Wettability, typically estimated through the contact angle, is a fundamental property of surfaces with wide-ranging implications in both daily life and industrial processes. Recent scientific interest has been paid to the surfaces exhibiting extreme wettability: superhydrophobic and superhydrophilic surfaces, characterized by high water repellency and exceptional water wetting, respectively. Both chemical composition and morphology play a role in the determination of the wettability “performance” of a surface. To tune surface-wetting properties, we considered coatings of carbon nanoparticles (CNPs) in this study. They are a new class of nanomaterials synthesized in flames whose chemistry, dimension, and shape depend on combustion conditions. For the first time, we systematically studied the wettability of CNP coatings produced in a controlled rich ethylene/air flame stabilized over a McKenna burner. A selected substrate was intermittently inserted in the flame at 15 mm above the burner to form a thin coating thanks to a thermophoretic-driven deposition mechanism. The chemical-physical quality and the deposed quantity of the CNPs were varied by opportunely combing the substrate flame insertion number (from 1 to 256) and the carbon-to-oxygen ratio, C/O (from 0.67 to 0.87). The wettability of the coatings was evaluated by measuring the contact angle, CA, with the sessile drop method. When the C/O = 0.67, the CNPs were nearly spherical, smaller than 8 nm, and always generated hydrophilic coatings (CA < 35°). At higher C/O ratios, the CNPs reached dimensions of 100 nm, and fractal shape aggregates were formed. In this case, either hydrophilic (CA < 76°) or superhydrophobic (CA ~166°) behavior was observed, depending on the number of carbon nanoparticles deposed, i.e., film thickness. It is known that wettability is susceptible to liquid surface tension, and therefore, tests were conducted with different fluids to establish a correlation between the flame conditions and the nanostructure of the film. This method offers a fast and simple approach to determining mesoscale information for coating roughness and topographical homogeneity/inhomogeneity of their surfaces.