Unmanned Systems Technology Dec/Jan 2020 | Phoenix UAS | Sonar focus | Construction insight | InterGeo 2019 | Supacat ATMP | Adelan fuel cell | Oregon tour | DSEI 2019 | Copperstone Helix | Power management focus

25 The baseline configuration is a sphere, with sufficient volume to hold enough lift gas to carry the aircraft and its payload aloft. However, we know that this shape is aerodynamically unacceptable, so we changed to a more streamlined shape which, because it uses more surface area to hold the same volume, consists of more fuselage fabric than the baseline. More fabric means extra weight of course, which will mean needing a larger volume of lift gas to achieve buoyancy. Additional volume leads to a larger fuselage, more fabric, more weight and thus a further increase in volume, and so on. This can be countered in two ways – first, use a stronger fabric, although stronger usually means heavier and that can feed the weight spiral; and/or second, determine whether there is a ‘sweet spot’ in the design space where, instead of the weight spiral diverging, it converges on the optimum combination of fabric, shape and weight. Using the CAD model – which, by including the material properties of all the aircraft’s components, gave us total weight – plus the aerodynamics results from the CFD calculations, and the loads results from the FEA calculations, an iterative optimisation process was performed that settled on a fuselage shape based on a NACA 0030 aerofoil. The NACA 0030 section chosen has quite a large thickness ratio, but the numerical simulations suggested that there would be no flow separation over the angle-of-attack range (the angle between the fuselage centreline and the oncoming air) envisaged for the aircraft’s operation. However, wind tunnel tests were performed on a scale model of the provisional design to prove whether that was the case. Consequently, tests were performed at one of the facilities at the University of Sheffield, and at the large low-speed wind tunnel at Perth College, one of the campuses at the UHI. The latter tests used non-intrusive, laser-based particle image velocimetry. The determination of the overall configuration of the aircraft included the sizing and positioning of the stabilising and control surfaces. The wings were designed to provide sufficient area to provide some aerodynamic lift should it be needed, and to provide enough of a moment arm for the wingtip-mounted ailerons to deliver sufficient control of roll in various pitch and yaw angles. The sizing of the vertical and horizontal tails, together with the rudders for controlling yaw and elevators for controlling pitch, was driven by the need to maintain stability and control in a sidewind. The vertical tail provides directional stability, keeping the nose aligned to the correct yaw angle to achieve the desired flight direction. However, in a sidewind it also acts as a weathercock, the sideways component of the wind trying to push the tail around so that the nose points into the wind. That is countered by the rudder, which deflects to provide a force opposing the weathercocking and maintain the correct orientation of the nose. The size and shape of the vertical tail and rudder were designed to be able to counter a sidewind of up to 22.5 m/s, above which the aircraft would be allowed to weathercock and suspend its mission. To maintain simplicity the wings, the horizontal and vertical tails were all the same constant-chord, constant-section shape using a NACA 0012 aerofoil section. The ailerons, rudder and elevators were all of the same size and shape, based on the crosswind requirement which was the most demanding criterion. The positioning of the wings and the centre of gravity (CoG, derived from the CAD model) were calculated so that the aerodynamic centre and CoG were co-located with the buoyancy centre (equivalent to the fuselage’s centre of volume). This requirement for the position of the CoG also determined the position and alignment of the gondola containing the battery, power management system, flight control computer, pumps and valve. Materials and construction The material for the fuselage is a Vectran-based woven fabric (LCP706) with a thermoplastic polyurethane (TPU) coating. Vectran is an aromatic polyester spun from a liquid crystal polymer in a melt-extrusion process. This process orients the molecules along the fibre’s axis, resulting in a high-tenacity fibre which, weight for weight, is five times stronger than steel and 10 times stronger than aluminium. Various TPU coating thicknesses were tested to determine the minimum required to provide sufficient helium retention, settling on 65 g/m 2 . In conventional fabric coating processes, there is the possibility of a ±15% weight variation, so project materials specialist Banks Sails worked with the manufacturer (Arville) to provide a more accurate process, finally achieving a ±5% variation, although this was a more expensive operation. The fuselage was constructed from Unmanned Systems Technology | December/January 2020 The lift gas – helium – is contained within the fuselage, so the volume of the fuselage determines the aircraft’s buoyancy and thus its lifting capability