24th International Colloquium Tribology - January 2024 219 Limit Values for the Evaluation of Lubricant Analyses Stefan Mitterer OELCHECK GmbH, Brannenburg, Germany stm@oelcheck.de 1. Introduction The analysis of lubricants, but also coolants or fuels, is an established field for the condition monitoring of machines. This results in recommendations for action that are included in the maintenance of the machines. The analyses are used in the areas of R&D, maintenance or during operation at the end customer. Important requirements for monitoring are an analysis scope suitable for the application and the use of standardized test methods. An analysis in the laboratory can include 40-50 measured values per sample, which provide information on the condition of the lubricant and the machine itself. 2. Limit values and their benefit In order to derive recommendations for action from measured parameters, appropriate limit values are required as orientation in addition to the knowledge and experience of an evaluating tribologist. The limit values can be defined for the various categories on a laboratory report. These include: - Wear, e.g. iron, copper - Impurities, e.g. water, sodium - Oil condition, e.g. viscosity, oxidation - Additives, e.g. calcium, phosphorus - Various additional tests, e.g. cleanliness class, MPC, i-pH value Lubricant and fuel analysis are used in a wide range of industries. Roughly, this can be shown in the following overview. These are the branches in which OELCHECK is active with analytics: Diagram 1: Main branches for lubricant analyses This very large scope already suggests that the performance and evaluations of analyses must be adapted to the respective concerns. Obviously, there is no fixed evaluation scheme that can be used for all analyses in the same way. More precisely, the following aspects play an important role in the diversity of evaluation criteria that must be used in the evaluation of analyses: 3 Different applications require individual limits  no general scheme available 3 Stress level is very different (hydraulics, engines, gears, transformers…) 3 Within one application: “hydraulic system is not just a hydraulic system” 3 Wide range of surrounding conditions 3 Different requirements for the evaluation of the analyses 3 Optimizing the availability of machines 3 Avoidance of unplannend downtimes The question arises, which possibilities are available to define corresponding limit values and to ensure their relevance. For the development and implementation of such values, various procedures are presented in the presentation: 1. Limit values according to existing specifications 2. Limit values from the laboratory because of the data basis and the experience from many analyses 3. Limit values in agreement with customers and as a result of feedback from the field It must be ensured that the stored limits are automatically made available during the evaluation of samples in order to be able to provide the customer with a targeted recommendation. With the help of a sophisticated system, these parameters can be used for general applications (e.g. gearboxes of wind turbines) as well as for a single machine. The system for processing must be easy and intelligent to ensure adjustments at any time. However, the requirement for many lubricant analyses is not only to evaluate according to fixed limit values. Information such as oil operating time, oil volume or the trendline of samples should also be included in the evaluation to avoid misinterpretations. 3. Complex analysis options With the help of limit values not only individual samples from individual machines can be analyzed and evaluated. Rather, evaluations of large amounts of data can then also be carried out. However, an important prerequisite for this is to have a stable data structure available. This includes correct information about the machines, the lubricants used, running times, etc.. If you want to evaluate not only individual laboratory reports, but also a large number of samples and associated machines as part of a data analysis, it is important to have knowledge of the existing data structure. When important information about the application or the lubricants is missing, a Big Data analysis becomes more difficult. If information on the oil types used is missing, a more in-depth evaluation with regard to changes in fresh oil-specific val- 220 24th International Colloquium Tribology - January 2024 Limit Values for the Evaluation of Lubricant Analyses ues (additives, viscosity-index, changes in the infrared spectrum...) can only be partially possible. If the above-mentioned prerequisites are met, extensive evaluations can be implemented on the basis of the analysis data. Various concerns and questions can now be looked at more closely for customers. Some of these questions are, for example: Which machine types often show conspicuous analysis values? Which analysis values are mostly critical? What correlations can be derived from this? These are just a few concerns that may be of interest to a maintenance or service department. As shown in diagram 2, it is possible to evaluate which values were mostly critical in the analyzes and which conclusions can be drawn from them. In the case in question, the customer’s machines frequently have high water values, which result in reactions and thus problems with the additive package (phosphorus). The problem here is therefore not primarily due to overheating of the oil or mixing of lubricants or mechanical wear, but was primarily to be found in defective seals in moisture entering. In a further evaluation, individual machine types of the customer could be identified which were particularly affected here. This makes it possible to identify which adjustment screws on machines need to be addressed and optimized in order to plan service activities better or to avoid unexpected problems. Diagram 2: Critical values in a machine fleet 4. Meaningfulness The definition of limit values is an important tool for ensuring that lubricants remain in use for as long as possible. Ultimately, this also contributes to the conservation and sustainable use of resources. Using practical examples, the presentation will show procedures for creating and optimizing threshold values, as well as how to deal with these values during evaluation. Furthermore, the digital possibilities for the evaluation of a large amount of analytical data will be shown.