Gear-drive inspection reveals the effects of residual stresses on face contact in high-speed gear units

High-speed standards usually require verification of the mating accuracy, and that this record be maintained. This entails recording profile, lead, and pitch deviations. The mating gears are considered as a matched set that must be checked for contact on a checking stand and then when installed in the housing. A thin coating of a material such as the popularly used Prussian Blue is applied at three locations 120º apart. Four or more teeth are covered on a dry degreased gear. The gears must be properly aligned with parallel shafts and at a set distance within the closest of tolerances. The gears are rotated with a nominal torque in their operating direction. Up mesh is preferred, pinion pulled upwards/gear downwards. The weight of any pulled-upwards rotor is countered by the tangential load. The acceptable contact pattern has been previously determined and made available to the purchaser’s representative. This is especially necessary as the manufacturer may have made lead modifications for torsion, bending, and temperature. Without modification—which is unusual with high-speed gearing—the contact area would be expected to cover 80 percent of the tooth face. All the records are expected to be maintained for 20 years.
The contact pattern is lifted from the teeth with an adhesive paper that is attached to a notated sheet that has been witnessed and signed. After the running tests at 110 percent maximum continuous speed at partial or even full power, the top half of the housing is removed and the gears are again inspected for any damage and further verification of the contact pattern. The gear’s performance is dependent on the designed contact pattern being maintained throughout its life. The complete drive and auxiliaries are designed for a minimum life of 20 years. It is also expected that when the unit is placed into service it will operate without interuption for three or even five years. It is generally accepted that face load factors greater than 1.1 can be used when the manufacturer can demonstrate that their correction factors for the elastic and thermal distortions expected to occur in service have been proven to provide the required contact pattern.
Some 10 years ago, after 10,000 to 20,000 hours, a gear unit had a tooth breakage at the extreme end of the pinion tooth. The cause was determined to be misaligned shaft centerlines that concentrated the tooth forces at the breakage point. The misalignment itself was caused by foundation distortion and the inexperience of the erection people—a classic case of failure due to improper tooth contact. The second incident on another unit occurred between 30,000-100,000 hours, when damage was seen in the mid-tooth section of the pinion. The gear wheel and pinion revealed that distorted geometry had concentrated the tooth forces in the midsection. These geometrical changes from residual stresses had happened over a prolonged time period. As a result an extensive inspection of numerous gear units was undertaken for the sake of collecting data. The size, pitchline velocity, hours in operation, previous inspection reports, and the profiles were recorded to create a service-versus-failure profile. Approximately 10 percent of all high-velocity drives were found to be in need of corrective action. This required realignment and, in some instances, regrinding of the profile.
Different designs of gear units were involved. The corrective action deemed necessary was a reeducation of both service and installation personnel, a tooth contact check after 2,000 hours of operation, and a change in manufacturing procedures. These include but are not limited to reduction of residual stresses by subsequent machining, low-distortion hardening methods, and new nitrided steels that provide increased hardness penetration depth. ME quality steels vacuum degassed with 5 grain or finer. Forging ratio ≥3.5 and further additional testing of mechanical properties. Gears are now operating at speeds and powers that until recently were considered impossible. Material is literally being stretched by centrifugal loads and, although the drives can still perform within the required criteria, more-frequent inspection of the contact pattern would seem a necessary requirement for these state of the art drives.
Previous articleQ&A with Robert Bruce
Next articleTerry McDonald: Site Safety
is former director of the National Conference on Power Transmission, as well as former chairman of the AGMA's Marketing Council and Enclosed Drive Committee. He was resident engineer-North America for Thyssen Gear Works, and later at Flender Graffenstaden. He is author of the book Design and Application of the Worm Gear.