Abstract Normal gravity model tests were carried out in a trench to study the influence of several design parameters on the behavior of nailed soil-retaining walls. Six nailed walls were built by varying the nail length, nail inclination, and the method of nail installation. The construction of each wall involved gradual placement of 36 m3 of sand using a clamshell attached to an overhead traveling crane, followed by excavation (in front of the facing) and installation of model soil nails. Each completed nailed soil wall and its retained sand measured 3.0 by 3.0 m in plan and was 2.4 m high before placement of surcharge fill. This paper describes the method adopted for reducing side wall friction, the sand placement procedure, the checks on uniformity and reproducibility of the relative density of the sand, the instrumentation, and the stage-by-stage excavation and nailing process. A summary of monitored results of all the tests is presented, and the results of Test 1 are discussed in detail.