Effect of Rapid-Prototyped Airfoil Finish on Loading at Low Reynolds Numbers

Abstract

T HEproliferation in the availability and implementation of rapidprototyped (RP) models for wind-tunnel testing has increased significantly in recent years. Marked time and cost saving as well as increases in model complexity may be achieved with little effort compared to traditionally machined models. Consequently, especially in an educational setting, rapid-prototype manufacture has achieved a notable foothold. Rapid-prototype manufacture may include the fused depositionmethod (FDM) using either acrylonitrile butadiene styrene (ABS) or poly ether ether ketone as the build material, stereolithography (SLA) using a photopolymer resin that hardens under ultraviolet laser light, and selective laser sintering (SLS) that uses a powdered deposition of the build material that is fused in layers, typically by a carbon dioxide laser [1–4] and then sintered to form the model. All methods build the model up in layers. Generally, FDM machines build from a ground platform up in horizontal layers through deposition of a plastic bead generally 0.01 in. in diameter. Support material (dissolvable in a lye bath) is also used to ensure model stability during the build. SLA builds the model fromavat of photopolymer epoxy resin in layers, by hardening selected elements on the layer through laser impingement. This layer is then dropped down and the next layer is drawn and solidified above it. SLS uses an analogous approach; each build layer is created by a laser that fuses a powdered deposition of the build material. The model is constructed through successive powered layer deposition and selective fusion. However, rapid-prototype manufacture is not without issue; model fidelity and finish is a major concern with models typically displaying a stepped appearance as they are constituted of finite layers. Studies examining the suitability of rapidprototype manufacture for wind-tunnel test models have suggested that the technique may be suitable; however, model rigidity, dimensional accuracy, and surface finish may be of concern [1,2]. These studies did not provide details on postmanufacture finishing (sanding, etc.), if any. A study by Aghanajafi et al. [5] investigated surface-finish effects through variation in the layer thickness of models produced using SLA. Their results indicated that greater layer thickness impacted the axial force (and consequently drag) to the greatest extent. The focus of this study is on the FDM technique, due to its widespread use educationally. Common FDM RP machines have a build resolution of 0.01 in. (0.254 mm). Individual layers of deposited plastic leave ridges that meld to form the surface. The ridges are circular in form and have a diameter close to 0.01 in. To eliminate the ridges, model finishing normally incorporates sanding, painting, or a combination. This, however, raises potential issue. The plastic is relatively soft such that the final resulting model form may no longer be representative of that desired; consequently, airfoils, etc., may be deformed. As a result, an experimental study was undertaken to establish the effect of RP model surface finish. A low-speed airfoil section, an S8036, was tested in various stages of surface preparation. Finishing was accomplished using paint and sanding. Finishing in general, for the wings examined, took 3–6 hrs each. Great care was taken to conserve the airfoil shape and minimize deformation; this may not typically be the case. The results from the wind-tunnel tests are reported in this paper.