Already in 1978 I have performed wind tunnel tests on a 2-bladed rotor with a diameter of 1.8 m and with tapered PVC blades made out of a PVC pipe. However, the rotor suffered from flutter already at very low wind speeds and therefore I would not advise to use PVC for rotor blades. Flutter is a combined bending and torsion vibration and the main cause for flutter is a too low torsion stiffness. The torsion stiffness is low if the thickness of the pipe is low with respect to the chord and the blade length and if a material is used which has a low modulus of elasticity like PVC. The measurements are described in the Dutch report R 343 D of the University of Technology Eindhoven but this report is no longer available. In November 2016, I have made a new report KD 616 called: "Translation of parts of report R 343 D of June 1978 from Dutch into English. R 343 D gives wind tunnel measurements for a rotor with tapered blades made out of a cylinder". This report can be copied for free from my website:
www.kdwindturbines.nl at the menu KD-reports.
Three different materials have been tested, PVC, aluminium and stainless steel. PVC has been tested for a thickness of 3 mm. Aluminium has been tested for three different thicknesses 2 mm, 3 mm and 4 mm. Stainless steel has been tested for 2 mm. Only the rotor with stainless steel blades gave an acceptably high flutter speed so only these measurement are given in KD 616. But in chapter 2 of KD 616, the problem with the PVC blades is described shortly. The problem was described more in detail in the original Dutch report R 343 D. So I now will give a translation of this description.
"It was started with a model made out of 3 mm PVC made from a standard PVC pipe (with an outer diameter of 300 mm). This went totally wrong because the rotor started to vibrate above a wind speed of 3 m/s and it rotated only very slowly. If the wind speed was increased op to 14 m/s, the rotational speed wasn't increasing but the amplitude of the vibration was enormously and the blade bent backwards like an umbrella. This is probably caused by the low modulus of elasticity of PVC and the high density".
So PVC blades give flutter problems if the torsion stiffness is too low. The torsion stiffness can be increased if thicker material is used. However, the aerodynamic characteristics of a cambered airfoil are only acceptable if the thickness is small with respect to the chord. So if you increase the thickness, the airfoil will get more drag. For the 4 mm aluminium blades, we have tried to modify the cambered airfoil into about an Eppler airfoil (Gö 804) but this requires modification of the airfoil nose and sharpening of the airfoil tail which is a lot of work. For the final stainless steel model the airfoil nose and tail were only rounded.
In public reports KD 617 and KD 656 I give the methode how to design a rotor with tapered blades using a cambered airfoil with increasing camber for decreasing radius.