JEC COMPOSITES MAGAZINE - Issue #70 - January/February 2012 - 31

epoxy resin utilizing an obsolete production wind turbine blade root ring insert mould, a large-scale infusion trial was conducted. Root ring insert moulding trials were conducted using the Hexion EpiCote® 135i/ Hexion EpiKure® 1366 epoxy resin system mixed as reported previously, the Reichhold Dion 9102-75 vinyl ester/Syrgis Norox® CHP mixed as reported Fig. 3: Epoxy, vinyl ester and polyurethane root ring mouldings previously, and the Bayer systems being evaluated. The thick 840871/840859 polyurethane system laminates were constructed as follows: mixed as reported previously. Each resin the reinforcement package consisted of system was infused into a 21-ply laminate 12 plies of E-glass fabric measuring 330 configuration in the root ring insert mm x 2,440 mm consisting of 8 plies of mould consisting of 14 plies of Vectorply Vectorply E-BX- 2400-5,810 gm/m², ±45 E-BX-2400-5 (described previously) and degree bias fabric (BX) and 4 plies of 7 plies of Vectorply E-T-2900 (described Vectorply E-T-2900, 1,000 gm/m², unidipreviously) in a balanced configuration of rectional fabric (TX) plied in a balanced (BX/LT/BX)7. laminate configuration (BX/TX/BX)4 For each infusion resin system, the glass with the unidirectional fabric (TX) fabric package was double-bagged under positioned perpendicular to the direction 70 millibars vacuum at 23°C for 12 hours of resin flow. A single ply of Airtech flow before infusion. Each resin was mixed media measuring 330 x 2,440 mm was and immediately degassed for 5 minutes placed on the top of each glass fabric at 70 millibars prior to infusion. In the stack. Each reinforcement construction case of the polyurethane resin system, was double bagged with 0.08-mm nylon mixing was achieved through a Langeman film and placed under a vacuum of 70 meter/mix unit to minimize the millibars from a central vacuum system mass effects of the polyurethane exothrough a multiport manifold to ensure therm during the infusion process. The that the level of vacuum was consistent infusion set-up for this trial consisted of and equivalent for each resin system two resin inlet ports feeding a spiral tube being infused. After mixing each resin laying along the entire part front edge and according to the manufacturer’s direcextending partway along the flanges, and tions as reported previously, the infusion a single vacuum port at the part’s lower was initiated simultaneously for each rear centre. In each case, the resin infusystem. sion took place at room temperature (22The infusion times and flow distance as 24°C). Once the infusion was complete, well as resin temperature were recorded the resin inlet port and the vacuum port at various intervals. were sealed and the mould was heated at Test samples were sectioned from each a rate of 2°C per minute to a temperature laminate at the 457 mm, 1,067 mm, and of 60°C. The infused part was allowed to 1,981 mm lengths for void content analyremain at 60°C for 12 hours before coosis per ASTM D-2734. ling to room temperature for debagging and demoulding. Production part scale-up infusion trials Each moulded root ring insert was placed Continuing the characterization of on its side in an unconstrained position polyurethane resin versus incumbent for 24 hours at room temperature as

shown in Figure 3, before being evaluated and their properties analyzed. The dimensional properties of each root ring insert moulding were then measured and compared to the mould, and the parts were then subjected to a static load/ deflection test to compare relative stiffness and stiffness retention. A 1,180 kg load was applied to the centre of each moulded part in 45 kg increments and measured for incremental deflection using a digital deflection gauge contacting the centre underside of the part. The test protocol for this load/deflection testing is shown in Figure 4.

Fig. 4: Static load/deflection testing of root ring insert mouldings, 1,180 kg applied in 45.5 kg increments

Test procedures All testing was conducted on composite samples prepared using 6 layers of biaxial glass fabric, at 45° to the fibre direction. Fatigue testing was conducted using a method described by Grimmer et. al.5 and tested at Accutek testing laboratory using a servohydraulic test frame, bank 4-axis 1; load cell 1010ACK-5k, S/N 334937A. Dog-bone test samples, 20.32 cm x 2.86 cm wide, 2.54 cm gauge length and with 0.64 cm hole for stress concentration were mounted in the test frame using compression grip, and loaded axially in tension at a rate of 3 Hz with an R-ratio of 0.15 in force control. Tests were run in ambient lab conditions and continued until 1 million cycles or failure. G1C interlaminar fracture toughness tests were run in accordance to ASTM D5528 at Accutek laboratory.

Feature Wind Energy

Results
System development and viscosity

No70 January - February 2012 /

jec composites magazine 31

JEC COMPOSITES MAGAZINE - Issue #70 - January/February 2012

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #70 - January/February 2012