Open Cut » Maintenance & Equipment
All hydraulic piston pumps are high speed components with extreme manufacturing tolerances that produce high pressures and forces. With such fine tolerances, failure is inevitable, causing expensive unplanned downtime and maintenance. This project identified a way of predicting these failures so components can be replaced under planned maintenance conditions while obtaining full life from the piston pump. The system highlights a means to extend pump life through trending data that is not currently available.
Presently hydraulic pumps and motors are changed at predetermined intervals, not on the condition of the pump. These centre lines are usually set by the OEM but are tweaked by different sites to fit into their planned maintenance programs. Oil sampling is ineffective to determine the condition of a single pump as all oil returns to a common tank from which the sample is taken. Magnetic particle detection gives end of life a pump but not lead time for an upcoming failure hence there is no planning for changeout, and machine is in downtime. Vibration monitoring is also ineffective due to the variable displacement of the pump meaning the pump mass constantly changes which effects the resonant frequency of the pump. The constant movement of the excavator also creates undue noise with vibration monitoring.
The objective of this project was to install monitoring equipment on the hydraulic pumps on a large mining excavator. The monitoring equipment can be remotely accessed by the maintenance planner showing pump serviceability trend lines. The trend lines will give the maintenance planner the ability to change hydraulic pumps prior to failure, during a PM service interval. The pump will be at end of life but inexpensive to rebuild, with only a presumably barrel kit needed. The monitoring sensors that are installed give necessary data for pump health. The data is defined by the OEM in technical specs with min - max limits for pump end of life. By using the OEM specifications as a guide to pump end of life, the trend lines displayed give real time tracking of pump serviceability.
The pumps are monitored individually with commonly sourced pressure and flow meters. The data logging equipment can also be sourced within Australia but due to site regulations with entry and 24/7 operation a decision to build a data logging unit in-house was made.
It was found that there was an increase of case drain flow from the different pumps overtime, but the monitoring equipment was not on for a long enough period to witness either exponential increase of flow or a pump failure.