How Human Error Wrecked a Gearbox

David Trocel, Fertinitro
Tags: gear lubrication

A five-year old single reduction conveyor drive was taken out of service for a scheduled preventive maintenance inspection and was disassembled. Prior to the inspection (disassembly), the unit had been in operation with routine vibration based surveillance for a full year since the previous PM inspection. There was no indication of any mechanical defect based on the vibration-based condition assessments.

The craftsman was supposed to disassemble the gearbox, allow for a visual inspection and then reassemble the gearbox. After inspecting the gearbox, a supervisor decided that components were in sufficiently good condition for continued use and instructed the craftsman to clean and reassemble the unit. The only work performed on the gearbox was cleaning and an oil change. The gasket was replaced with one made in the company shop.

During assembly, the gasket was improperly installed in such a manner that the gasket blocked the vent port. Once the unit was put into operation, internal heat lead to expansion which, with the plugged relief port, lead to the pressurization of the gearbox, rupture of the oil seal and loss of all of the lubricant. The gearbox ran for 48 hours before catastrophic failure of the pinion teeth occurred, as shown in Figure 3.


Figure 3. Severe Wear on the Pinion Gear

As a result of the failure, the site initiated a triple-redundant visual inspection and check-off requirement for all machines disassembled for PM activity, wherein the machine must be inspected by three different personnel prior to final closure and assembly.

In the final synopsis, the site concluded the following:

  1. The root cause of the failure was breather port blockage,
    as shown in Figure 2.

    Figure 2. Vent Hole Behind the Gasket

  2. The mechanic’s homemade seal contributed to the failure,
    as shown in Figure 1.

    Figure 1. Gasket Blocking the Vent Hole

  3. Lack of supervision contributed to the failure.

  4. The mechanic’s low knowledge and skill level also contributed to the failure.

Additional Thoughts by the Technical Editor
A “depot level” disassembly is a thoroughly invasive procedure. Under the best of circumstances, start-up failure risk is high following a rebuild. Based on the apparent condition, without the invasive human intervention this unit would probably not have failed. What were the options that could have prevented this occurrence?

  1. Add oil analysis as a routine surveillance tool. The unit was under surveillance with vibration technology, a useful tool by itself. Routine oil analysis could have provided precisely the type of information that reliability management at the site required to decide to NOT disassemble the gearbox.

  2. Training in the use of the available tools. Vibration-based analysis conducted properly can tell much of the integrity of the rotating components, including gearing condition. Gear condition analysis is, however, a fairly complex and difficult exercise. Whether vibration- or oil analysis-based condition assessment is used, the workman and supervisors all need to thoroughly understand how to conduct the measurement and how to interpret the results in order to believe the analysis once the data is compiled.

  3. Precision repair and rebuild training. It is a seemingly small thing to make a seal at the site rather than purchase a seal designed for the unit from the OEM. If the seal were properly installed, it may have been irrelevant. However, a combination of poor understanding of the seal functional requirement, and a poor understanding of the need for precision in installation led to the failure.

  4. Review the philosophy of “disassembly for inspection.” The risk of catastrophic failure is great following a rebuild. Under the best of circumstances, it is difficult to rebuild anything without changing some conditions to a less desirable state. This particular failure aptly reveals the problem with scheduled replacement or disassembly of machines.