 0.005), although the power of the test to detect this difference was low (power = 0.107). We anticipate that, when surgery is indicated for treating acromioclavicular joint injury, an augmented Weaver-Dunn reconstruction will yield a joint that is less painful, more functional and less likely to require revision.