UAV photogrammetry is one of the most widely used techniques for the acquisition and generation of 3D models for small areas. There are many studies focused on the analysis of the influence of several factors on the model accuracy, most of which conclude that the number and distribution of the Ground Control Points (GCPs) is one of the most important parameters. Traditionally, the GCPs are uniform distributed in two dimensions along the study area. Considering the representation of elements with a noticeable altimetric component, this research proposes a uniform three-dimensional distribution, placing the GCPs in different heights and planes along the modeled object. This alternative has been designated as Multi-control. This research proposes the generation of four 3D models of a mining derrick with different GCPs distributions and their comparison by the RMSE in Check Points and distances, as well as a cloud-to-cloud comparison. The results obtained show that a Multi-control distribution, allows reducing the RMSE and obtaining fewer variable results by presenting a lower standard deviation, providing a better result in the model with the GCPs distributed at different heights along the structure.