Performance of pin-loaded carbon fibre reinforced polymer straps at elevated temperatures

Abstract

Fibre reinforced polymer (FRP) composites are ideal for demanding structural engineering applications which require high-strength, stiffness, and durability. This is particularly the case for carbon fibre-reinforced polymers (CFRPs). In structural engineering, the use of CFRPs is reasonably widespread for providing additional (externally bonded) reinforcement to existing structural members, but CFRPs may also be used in standalone structural engineering applications. One such application is their use as tensile elements in the form of looped CFRP members (straps). Until now, all research that has considered CFRP straps (either in static or fatigue loading) has been performed at ambient temperature conditions. The absence of experimental data at elevated temperatures damages confidence when utilising CFRP straps in applications where elevated temperatures may occur. This thesis investigates the tensile performance of model, laminated, unidirectional (UD), titanium pin-loaded CFRP straps as temperatures increase up to 600 °C. In addition, it also investigates the fretting fatigue performance of pin-loaded straps at sustained service temperatures of 60 °C.

Two main experimental series are presented, including tests both at ambient and at elevated temperatures (in both steady state and transient thermal conditions). The UD CFRP straps were manufactured either as non-pretensioned – using an existing mould – or pretensioned during manufacture using a new pretensioning mould developed as part of this thesis.

First, the performance of pretensioned, laminated, UD, CFRP straps is assessed at elevated temperatures. Two types of tests are presented: (1) steady state thermal, and (2) transient state thermal. Nine steady-state target temperatures in the range of 24 to 600 °C were selected based on results from dynamic mechanical thermal analysis (DMTA) and thermogravimetric analysis (TGA). The transiently-heated pretensioned straps sustained three tensile load levels, namely 10, 15, and 20 kN, corresponding to 25%, 37%, and 50% of their ambient temperature ultimate tensile strength (UTS). The straps were able to retain about 50% of their ambient temperature UTS at 357 °C. Similarly, the strength degradation curve of non-prestressed straps at eight target temperatures between 24 and 410 °C was obtained.

To better understand the stress concentration regions within the straps leading to failure, an 1/8 finite element (FE) model is presented. A series of parametric studies is used to show that the most influential parameter is the pin geometry; an elliptical pin could decrease the stresses of the straps around the vertex area (at the contact region between the pin and the strap) by almost 19%. In addition, the strains of the outermost plies of the 1/8th scale FE model were compared to the strains obtained with optical fibre measurements made experimentally, and are found to be in reasonable agreement. Resultant stresses from the FE modelling are implemented in a failure criterion by Schürmann to predict the load at which first fibre failure is likely to occur; this is found to be in reasonable agreement (less than 3% difference) compared to the experimental load at which fibre breakage occurs.

The tensile performance of larger pretensioned CFRP straps is investigated following the same testing conditions as noted above (same thickness and width; longer straight shaft). Initially, two configurations of large straps are transiently heated for which their winding start/end point and a ±45o ply is located: (1) at the centre of the straight shaft length (inside the chamber) and (2) close to the vertex area (outside the chamber). The two configurations were chosen in order to determine whether the large straps’ performance with increasing temperature would be affected. The transiently heated large straps sustained the same three tensile load levels as the standard straps: 25%, 37%, and 50% of their ambient temperature UTS. It is found that the large straps with the start/end point of the winding and a ±45o ply nearer to the vertex area perform better in the transient tests. Subsequently, steady state tests of large pretensioned straps at eight target temperatures between 24 and 600 °C are presented, in which the straps’ winding start/end point with a ±45o ply was outside the chamber. Overall, the large pretensioned straps are able to sustain 50% of their ambient temperature UTS at 550°C.

With respect to the second experimental series, the fretting fatigue performance of laminated, titanium pin-loaded standard CFRP straps at ambient and elevated service temperature is examined. A frequency of 10 Hz and a load ratio of R=0.1 up to three and/or 11 million loading cycles are used in all fatigue tests in order to establish the maximum stress over the number of cycles to failure (S-N) curves. Prestressed straps are fatigue tested at ambient (laboratory) conditions at six upper stress levels (USLs) between 800 and 1300 MPa, while non-prestressed straps are fatigue tested at a slightly elevated service temperature (60 °C) at nine USLs between 650 and 1400 MPa. The fatigue limit observed from the S-N curve at 24 °C shows a slight increase of 50 MPa, when compared to the fatigue life determined from non-prestressed straps at 24 °C.

Additional tests at 11 million loading cycles could provide a complete picture of the results obtained in this thesis regarding the prestressed straps at 24 °C. Regarding the fatigue limit of non-prestressed straps at 60 °C (proposed maximum service temperature), it is concluded that the fatigue of the non-prestressed straps is not obviously affected at higher USLs, and only exhibits a slight decrease of 50 MPa when compared to the fatigue limit derived from the S-N curve of non-prestressed straps at 24 °C.

The findings of this thesis provide evidence for designers as to the tensile and fatigue performance of CFRP straps; this evidence can potentially be used to enable the application of such straps for bridge deck suspender cables or shear reinforcements for reinforced concrete slabs and beams.