Seepage barriers have been used extensively to mitigate seepage problems in dams and levees. Although the design of many of these dams and levees has been based on intact barriers, seepage barriers have been shown to be susceptible to deformation and cracking when high differential hydraulic pressures act across the barrier. Under certain conditions, these cracks can lead to serious seepage problems, which could potentially lead to the development of a low-resistance seepage pathway. Three scenarios have been identified where there is the potential for erosion to occur adjacent to a crack in a barrier: 1) erosion at the interface between a fine-grained soil and a course-grained soil, 2) erosion of overlying soil due to flow along a joint in bedrock, and 3) erosion of the barrier material. The objective of this study is to investigate the first mode of erosion and identify the conditions at which more serious seepage problems can develop. The research has been performed using a laboratory model to simulate conditions near a seepage barrier crack under the scenarios described above. The results from the laboratory testing were compared to finite element seepage models for each scenario to estimate the flow velocities near the crack. The flow velocities were compared to estimated critical velocities of the soil to asses where erosion is likely to occur. A comparison was made between the observed behavior in the model and the behavior predicted with the computer model. The results of the research will be used to develop a method to assess the potential for erosion to occur and develop into a failure mode based on conditions near seepage barrier cracks.

Abstract With the development of new technology and the stringent requirement for the international environmental regulation, the silver-oxide cadmium composites with the best contact performance will be prohibited due to the anxiety to cause cancer. Therefore, this study intends to develop a new contact materials with the best contact performance and cost. This study intends to use the silver powder (2~4£gm) as the major component, and the alumina ceramic powders (0.5£gm) as the minor component. The self-development mixer is used to uniformly mix silver powder with alumina powder, and then the powder metallurgy is used to make the electric contacts. Under the supply voltage and current, as well as the mixing conditions, the contact erosion amount, the arc energy, and the contact resistance are measured to investigate the erosion mechanism. Results show that the electric contact with 4%wt alumina ceramic powder has the best performance. Furthermore, the new electric contacts of silver base alumina composite material appear the acceptable performance compared with the commercial electric contacts.

Centrifugal separation is a crucial piece of technology, used in a wide range of industries. In cases where solid particles are included in the working medium, erosive wear of componentsinside a centrifugal separator is one of the reasons limiting its service life. In the present work, erosion of some candidate materials of metals, hard metals and polymers were tested to represent the erosive conditions inside a centrifugal separator, and to study the erosion behavior between different material groups. A literature study was conducted to give an overview of the field of erosion, and to evaluate which erosion parameters to consider for testing. A centrifugal erosion test rig was used for the testing, using natural sand, sieved to a size interval of 500 to 700 µm, as erodent, at both 90° and 30° impact angle. Every material was tested at 100 m/s impact velocity, while only metals and polymers were tested at 70 m/s. The erosion rates, volumes of material removed per mass of impinging erodent, were measured and used to rank the candidate materials within each respective material group. Surface analysis of the eroded surfaces were conducted, using SEM, to determine the mechanism of material removal. Relation between erosion resistance, erosion mechanism and material property were investigated.