Pits form from wear, electric erosion, chemical attack, thermal damage, or porosity that arises during material casting or processing. Grinding burns always lead to a reduction in fatigue life and pitting. In Failure Analysis for Hertz Contact Machine Elements Tallian states that spalling, pitting, or flaking are used indiscriminately in tribological literature for contact fatigue. The book has adopted spalling as the only designation for macroscale contact fatigue. The term pitting is reserved for the formation of craters by processes other than fatigue cracking. The gear industry, on the other hand, has clearly defined the various terms to be applied to each stage of Hertzian fatigue. AGMA’s gear tooth nomenclature designates all contact fatigue as surface fatigue.
Macropits are similar to micropits, but several times larger. Macropitting is frequently associated with flaking, although exfoliation can occur without growing into pits. When the pits grow they join together and form irregularly shaped craters that are frequently arrow-shaped and with a “tit.” The tit is a shallow projection from the front edge of the macropit. This is known as the crater stage. The phenomena can also be described in two other stages. Prior to the spall stage the macropitting can be in a prior progressive stage, or at a non-progressive level. Non-progressive pits commence as small pits < than .020”in localized areas. Pits can become > 0.032” and then cease after the gears have been run in. Flake macropitting has taken place when there exists large, relatively shallow, triangular shaped pits of constant depth over a large surface area. The fatigue cracks commence from a central point and extend outwards in all directions until thin flakes of material break away.
Two types of spalling are surface and subsurface. Repeated contact stresses induce surface or subsurface stresses resulting in cracks and loss of material. Macropitting can even occur with low stresses following a number of cycles, because there is no endurance limit. Accordingly, gears must also be designed based on life cycles. The macropitting life is extended by maintaining low contact stress, high-grade steel, and ensuring a suitable film thickness. The gear design can be selected with an optimum geometry. The steel will have a high hardness and strength combined with a honed or ground surface finish. Because gears are made with different hardnesses, fatigue resistance and surface finish the prevalence of macroscale contact fatigue compared to other failures cannot be established by a single variable, such as hardness. The resistance to all failure modes is influenced by hardness.
The macropitting can initiate both on the surface and subsurface at the location of an inclusion. The interaction between asperities and defects such as nicks or grooves will create surface-initiated cracks. High-speed gears have superior surface finishes and well planned lubricant thicknesses, so the macropitting is most likely to be subsurface. It will have started because of a subsurface inclusion that caused a stress concentration at that point. The lubricant must be clean since water or abrasive particles assist in developing the macropitting. Fine pitting over the entire tooth surface is caused by grain boundary oxidation.
On occasion rust particles appear as a result of water or chemicals. Large gear drives transported over rail lines across Canada were found to have a similar condition due to the constant vibration on static gears. Although not considered as a pitting failure, case crushing can have a similar surface appearance. The failure appears on only one or two teeth. In this type of failure the cracks enter the surface of the tooth normal to the surface and penetrate to the case core interface. At this point the cracks take a 90-degree turn and follow the case core interface, usually in both directions. No low hardness material is known to have a good resistance to pitting fatigue. The resistance increases in direct proportion to the hardness. When there is a sub-case failure the gradient of decreasing hardness is steeper than the gradient of Hertz shear stress then the strength to stress ratio is at a less favorable depth location below the normal maximum Hertz shear stress region.
An interesting case history involved a compressor drive. The spray was specified to be outgoing but was installed ingoing. As a compromise a set of sprays were supplied to be used for the ingoing mesh in addition to sprays for the outgoing. They arrived too late for the start-up. The gears were in continuous service for two years, and upon examination were found to be in perfect condition. The mechanic then installed the ingoing sprays instead of the existing sprays. Unfortunately they were sized for only a third of the required oil. Due to unusual noise the unit was shut down after 26 days and found to have extensive pitting on the pitch line. The reduction of oil from 18gpm to 6gpm affected the film thickness. Calculations for film thickness should consider oil flow quantity and the point of application in high performance drives. Unlike steel gears, worm gears remain functional after severe pitting.
