Shear Pin Failure Prediction Through the Use of Acoustic Emission Sensing and Analysis
Project ID: 8542
Principal Investigator: John Germann
Research Topic: Condition Assessment
Priority Area Assignments: 2015 (Renewable Energy and Energy Conservation), 2016 (Renewable Energy and Energy Conservation)
Funded Fiscal Years: 2015
Keywords: shear pin, acoustic emissions, testing, shear pin monitor
Can Reclamation use the science of acoustic emissions to predict crack propagation in turbine wicket gate shear pins?
Can Reclamation build from this technology to develop a practical shear pin failure detector?
Need and Benefit
Understanding the fundamentals of how a shear pin cracks and fatigue related issues of metal shear pin failures will lead to better methods to protect hydroelectric turbines from catastrophic failure and reduce maintenance requirements. Generation production can be significantly increased through the reduction of down time caused by an inadvertent failure of a turbine component.
In the past, when hydroelectric units were manually controlled, it was common to staff plants with a plant operator(s) twenty-four hours a day, seven days a week. A major element of the operator's job was to monitor the unit; listen for abnormal sounds and to take action to shut the unit down in the event of a shear pin failure. Today it is more common to remotely control plants or units. Most Reclamation plants are not staffed on weekends and some smaller, remote located plants only see occasional maintenance staff visits during the work week. This puts the power generation units at high risk.
It is necessary to detect as soon as possible the breaking of a shear pin and to identify and replace the defective shear pin to re-establish the proper function of the gating system. Oddly enough, shear pin failure detection devices are not commonly used in the industry. This is partially due to the limited designs for an effective shear pin failure monitor. There are two current detectors designs that are occasionally used on units. The first type uses a wire inserted in a bore at the center of the shear pin. The bore is filled with epoxy to hold the wire in place. The wire is connected electrically through an electric circuit. Each detector is connected in series or parallel with the others up to a common alarm point. When a shear pin breaks, the wire is cut and the loss of power triggers an alarm. Each pin is modified to incorporate the wire and connection. The second common design is to use air. Pressurized air is fed into a well machined into the center of each pin. When a pin breaks, the line supplying the air loses pressure. A pressure switch in the air circuit recognizes the loss of pressure and subsequently triggers an alarm.
The two existing designs have some inherent issues. Wires and air lines connecting each detector are sometimes snagged and wires cut or disconnected. Furthermore, the shear pins are often bathed in water, oil and grease. In the long run, these conditions finally attack the cables, hoses and the connectors. The air type system is also subject to chronic air leaks which create false alarms. Often this occurs from a cracked hose, loose fitting or a slight crack in the pin itself. Both types of detection systems require that the shear pin be modified to incorporate the design. This adds to the cost of the system and the production of replacement pins.
Additionally the electrical type system often has grounding problems and troubleshooting is difficult Since the wires are embedded in epoxy, it is then necessary to remove the shear pin completely, clean it from the epoxy and re-install a new detector.
An object of the present research is to explore the use of acoustic emission sensor(s) attached to a shear pin to detect the start of a shear pin crack prior to breakage. The ultimate goal of the research will be to design a detector which is non-intrusive to the pin, easily adaptable to any kind of shear pin, is robust, reusable, cost effective, and maintenance free.
Contact the Principal Investigator for information about partners.