Metallurgical Analysis of High Pressure Gas Pipelines Rupture

Abstract

On 6 July 2004, two parallel-running gas pipelines (18-inch and 24-inch diameters), in the main transmission network of SNGPL (a gas company in Pakistan) were ruptured. The ruptures occurred in the early hours of the morning about 8 miles downstream of the compressor station AC-4. The ruptures were indicated by the increased gas flow at the outlet of AC-4 [1], first at about 0648 hours and then again about 20 minutes later. The gas escaping from the ruptured lines had caught fire, and the flames had also ‘affected’ a third parallel-running pipeline of 30-inch diameter, lying next to the 24-inch line. The metallurgical examination of the two ruptured lines showed that the 24-inch line was ruptured with the help of an explosive device that had been placed on the underside of the pipe. An examination of the 18-inch line showed that this pipe had failed as a result of the heating of the pipe-wall, presumably, by the flame emanating from the 24-inch line. These two observations clearly suggested that the 24-inch line was the first to rupture (by explosives), and the fire following this rupture had heated the 18-inch pipe to a temperature where its yield strength was unable to support the inside gas pressure. The 20 minutes time interval between the two ruptures was obviously the time taken by the 18-inch pipe to be heated upto the level where it started to yield. The 30-inch line lying next to the 24-inch line was affected to the extent that its coating had been burnt-off over a length of about 40-50 feet. However, the pipe did not exhibit any signs of deshaping or deformation what-so-ever. A replica metallographic examination indicated that the microstructure of the pipe was not measurably affected by the heat. It was thus decided not to replace the ‘affected’ part of the 30-inch pipe, but only to re-coat this affected portion.