Abstract

The U.S. Department of Transportation's Pipeline and Hazardous Materials Safety Administration (PHMSA) is responsible for ensuring the safe, reliable, and environmentally sound operation of the nation's natural gas and hazardous liquid pipelines. Regulations adopted by PHMSA for gas pipelines are provided in 49 CFR 192, and spacing requirements for valves in gas transmission pipelines are presented in 49 CFR 192.179. The present report describes the findings of a scoping study conducted by Oak Ridge National Laboratory (ORNL) to assist PHMSA in assessing the safety impact of system valve spacing. Calculations of the pressures, temperatures, and flow velocities during a set of representative pipe depressurization transients were carried out using a one-dimensional numerical model with either ideal gas or real gas properties for the fluid. With both ideal gas and real gas properties, the high-consequence area radius for any resulting fire as defined by Stevens in GRI-00/0189 was evaluated as one measure of the pipeline safety. In the real gas case, a model for convective heat transfer from the pipe wall is included to assess the potential for shut-off valve failures due to excessively low temperatures resulting from depressurization cooling of the pipe. A discussion is also provided of some additionalmore » factors by which system valve spacing could affect overall pipeline safety. The following conclusions can be drawn from this work: (1) Using an adaptation of the Stephens hazard radius criteria, valve spacing has a negligible influence on natural gas pipeline safety for the pipeline diameter, pressure range, and valve spacings considered in this study. (2) Over the first 30 s of the transient, pipeline pressure has a far greater effect on the hazard radius calculated with the Stephens criteria than any variations in the transient flow decay profile and the average discharge rate. (3) Other factors besides the Stephens criteria, such as the longer burn time for an accidental fire, greater period of danger to emergency personnel, increased unavoidable loss of gas, and possible depressurization cooling of the shut-off valves may also be important when deciding whether a change in the required valve spacing would be beneficial from a safety standpoint. (4) The average normalized discharge rate of {lambda}{sub avg} = 0.33 assumed by Stephens in developing his safety criteria is an excellent conservative value for natural gas discharge at the pressures, valve spacings, and pipe diameter used in this study. This conclusion remains valid even when real rather than ideal gas properties are considered in the analysis. (5) Significant pipe wall cooling effects (T{sub w} " -50 F or 228 K) can extend for a mile or more upstream from the rupture point within 30 s of a break. These conditions are colder than the temperature range specifications for many valve lubricants. The length of the low-temperature zone due to this cooling effect is also essentially independent of the system shut-off valve spacing or the distance between the break and a compressor station. (6) Having more redundant shut-off valves available would reduce the probability that pipe cooling effects could interfere with isolating the broken area following a pipeline rupture accident.« le Document type: Report