industry university crosstalk
play

Industry-University Crosstalk Andrew George Brigham Young - PowerPoint PPT Presentation

Industry-University Crosstalk Andrew George Brigham Young University Andy_George@byu.edu / 801-422-7878 1 Nov 2017 Production challenge - aero composites Commercial Airplanes Manufacture Backlogs Boeing/Airbus all models = 9K


  1. Industry-University Crosstalk Andrew George Brigham Young University Andy_George@byu.edu / 801-422-7878 1 Nov 2017

  2. Production challenge - aero composites • Commercial Airplanes Manufacture ✓ Backlogs • Boeing/Airbus – all models = 9K (2/’13), 12K (6/’15 and 9/’17) • ~8-10 years of production • Boeing 787 = 683 (Sept 2017) • Airbus A350 = 744 (Oct 2017) • Producing planes at 10-12/month (ATL/AFP/autoclave) ✓ 20 year forecast (to 2036, by Boeing) = 47K airliners / $6T • Trend towards higher composites content • Need to produce high performance composites faster • Global competition (JBRIC) = need to reduce costs 2 B RIGHAM Y OUNG U NIVERSITY

  3. OoA Out-of-Autoclave (OoA) ✓ Conventional oven instead of autoclave: • Lower costs: capital investment / energy / nitrogen pressurization • Lower cycle time • Higher propensity for voids, but recent R+D has helped (minimize amount and effect of voids) 3 B RIGHAM Y OUNG U NIVERSITY

  4. LCM Liquid composite molding e.g. resin transfer molding (RTM) and vacuum infusion (VI) ✓ Dry fabrics instead of prepreg: • Lower material costs / scrap • Less freezer storage requirements (less energy) • Greater flexibility in geometry and preforming (less post-processing) • Adds engineering challenge of infusion – Optimization of LCM process with simulation tools 4 B RIGHAM Y OUNG U NIVERSITY

  5. LCM PAM-RTM; Mesh provided by ESI 5 B RIGHAM Y OUNG U NIVERSITY

  6. “Effects of defects” - voids LCM = never homogeneous void distribution, so total void content in part is inadequate S. Sisodia, et al. J. Comp Mat 6 B RIGHAM Y OUNG U NIVERSITY

  7. NDI for void content (ultrasound) 7 B RIGHAM Y OUNG U NIVERSITY

  8. Void Characterization: xCT imaging z Collection of 1 mm voids in 90 ply x y z y = 90º 45º S. Sisodia, et al. ICCM20 −45º Biaxial yarns Voids Copenhagen x = 0º direction of B RIGHAM Y OUNG U NIVERSITY production

  9. Void Characterization: xCT imaging How processing and reinforcement affect voids in LCM Low-velocity through- thickness impact damage in UD weave. Damage propagates from S. Sisodia, manuscript voids that accumulate by the in preparation weft roving cross-over points 9 B RIGHAM Y OUNG U NIVERSITY

  10. Void Characterization: xCT imaging S. Sisodia, et al. quasi-static indentation damage, manuscript in preparation B RIGHAM Y OUNG U NIVERSITY

  11. Voids: Properties Correlation 11 B RIGHAM Y OUNG U NIVERSITY

  12. Void (bubble) Formation Size/location of voids is dependent on flowrate Optimal = when flowrate is equal between, and within the tows Ruiz, et al, Comp Sci Tech, 66 (2006) 12 B RIGHAM Y OUNG U NIVERSITY

  13. Void Measurement: photography Void measurement methods – macro lens photography and automated image analysis 13 B RIGHAM Y OUNG U NIVERSITY

  14. Bubble photography during infusion Carbon Weave 14 B RIGHAM Y OUNG U NIVERSITY

  15. Bubble photography during infusion Carbon Biax NCF 15 B RIGHAM Y OUNG U NIVERSITY

  16. Filling Simulation: Void Formation • Void formation coupled to filling simulation ✓ Optimize for lowest formation, or least detrimental shape/location B RIGHAM Y OUNG U NIVERSITY 17

  17. Filling Simulation: gate location opt. • Gate Optimization ✓ Optimize resin inlets and vacuum vents to prevent dry spots / reduce filling time Vent Inlet 18 B RIGHAM Y OUNG U NIVERSITY

  18. VI thickness gradient simulation • Simulate vacuum bag’s thickness gradient • Determine how long vacuum on inlet required to to reduce the gradient B RIGHAM Y OUNG U NIVERSITY

  19. Resin kinetics / viscosity modeling • Model the rate and extent of cure ✓ Rheometry and DSC testing 200 C 200 C Cure rate vs time 180 C 160 C Red = full model 180 C Black = experiment Green = model in literature Blue = simplified model 160 C Cure extent vs time 20 B RIGHAM Y OUNG U NIVERSITY

  20. Cure and fill simulation / optimization • Temperature gradient, exotherm, viscosity ✓ Following Hexcel advice: infuse at 80°C into 120°C tool; what if 200°C tool? 21 B RIGHAM Y OUNG U NIVERSITY

  21. Industry-Academia in Europe • Students are inexpensive R+D sources for all large companies ✓ Industrial thesis – time at both school and work ✓ Workforce development • Less of an IP barrier to publication ✓ Companies looking for publications more than patents • Co-authored by school and company • Thesis can be restricted to closed audience, as long as some of it is published B RIGHAM Y OUNG U NIVERSITY

  22. Industry-Academia in Europe • Cross-industry collaboration ✓ Seed-investment by government to bring multiple companies and universities together into a project ✓ Generic “demonstrator” part used to solve an industry -wide problem ✓ e.g. FP7-Infucomp • companies: Daher, ESI, Hexcel, Short Brothers, Israel Aerospace, INASCO, Piaggio • Institutes: ARMINES, SICOMP • Uni’s : Cranfield, Patras, KUL, Stuttgart B RIGHAM Y OUNG U NIVERSITY

  23. THANK YOU 23 B RIGHAM Y OUNG U NIVERSITY

Recommend


More recommend