Computational Fluid Dynamics (CFD) is widely used to model fluid flow and optimize industrial processes. The main obstacle to implement these methods to small businesses might be the price of commercial software. However, many cases do not necessarily require state-of-the-art algorithms and use of open-source software might be the best solution. The aim of the article was to study and present the capabilities of a solver Elmer FEM. Meshes were generated with Gmsh software and data visualization was done with the use of application ParaView, all of which are open-source software. The chosen software satisfied criteria of being simple to learn, being compatible with widely used file formats and having an active community. What is more, it allows the user to solve multiphysics problems. The object of simulations was a prototype of a graphene flow meter, developed by Industrial Research Institute for Automation and Measurements. The article focused on a key issue in most of the CFD simulations, which is turbulence modelling. Both RANS-based and Large Eddy Simulation models were tested. The process of setting up simulations and its results are presented.