Abstract
Pipeline repair sleeves made of materials other than metals (composite sleeves) are commonly used in situations where welding is difficult or the cost of a steel repair sleeve is considered to be excessive. The use of these composite sleeves is based on a long history of development and testing by both industry and several research organizations.
Because these sleeves were developed for use on gas transmission pipelines without much consideration for the special conditions existing on hazardous liquid pipelines, there was reluctance to use them without further testing. For this work, the Westhollow Technology Center decided to test full ring specimens rather than individual lamina. This was driven by our conviction that the performance of a composite structure was likely influenced by variables introduced during assembly that could not be evaluated by testing of components. Subsequent testing confirmed this suspicion as we saw results that did not match either the published data or our own expectations. One of the main discoveries was a large variability in performance between supposedly identical sleeves. Differences were seen between the generic types of composite sleeves tested, between examples of the same model with different diameters and even between identical examples of the same product. In addition, at least three different modes of failure were discovered and several effects of long-term exposure to water were found.
