Sliding friction and wear of a continuous glass fibre/nylon composite against steel

Abstract

Nylon 6 has achieved wide use in many fields of bearing applications, especially as the lightly loaded bearing with poor or no lubrication. In many cases, it is used for its good dry bearing performance at low pv-factor (pressure-speed) with other desirable mechanical, physical and chemical properties [1, 2]. The possibility of adding internal lubricants to the base Nylon polymers allow their use as self-lubricated matrices for low performance bearing applications, but water absorption, thermal expansion and low loadability have limited their use in wet and high-load conditions. Various techniques have been used to modify their properties, one of which is to use fibres as reinforcement. Furthermore, Nylon 6 is a semi-crystallized polymer and its physical and mechanical properties are processing dependent. In this letter, an investigation of the sliding wear performance of continuous glass fibre (GF) reinforced Nylon (PA6) composites under dry sliding condition against a smooth steel surface is described. The main objectives were (i) to examine the wear performance of these materials subjected to different processing histories when tested at room temperature, and (ii) to study the effects of fibre orientation on the friction and wear behaviour. The GF/Nylon 6 commingled yarns, supplied by Toyobo Co., Japan, consisted of a 60:40 (wt%) mixture of two components. Unidirectional prepreg mats as well as composite laminates were manufactured within a steel mould using a heat press, as described in [3]. Each laminate contained 14 layers of GF/Nylon 6 mats, which produced a thickness of fully consolidated laminates of 4.5 mm with fibre volume fraction about 40%. The mould containing the GF/Nylon 6 mats was first heated to 245 °C, then a pressure of 1 MPa was applied over a period of 15 min. Depending on the thermal history preceding sliding wear tests, all of the consolidated GF/Nylon 6 composites may be classified into the following three groups: (i) rapidly cooled from the melt (I; approximately -60°Cmin-1); (ii) cooled in air at room temperature (II; approximately 5 °C rain-l); and (iii) cooled with the press machine (III; approximately 1 °Cmin-1). Sliding pin specimens, with a cross-section of about 4.5 m m x 5 ram, were cut from the consolidated unidirectional composite panels, and three fibre alignment directions were introduced regarding the sliding direction, i.e. parallel (P) along the fibre direction, normal (N) and antiparallel (AP). A pin-on-disc sliding wear machine was used to study the friction and wear performance of the GF/Nylon 6 composites. A mild steel (0.15% C maximum) was selected as the counterpart disc and its surface was grinded before testing by a grinding paper of 1200 grit, which produced a counterpart surface of roughness Ra = 0.04-0.05/~m. The applied pressure (p) on the sliding pins was 2.4 MPa, and four sliding speeds (v = 0.126, 0.168, 0.251 and 0.335 ms -I) were selected in the sliding tests. The friction torques were measured by a load transducer at the steady states [4] of wear performance, which produced the coefficients of friction. The depth wear rate, wt, was directly evaluated from height reduction of the sliding pins at the steady states, which was measured by a linear variable displacement transducer. It was found that the coefficient of friction was increased as a general trend when the sliding speed was increased. However, the variation of it within the range of the selected sliding speed was less than 20% and could be considered as a constant. Fig. 1 illustrates the relationships between the mean values of coefficient of friction, fibre orientations and the material processing histories. It can be found that with a given cooling history, the mean value of the coefficient of friction was almost independent of the fibre alignment directions. It seemed that the coefficient of friction was reduced as a general trend when the cooling rate was shifted from high to low. However, the reduction was very small and within scatter bands of the testing data. It was reported that /~ = 0.37 for an extruded Nylon 6 polymer when