An analysis of unsteady modes present in the flowfield of an airfoil with a leading-edge horn-ice shape was performed in the current study. An NACA 0012 airfoil was tested in a subsonic wind tunnel at Re = 1.8 × 106. In addition to the clean configuration, the airfoil model was also tested with a set of boundary-layer trips, a two-dimensional extrusion of a horn-ice shape casting, and an array of simulated icing configurations created using simple geometries. Time-averaged and unsteady static pressure measurements were acquired about the airfoil surface, along with unsteady wake velocity and surface hot-film array measurements. Additionally, surface and off-body flow visualization techniques were used to visualize the airfoil flowfield. A technique was also developed to determine the unsteady shear-layer reattachment location of the ice-induced laminar separation bubble downstream of the horn-ice shape using the surface hot-film array measurements.

The maximum amount of unsteadiness in the iced-airfoil flowfield was observed to increase with increasing angle of attack. For a fixed angle of attack prior to stall, a change in the feature height of the simulated ice shape led to a change in the distribution of flowfield unsteadiness, but did not change the maximum levels of unsteadiness present in the flowfield. The iced-airfoil flowfield unsteadiness was primarily associated with three different frequencies. The first was represented by an increase in spectral energy across a broad-band frequency range, and was observed just upstream of shear-layer reattachment as well as downstream of shear-layer reattachment. This increase in spectral energy was caused by the regular mode of unsteadiness due to vortical motion in the separated shear layer and vortex shedding from the separation bubble. The average Strouhal number of this regular mode corresponded to StL = 0.60, and the average vortex convection velocity was observed to be 0.45U_inf. These values were highly consistent with those reported elsewhere in the literature.

The other two frequencies were much lower and were observed as narrow-band peaks in the spectral content of the acquired measurements that were primarily present in the region covered by the ice-induced separation bubble. The first was attributed to the shear-layer flapping phenomenon and was particularly dominant in the upstream portion of the separation bubble. The Strouhal number associated with this shear-layer flapping mode corresponded to Sth = 0.0185, which was consistent with those reported in studies of separation bubbles about canonical geometries. The second frequency was lower than that of shear-layer flapping and was associated with a low-frequency mode of unsteadiness that can occur prior to static stall for airfoils of thin-airfoil stall type. This low-frequency mode was characterized by a low-frequency oscillation of the airfoil circulation, and it was clearly identified in the spectral content of the iced-airfoil lift coefficient. The resulting values of Strouhal number exhibited a dependence on the airfoil angle of attack and corresponded to a range that was consistent with the Strouhal number values reported in prior studies of the low-frequency mode in the literature.

Using the method for determining the unsteady shear-layer reattachment location, the average time-dependent relationship between the reattachment location and the lift coefficient was calculated. It was discovered that at the low-frequency mode, the lift coefficient leads the shear-layer reattachment location by a phase of pi/2. This phase relationship occurred due to a feedback between the airfoil circulation and the separation bubble length. This improved understanding of the low-frequency mode in the iced-airfoil flowfield was utilized in a practical example to improve the predictive qualities of a hinge-moment-based stall prediction system. This improvement in the predictive qualities was performed by identifying the intermittent signature of the low-frequency mode in the wavelet transform of the hinge moment coefficient, which allowed the iced-airfoil stall case to be isolated from the other clean airfoil and leading-edge contamination configurations.