The content of this article presents research on used and fresh engine oils. The aim of the experiment is to preliminarily develop a method for assessing the condition of engine oil subjected to service. A four-ball tester was used to compare the lubricating properties of the engine oil as one component of the tribosystem under laboratory conditions. The method used to determine the mashing load consisted of subjecting the kinematic node to a linearly increasing load with a build-up rate of 409 N·s-1 under operating conditions of approximately 20°C and a spindle speed of 500 rpm. The presented article is a continuation of the consideration of the lubricating properties of engine oils subjected to operation. The tests carried out made it possible to observe that fresh oils are characterised by their ability to carry higher loads in relation to oils subjected to service. This is evidenced by the obtained values of scuffing loads, which have a higher value for fresh oils (The average percentage increase in scuffing load for fresh oils was 62.23%). Comparing the friction torque characteristics with each other, it can be seen that the values of maximum friction torque are also higher for the fresh oils group. The modelling process made it possible to characterise changes in the tribological properties of the lubricating oil being used. In the future, the described model will be extended to include further input parameters (viscosity, contaminant content, fractional composition, etc.), which will allow a multi-parametric assessment of lubricating oil wear.

A problem has been noted regarding the admixture of fuel to a low viscosity lubricant in hybrid electric vehicles (HEVs). This is very detrimental to the wear and tear of engine operating components. In this study, the operating conditions of HEVs were analysed. Using X-ray fluorescence spectrometry (XRF), engine oils of two different viscosity classes were compared after the operating process and these data were compared with fresh reference samples. Attention was paid to the content of elements such as Ca, Zn, Mo, Sn, Cd, Fe Pb, Si, Cu, and Ni. The depletion of anti-wear additives, as well as the higher content of metallic wear products relative to the operated 5W30 (engine oil), contributed to the overall assessment of the lubricity of the 0W30 oil, as well as to the tribological results. Then, under laboratory conditions, oil samples contaminated with up to 1 to 8% fuel were subjected to rheological (mini AV-X viscometer) and tribological (four-ball tester) tests. The dependence of the local pressure at the metal-to-metal contact point in the kinematic node on viscosity showed the dissimilar nature of the used and fresh oil and the divergence of the domains for the two groups of samples. Increasing the fuel contamination of used oil above 4% drastically reduces the pressure responsible for maintaining the oil film. In order to improve lubricant performance during HEV operation in urban conditions, it was proposed to carry out extra-urban traffic driving in order to evaporate the fuel from the engine oil. A shorter oil change interval is also recommended.

The research presented in this paper takes another step towards developing methods for automatic condition verification to detect structural damage to vehicle wheel rims. This study presents the utilisation of vibration spectra via Fast Fourier Transform (FFT) and a neural network’s learning capabilities for evaluating structural damage. Amplitude and time cycles of acceleration were analyzed as the structural response. These cycles underwent FFT analysis, leading to the identification of four diagnostic symptoms described by 20 features of the diagnostic signal, which in turn defined a condition vector. In the subsequent stage, the amplitude and frequency cycles served as input data for the neural network, and based on them, self-organizing maps (SOM) were generated. From these maps, a condition vector was defined for each of the four positions of the rim. Therefore, the technical condition of the wheel rim was determined based on the variance in condition parameter features, using reference frequencies of vibration spectra and SOM visualisations. The outcome of this work is a unique synergetic diagnostic system with innovative features, identifying the condition of a wheel rim through vibration and acoustic analysis along with neural network techniques in the form of Kohonen maps.

The article presents the results of tests, replaced according to the vehicle manufacturer's recommendations, of engine oils. The sample of engine oils in service came from spark-ignition and compression-ignition vehicles used in urban or mixed mode. During their collection, the type of drive unit, the mileage of the car and the number of kilometres the oil was used for were recorded for each sample (this was the main criterion for differentiating samples). In addition, a control group of samples consisting of fresh oils of the same viscosity grade and distributed by the same producer was set up to observe changes in the parameters of individual lubricants after the operating period. The first part of the empirical study consisted of determining the physico-chemical properties of the lubricants, i.e.: kinematic viscosity, density and water content. The second part involved anti-wear tests using a T-02U tribotester. The use of the tribotester made it possible to record the anti-wear parameter, i.e. moment of friction, and also the load imposed on the friction node, as a result of which it was possible to calculate the friction force and friction coefficient. The research was complemented by an analysis of worn surfaces of the friction node on a microscope. The tests carried out can be used for predictive purposes, in terms of assessing the condition of a lubricant subjected to an operating process in an internal combustion engine.