Date of Award

8-19-2024

Document Type

Open Access Dissertation

Department

Mechanical Engineering

First Advisor

Michel van Tooren

Second Advisor

Paul Ziehl

Abstract

The current large scale application of thermoset composite materials in large, twin aisle, civil transport aircraft have resulted in significant performance enhancements, such as aerodynamic efficiency, weight reduction, and durability. These large aircraft are, however, produced in relatively low numbers. For the next generation smaller single aisle aircraft, large scale applications of composites are of interest in the airframe. Application of composites in single aisle aircraft will require not only proof of similar and/or significant performance improvements, but also proof of adequate manufacturing rates. One option to achieve both is large scale application of thermoplastic instead of thermoset composites.

Thermoplastic composites allow for high-rate manufacturing techniques, such as stamp forming, for high rate manufacturing of parts. In addition, thermoplastic polymers allow for repeated global and/or local melting and reconsolidation. Use of local melting and reconsolidation enables different types of fusion joining to form welds as an assembly method for joining parts. The alternative part manufacturing and assembly methods could enable performance enhancements and manufacturing rate improvements the aerospace industry is looking for in single aisle aircraft.

In this dissertation the applicability of induction welding for the reduction of mechanical fastening and replacement of adhesive bonding for airframe assembly is investigated. Induction welding, a specific type of fusion joining, has already been successfully applied in the aerospace industry for the assembly of carbon fabric reinforced PPS-based skins, spars, and ribs of control surfaces for business aircraft and military unmanned aerial vehicles. However, welds in these applications are partially welded, meaning the interface between the parts being welded is not welded full length (end-to-end) or over the full width of the interface. For the application of welding of primary structures, such as stiffened shells in the fuselage, induction welding will need to be further developed to weld UD-tape based PAEK materials and to deliver full length, full width welds.

The research discussed in this dissertation focuses on induction welding of UD carbon fiber reinforced PEKK. The influence of the fiber direction differences in the two plies adjacent to the weld surface is studied computationally and experimentally. The results show that interply differences between fiber angles less than 90 degrees intrinsically found in woven fabric-based materials can provide sufficient heat generation for melting. Additional experiments performed on welding of quasi-isotropic laminates showed that end-to-end welding and full width welding can be achieved. The end-to-end full-width welds were created by varying only the coil speed (velocity) along the weld. All other welding parameters such as welding current, coil shape, distance of the coil from the surface, pressure, cooling of the tooling and stacking sequence of the weld pieces, were kept constant. Welding was performed with the KvE /UofSC weld tooling and standard KvE / Dahar single turn coil.

A three-step procedure was developed to achieve welding recipes for such welds. Step 1: welding trials at constant coil velocity to explore differences in thermal profiles and peak temperatures along the weldzone. Step 2: division of the weld length in segments and variation of the coil velocity per segment until peak temperature achieved in all segments is at or below melting temperature. Step 3: the velocity envelope found in Step 2, is adjusted with the current setting such that end-to-end full-width welding occurs and peak temperatures are above melting temperature for the material along the weld length. The current setting is constant along the weldzone in all steps, only the current level is adjusted for the final step.

Photomicrographs and ultrasonic testing C-scans were used to investigate the quality of the welds at different positions along the length of the weld and classify defects/damages associated with the induction welding process. The defects seen in the micrographs were classified in different types. The occurrence of fiber distortions, ply slippage/wash out, deconsolidation and polymer degradation specific to welding will need more investigation.

A first impression of the variation of the process to contamination was obtained through trials to weld parts with interfaces contaminated with fingerprints. The resulting welding quality was influenced and showed the importance of proper cleaning of the surfaces prior to welding. However, no special surface pretreatments are necessary to achieve high quality welds as used for adhesive bonding.

Mechanical testing was performed on lap shear coupons to characterize the mechanical performance of the welds. It was shown that full-width induction welding can deliver 76% of static joint load carrying capability of baseline hot press co-consolidated lap-joints. The fatigue joint performance was found to be influenced by ply slippage welding defects.

The methodology developed in this dissertation was applied to a panel with stiffening element welded in two orthogonal directions. The flanges of the stiffener in one of the two directions were interrupted to allow the stiffeners in the other direction to cross. All flanges were successfully welded showing the feasibility of the technique for the assembly of realistic and complex structure.

Rights

© 2024, Jaspreet Singh Pandher

Download

Share

COinS