Abstract
The treatment confines itself to the analysis and design of components compression-moulded in glass-reinforced polyester systems, with particular reference to fan wings and associated mouldings. The first of the three parts develops approximate analysis to express the behaviour of simplified standard fibre arrays. No claim for rigorous accuracy is made, the intention being to demonstrate the mechanics of the composite material rather than to express its performance in precise terms. Expressions are, however, derived for various parameters and carried into the second section for comparison with measured quantities. The second section deals with extensive mechanical tests carried out on standard fibre-matrix systems. The test pieces were compression-moulded under moderate control conditions to simulate commercial practice, so that some estimate of the variation in properties, to be expected in such conditions could be obtained. Under static test conditions this variation is found to depend upon the quantity being tested; tensile test results show, greater scatter in 'weak' than in 'strong' systems, due generally to the greater influence of matrix imperfections in the former categories, while shear results showed, considerable variation in all categories. Some variation, particularly in the latter case, could be accounted for by the simple test methods employed. Further tests included fatigue, creep and thermal expansion, but these were insufficiently exhaustive to show any decisive trends. Agreement with some of the formulae develpped in the first part is found to be good; in other cases an empirical correction offers reasonable correlation, while in some fields -particularly dealing with thermal expansion - no meaningful- basis for comparison can be established. The third section discusses the problems of design in the context of extensive testing of fan wings. Many of the test results unfortunately had to be discarded, either through irrelevance or through lack of information concerning the basic structure, all the test wings, having been moulded, elsewhere. It is also to be regretted- that the test programme was terminated before analysis was complete, so that no further tests could be carried out to resolve ambiguities in the final breakdown. Examination of failure modes, however, indicates clear errors in the design of the existing wings, and suggestions are made for an improved design. Further analysis makes it clear that the range of materials used in current production will not permit compression moulding of wings for fans greater, than 60 in. in diameter for 4-pole 50-c/s operation, or greater, than 48 in for 2-pole speeds, no matter how sophisticated the method of production may become. It is shown that design processes based upon the analysis generated in the first two parts may only be applied with advantage where the moulding charge is of the complex form required for high strength, implying an essentially preformed structure with little flow during moulding. Simpler charges for lower, strength duties may be formulated by simple rules, but can only develop optimum mechanical properties through, trial and error in the moulding sequence. No attempt is made to enter into detailed comparisons with the work of other experimenters. The basic theory in the first part indicates similar trends to those postulated in similar simple theory by Shaffer7, Corten8, Jones9 and Krenchel10, but the compression moulding practice using glass reinforced material in the second and third parts does not appear to have been covered elsewhere.