Jin, Xiaozhe
(2017)
Investigation of the interrelated effects of temperature and frequency on fretting wear.
PhD thesis, University of Nottingham.
Abstract
This thesis concerns both experimental and modelling investigations of the interrelated effects of temperature and frequency on fretting wear. Experiments were conducted using a cylinder-on-flat contact configuration with 304 stainless steel under various environmental temperatures and fretting frequencies. Analytical and finite element (FE) models have been developed to predict the thermal field near the fretting contact, the friction behaviour, the plastic deformation and the wear scars.
To study the effect of frictional power dissipation on the thermal field near the fretting contact, an FE model, which has a predictive capability for the local temperature rise induced by fretting wear of the cylinder-on-flat contact, was developed. Analytical models are also presented to benchmark the FE model before it is applied to more complex situations. The effects of both the geometrical development of the wear scar and the debris layer development on the local temperature field has been analysed via the FE model, where it has been shown that under the test conditions examined (i) the development of the geometry of the wear scar can result in a significant (up to ~ 25%) reduction in temperature rise in contact, and (ii) the presence of a typical oxide debris layer can result in a significant (up to ~ 50%) increase in both the average and the peak temperature rise in the contact; (iii) the effect of both of these on the thermal field is only significant very close to the contact (to less than 0.2 mm for the effect of geometry change and to only a few micrometres for the effect of the oxide debris layer), which makes experimental measurements of these effects very difficult.
The influences of environmental temperature (room temperature – 275 °C) and fretting frequency (20 Hz – 200 Hz) on the mechanisms and rates of wear for the 304 stainless steel contact were examined, and were found to be mainly attributed to changes in the mechanical response of the bulk material and to changes in the behaviour of the oxide debris formed in the fretting process. At low temperatures, wear proceeds by continual oxide formation and egress from the contact, whilst at high temperatures, the rate of wear is much reduced, associated with the development of oxide formed into a protective bed within the contact. The temperature at which the change between these two behaviours took place was dependent upon the fretting frequency, with evidence that, at this transition temperature, changes in behaviour can occur as the fretting test proceeds under a fixed set of conditions. An interaction diagram has been developed which provides a coherent framework by which the complex, interrelated effects of these two parameters can be rationalised in terms of widely accepted physical principles.
Wear scars of very different shape have been obtained from the experiments under different temperatures and frequencies. Previously, it has been shown that the geometrical interaction of the wear scars on the opposing specimens can cause the variation of tractional force during the gross sliding phase of a fretting cycle which will therefore influence the derivation of the coefficient of friction. In fretting tests (where it is the displacement amplitude that is typically controlled), this variation in tractional force can also result in a change of slip amplitude. To find a better way to derive the coefficient of friction (which will be used to analyse the wear behaviour and be applied in the wear model) from the experimental measurements, the existing fretting scar interaction model was extended to include the stiffness of the fretting system and was applied to analyse such behaviour. The results show that: (i) both the tractional force variation across the slip region and the slip amplitude are influenced by the system stiffness and the geometry of the wear scars; (ii) the energy coefficient of friction (ECoF) increases with increasing system stiffness and the depth over width ratio of the wear scar. On this basis, an alternative method to calculate the contact coefficient of friction (i.e. the geometry-independent coefficient of friction (GICoF)) is proposed which is independent of system stiffness and developments in the geometry of the wear scars.
Temperature has been proven to influence the wear behaviour significantly through influence the plastic deformation in this work. To improve the simulation of fretting wear behaviour in this thesis, a finite element model with inclusion of plasticity was developed to simulate: (i) the wear scar developed at room temperature which did not form with a typical “U” shape and (ii) the wear scar with significant material build-up at the sides at 125 °C. The wear algorithm has been improved at the same time by introducing the newly developed dynamic correlation method (DCM) into the model which reduces the numerical error compared to the old wear algorithm. The results showed that the newly developed model simulated the bulk material wear behaviour successfully for both cases. However, the lack of third body simulation in the model still means that the model is not fully predictive, and it is recognised that future work must be directed towards this issue.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[img]](/style/images/fileicons/application_pdf.png)