Since the introduction of electrical heat tracing, a major operations concern has been design considerations for adequately determining the heat loss of valves in piping systems. Analyses have been done on straight pipes and this part of the design is usually well understood and few problems arise with careful design and installation. Valves, supports, instrumentation, and other devices with "extended surfaces" (noted as "thermal features" in this paper), if not characterized properly, may cause problems with plant operations. The surface area of a pipe increases linearly with increase in diameter, but the surface area of a valve increases with the square of the pipe diameter for valves. This causes significantly increased heat loss and a lower maintain temperature. A frozen valve on the North Slope of Alaska instigated this paper, but the issues associated with this situation have wider application. A common valve heat loss factor, independent of pipe diameter, is sometimes recommended. Studies have found this causes small-diameter valves to be overheated and large-diameter pipes to be underheated. This paper investigates real heat loss of valve systems on the North Slope of Alaska. Two years of data (800 MB) were analyzed and interpreted, and dynamic experiments were conducted along with building a finite-element model. Thermal insulation of valves is shown to be an area of critical concern. Valves and other "thermal features" are usually thermally insulated with fabricated blankets with lacing or Velcro used to attach the insulation. This attachment method is used in order to allow access for maintenance. These blankets usually have a lesser thermal insulating capability (larger k factor) than the insulation for the straight pipe. The insulation thickness may also be less. The purpose of this paper is to highlight these parameters and make design recommendations for critical thermal features and large /spl Delta/T heat tracing applications.
Read full abstract