A Finite Model of Bi-Stable Woven Composite Tape-Springs
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesisAlternative title
En finit elementmodell av bi-stabila tejp-fjädrar av vävda kompositer. (Swedish)
The recent development of CubeSat nano-satellites shows that it is an effective way to send a payload onto orbit, as it is a relatively inexpensive and quick access to space. If the small size of these satellites is their main advantage, it is also the principal source of problems when it comes to designing it. The control electronics, electric power system and the payload are limited in mass and have to t in a tiny ten-centimeter cube. The necessity of a compact deployable structure to hold the payload once the satellite reached its orbit is one of the principal subject of study for the design of a CubeSat. In the CubeSat program SWIM (Space Weather using Ion Spectrometers and Magnetometers) that KTH takes part in, the deployablestructure developed consists of bi-stable semi-tubular booms made by a woven-composite fabric. Preliminary tests show that this structure is very compact and stable in the packaged con guration while being suciently long and stiin the deployed con guration. However little is known about the deployment phase, the physical model of the booms is very inaccurate in determining the deployment force and speed, because of the complexity of the material mechanics behind it. Modeling a woven composite material in a nite element analysis software is a dicult task due the structure of the material itself. The ber yarns interlace each other like in textile material, and they are impregnated in a soft matrix resin. Although in-plane properties of these materials can be calculated accurately using the classic lamination theory (CLT), the corresponding out-of-plane properties lack any accuracy for one-ply woven composites. Solutions are found through micromechanical approaches but these models are dicult to implement and are computationally expensive. The solution to this problem is to decline the CLT model of the material in two versions, each with a its own purpose. This paper presents rst a CLT model of the woven composite which aim is to predict in-plane properties accurately and giving a good estimation of the out-of-plane properties. The second version of the CLT model is developed with the aim of predicting accurately the amount of strain energy stored and the stable radius of the rolled-up con guration. The purpose of this version is to be used in deployment analysis. This paper also presents the main lines of a fully parameterized nite element model of the deployment analysis for future use.
Place, publisher, year, edition, pages
Engineering and Technology
IdentifiersURN: urn:nbn:se:kth:diva-104676OAI: oai:DiVA.org:kth-104676DiVA: diva2:566653
Tibert, Gunnar, PhD, Docent