How To Replace Oil as a Lubricant in Worm Gearboxes With Grease

For worm gearboxes, oils are mainly used in combination with splash lubrication. Consistent lubricants, such as greases, are used less frequently and are used, for example, in gearboxes where it is not possible or difficult to seal the housing . These applications include positioning gears, but also power gears with high sealing requirements, such as in the food industry. Grease lubrication has, in comparison to oil, a negative thermal effect on the worm, resulting in higher mass temperatures . However, this effect becomes less relevant in the application of positioning gears, so that the advantages of grease lubrication outweigh the disadvantages in these applications. At certain operating points, consistent lubricants also offer additional advantages over the more common oil lubrication. For example, in worm gears with low rotational speeds and sliding speeds, friction and wear can be reduced when using grease instead of oil . Grease lubrication places different demands on the calculation and design of gearboxes than oil lubrication. This applies to both the housing design and the design of the gearing. In terms of rheological behavior, grease can be described as a non-Newtonian fluid, which means that the viscosity varies with the shear rate. To describe this behavior, there are various models available, such as the power law or the Bingham model. These models assume a solid behavior of the lubricant at low shear rates and viscous behavior at higher shear rates. In contrast to oil lubrication, there are currently no calculation methods available for predicting friction in grease-lubricated worm gears. Such a physically based calculation method for friction in a grease-lubricated tooth contact enables application-specific gear design and optimization regarding friction reduction. This work presents the foundation for calculating the friction in the worm gear contact, which is the main difference between calculating the efficiency for gearboxes with oil and grease lubrication. The presented investigation includes the determination of boundary friction as well as the calculation of the film thickness, which is essential for the relationship between boundary friction and fluid friction. The combination of these two friction calculations represents the mixed friction conditions that are present in the tooth contact of worm gears. The calculation methods will be presented exemplarily for one grease. In addition to those investigations regarding the calculation of the friction, the grease distribution inside the gearbox will be researched by using fluorescent particles. The grease distribution is a key factor for the gearbox design to avoid starvation in the tooth contact. The results will be presented for two different greases, which are different than the one used for the calculation methods. Although one of them is grouped in the same NLGI class as the grease used for the analytical description, according to Ref. 4, the main goal regarding the grease distribution is to avoid dead spaces and to create a good mixing of the grease. Summarizing this paper answers the question “How to replace oil as a lubricant in worm gearboxes with grease?” with two different approaches. On one hand, an approach for an analytical calculation method is presented, and in addition to that, experimental investigations on the practical design of the gearbox are shown.

Grease lubrication in worm gearboxes is only little researched so far in contrast to the area of rolling bearings, where grease lubrication is widespread . Monz showed with an experimental approach the potential for grease lubrication with a wide range of different greases and operating conditions of the worm gear showing upsides compared to oil lubrication in different areas . For worm gears, there is no physically based calculation method for friction in the grease-lubricated tooth contact available, which is necessary for application-oriented gearbox design as well as for the optimization concerning friction reduction. Because of this lack of tools to design worm gearboxes for grease lubrication, most gear manufacturers use the design of the gearboxes for oil lubrication and simply fill them with grease. This working method is sufficient to achieve the goal of high sealing requirements, but neglects the potential for optimization. The calculation methods used for rolling bearings, for example, to determine the rating life, use a simplification to calculate the film thickness for grease lubrication. The calculation of this value for oil lubrication is well researched and broadly used, for example, through the equations by Hamrock and Dowson . To determine it for greases, common sense in the practical calculation is to use the properties of the base oil of the grease and calculate the film thickness according to the model for oil lubrication. This method is, around high sum velocities, appropriate because the film thickness of grease converges to that of its base oil with increasing sum velocity . Nevertheless, there are approaches to describe the film thickness of grease lubrication in a more detailed way, again with the background of lubrication in rolling bearings. Morales-Espejel et al. present a different way to describe the dynamic viscosity of the grease, in contrast to just using the viscosity of the base oil. They determine an experimental correction factor that is applied to the base oil viscosity and depicts the behavior of the grease in a more precise way. The factor depends on the sum velocity and converges to the viscosity of the base oil with increasing speed, because of the converging behavior of the film thickness from the grease to the oil. This adapted viscosity can later be used in the existing formulas for calculating the film thickness for oil lubrication. Another alternative calculation method is presented by Cousseau et al., where the properties of the bleed oil of lubricating greases are used. The film thickness measurements on a ball-on-disk tribometer showed that the behavior of the bleed oil differs from the base oil and is more comparable to the actual grease. Their suggestion for the calculation of grease film thickness is to use the properties, mainly the dynamic viscosity of the bleed oil instead of the base oil, to get a better match to the actual behavior of the grease . Friction in the Tooth Contact The analytical description of the friction in the tooth contact consists of the description of boundary friction combined with fluid friction. The approach used for determining these two frictions is different from each other, as shown in the existing model for oil lubrication described in Ref. 9. Boundary Friction The boundary friction is calculated by using a characteristic diagram, which is determined with a two-disk test bench for every grease under various operating conditions. The overall setup of the testbench, as well as its functionality and the execution of the test, is described in Ref. 10. The varied parameters to achieve different operating conditions are shown in Table 1. The low sum velocity and the high slide-to-roll ratio SRR (up to 100 percent) can be explained by the conditions in the worm gear contact, as shown in Ref. 9, and these circumstances are modelled with the given parameters on the two-disc testbench. The two discs are analogous to the materials used in worm gears, made of bronze and steel. In a worm gearbox, the worm is usually made of steel, and the worm wheel is made from bronze.

During the tests, different coefficients of friction were determined for positive and negative slide-to-roll ratios (SRR). The difference between positive and negative SRR results from the different speed distribution of the two discs. Investigations with positive SRR were made with the steel disc rotating faster, and the values for negative SRR were determined with the bronze disc rotating faster. The primary focus during this investigation is on the values for positive SRR, which means that the steel disc is rotating faster. This condition is the main use case in real worm gear boxes because the worm is rotating faster than the wheel. The used grease is classified according to Ref. 4 into NLGI class 1, and its base oil viscosity amounts to 680 cSt. During the tests, traction curves are determined for various temperatures and pressures, resulting in a characteristic diagram for the examined grease. As an example, Figure 1 shows the traction curves of the examined grease at 20°C with two investigated contact pressures.

The traction curves show a temperature and pressure-dependent behavior of the grease regarding the coefficient of friction. There is also a visible difference between the states of positive and negative SRR. The measured coefficients of friction with the bronze disc rotating faster are higher than those with a faster-rotating steel disc. A possible explanation for this effect is the rolling friction, which always works in the same direction, regardless of the sliding friction. The effect that the coefficient of friction in the area of positive SRR decreases with higher pressure can potentially be explained by a thinning effect on the grease, resulting in lower fluid friction.