PERFORMANCE OF A FILAMENT-WOUND GRAPHITE THERMOPLASTIC COMPOSITE RING-STIFFENED PRESSURE HULL MODEL

Citation
Ma. Lamontia et al., PERFORMANCE OF A FILAMENT-WOUND GRAPHITE THERMOPLASTIC COMPOSITE RING-STIFFENED PRESSURE HULL MODEL, Journal of thermoplastic composite materials, 8(1), 1995, pp. 15-36
Citations number
9
Categorie Soggetti
Materials Sciences, Composites
ISSN journal
08927057
Volume
8
Issue
1
Year of publication
1995
Pages
15 - 36
Database
ISI
SICI code
0892-7057(1995)8:1<15:POAFGT>2.0.ZU;2-7
Abstract
The design, analysis, manufacture, and hydrostatic testing results of a 24-inch-diameter ring-stiffened pressure-huh model are presented. Th e AS-4 graphite/PEEK cylinder is manufactured using a non-autoclave in -situ filament winding and tape laydown process. The cylinder incorpor ates 5 hoop-wound rings and a shell (hoop/axial construction) that are integrally wound during manufacture. Ultrasonic inspection in conjunc tion with optical microscopy of end rings indicates that high quality is achieved with a void content of less than 1%. Translation of coupon data into the structure is quantified by subcomponent tests that meas ure in-situ properties (A and B-basis allowables) including shell axia l compressive stiffness and strength and interlaminar shear strength. Analysis of the structure indicates that high interlaminar shear stres s exists in the ring fillet areas. Subcomponent tests with supporting analysis to design the test method indicated that this failure mode wa s not critical. Design and analysis efforts focused on midbay and endb ay performance. Axial compressive stress concentrations in the endbay were reduced by incorporating local increases in shell thickness and a steel insert ring that provided radial constraint of the shell to red uce the probability of an end-brooming failure mode. Steel hemispheric al heads were designed and manufactured to further reduce axial compre ssive stress concentrations in the endbay. The test model was instrume nted with strain gages and acoustic emission sensors and tested at Car derock Division, Naval Surface Warfare Center (CD-NSWC). Increments in pressure were followed by a 5-minute dwell time. Acoustic emission ce ased for all pressures except the last two increments at 5250 and 5500 psi. The collapse pressure of 5500 psi was within 3% of our B-basis p rediction. The cylinder weight-to-displacement ratio, W/D, was 0.212 ( unitless). Axial compression failure occurred in the cylinder midbay i n agreement with our analysis.