During this project, the 3D-printing of a drone, operating à 120°C, with Fused Deposition Modelling was studied. The used materials were chosen among a given list, with the help of Differential Scanning Calorimetry and Rheometry. Then, to study their composition and behaviour, they were characterised using the available experimental techniques and a limited budget. Finally, the drone was designed and printed. With more time and resources, the drone could have been improved in the following ways.

 

First of all, the choice of polymer was restricted. The nGen Colorfab would have been interesting to use if the available quantity was sufficient. Indeed, it is easy to print and has good mechanical and thermal properties [1]. However, as its glass temperature (85 °C) is below the operating temperature of the drone, it would be necessary to proceed with mechanical tests in order to verify its mechanical properties at high temperature. Another appropriate choice would have been the Polycarbonate. This material offers good thermal properties (its glass temperature is 145°C [2]) and  has a high toughness [3]. Nevertheless, it is difficult to print due to the high extrusion temperature required (260-290°C), which is too high for the Makerbot Replicator 2X. Moreover, if the need of high impact strength would have been emphasized, the PMMA would have been chosen with its value of 0.16-0.27 J/cm [4].

 

Regarding the 3D printer, the Fused Deposition Modeling was a relevant tool in our case : ease of use, low cost, wide range of possible materials. However, it has its limitations such as finishing, resolution and speed which could have been an issue if the requirements were higher. The Continuous Liquid Interface Production would bring a solution to the limited speed and finishing issues, but it is not yet available. The technique which seems the most interesting for improvement is the stereolithography (SLA). Indeed, it can be used to get layers with a height as small as five microns, which is way better than the 100 microns obtained with FDM and means a higher resolution. The printing is also quicker and the mechanical properties are better due to the chemical interactions between the layers. Moreover, the use of resins means that the printing is not restricted to thermoplastics.   

 

Concerning the characterisation, the budget was the limitation, to raise awareness of the their cost and the importance to make relevant choices. In addition to that, the lack of experience meant that we did not always know what was possible to do and what was the most appropriate option. The analysis could have been preceded by thermic cycles in order to determine if the properties of the materials change with repeated use of the drone at high temperature and a storage at room temperature. Besides, the TGA or mechanical test at different temperatures could have been used to find the highest temperature that the drone can withstand with its properties remaining unchanged. As regard to the mechanical properties, impact tests would have been necessary to determine the appropriate material and design of the bumper. The polymer chosen, the Flexifil, is a new polymer for 3D-printing and its exact composition was not known. A more advanced characterization of its composition would have been useful to understand better its properties. Another important aspect to investigate for both polymer is their chemical resistance. In fact, in the case of an industrial chimney, they would be exposed to many types of solvents which could deteriorate their properties.

 

Finally, some improvements could have been made to the design of the drone, with a finite element analysis to determine the best design parameters, including the shape, the dimension, the thickness, the density of infill or the weight. Also, It would be interesting and rewarding to finish this project with a fully-functional drone. Flying test can be realized in order to observe directly possible further improvements. Besides, sensor (chemical, thermal,…) or cameras can be added to the drone to respond to the need of real life industrial applications. Moreover, the aspect of the price during the material selection and the design could have been also considered.

​