A freeze in time: an ASME post-construction standard leads a refinery maintenance team through an unfamiliar but efficient repair.
When an unexpected problem emerges, it triggers a reaction by plant personnel. Piping must be repaired, components replaced-and often these things must be done very quickly to minimize costly plant downtime.
But exactly what is the best response? In an industry where safety comes first, there is little appetite for attempting novel repairs that are not carefully studied--and when process plants are down, often there is no business case for the delay that careful study of untried alternatives would require, when there are better-understood repair options whose costs may be significant but are known.
Reluctance to innovate, however, could be costing companies substantial sums of money. It was to address this dilemma that ASME published a standard in 2005--PCC-2 Repair of Pressure Equipment and Piping--to guide plant personnel in the swift and safe execution of several lesser known but very valuable repair techniques, regardless of their experience.
One such repair technique that is well documented in PCC-2 is the use of freeze plugs, which prevent flow in pipes to allow for downstream maintenance activity. It was this section of the document that solved a critical problem and avoided a shutdown at a busy refinery on the West Coast.
During a recent maintenance shutdown at the refinery operated by Chevron in Richmond, Calif., routine work had been planned to dismantle a heat exchanger for inspection and refurbishment. Heat exchangers are key pieces of equipment for refiners. They control temperatures of process streams and recycle heat to make processes run more efficiently.
A typical exchanger will employ anywhere from tens to thousands of parallel tubes in a bundle, configured so that one process stream flows through the inside of the tube, and a different one flows over the outside of the tube, exchanging heat through the tube wall. Over time, the integrity of this pressure boundary--the tube wall--is compromised by corrosion, and when it becomes too thin, the tubes must be replaced. This exchanger needed its tube bundle replaced.
Once the plant was shut down, cleaned up, and prepared for maintenance work, operators discovered that a key valve normally used to separate the heat exchanger from its supply piping was broken beyond repair, and would no longer close. Without closing this critical valve to isolate the bundle, the maintenance on the heat exchanger could not occur, and the plant would not be able to return to service.
To complicate matters further, this valve was on an eight-inch diameter branch line off a 50-year-old cooling water utility system that services several independent plants at the refinery, and only this plant was scheduled to be out of service for maintenance at the time.
One way of taking the line out of service to replace the inoperable valve would require that the entire utility be shut down, along with all the plants it serves. Unplanned shutdowns of this nature usually cost refineries hundreds of thousands to millions of dollars, so plant personnel were eager to identify other ways to safely isolate the valve for replacement.
* evaluating and planning
Besides a general shutdown of the cooling water system, two other options for isolating the exchanger were evaluated: A hot-tap and stopple, and a freeze plug.
The evaluation revealed that a nitrogen freeze plug provided the best solution for this work due to its low complexity and low cost. The more familiar hot-tap and stopple--in which a welded sleeve is attached to the pipe, a hole is drilled through the pipe wall inside this sleeve, and an inflatable plug is inserted in the pipe for isolation--was determined to involve higher complexity, and was expected to be more costly as well.
A freeze plug, unlike the hot-tap, does not compromise the integrity of the pipe pressure boundary by welding or drilling. The freeze plug is created by installing a bolt-on jacket around the pipe through which liquid nitrogen (at -321[degrees]F) is circulated until the water in the line freezes. The concept is simple, but execution must be well planned.
The company's engineers were aware of freeze plugs, which had been used successfully on many occasions elsewhere by Chevron, but because no one on the immediate team had direct experience with them, they resorted to PCC-2 Article 3.2 for guidance. A third-party contractor was brought in to perform the freeze plug, and Chevron's engineer worked closely with the contractor and other owner representatives to ensure the procedure was comprehensive. The contractor's trained and experienced personnel offered valuable insight into the job.
Prior to execution of the freeze plug, all parties involved in the work gathered to assess the risks to health, safety, and the environment. PCC-2 addressed the issues and concerns regarding freeze plugs. Some of the risks discussed include:
Flow in pipe preventing plug formation--a threaded connection on the section of pipe to be isolated was dripping. Even a small amount of flow can prevent plug formation.
Determining positive isolation prior to beginning maintenance--if the broken valve was unbolted prior to achieving isolation, the flanges connecting the valve to the pipe would leak and, with an influx of warmer water, the plug would fail.
Downstream effects of ice plug--if the pipe was returned to service prior to allowing the ice plug to completely melt, the plug could flow downstream and severely damage equipment and piping.
Physical setup for the freeze plug began with ultrasonic thickness measurements taken in a 1-inch square grid for the full length of the area to be occupied by the jacket. The data revealed that the pipe was well above the required minimum thickness for hoop stress required by the ASME B31.3 Process Piping code, and close to original thickness in many places. The ultrasonic data gave confidence that the plug location could endure the mechanical loads likely to be applied while it was below the brittle transition temperature.
One of the key concerns on this job was minimizing the potential for impact loading the frozen pipe. Before initiating the freezing operation, all bolts connecting the valve to the pipe were changed out one by one--in case any of them had seized during the course of their 50-year life--with new lubricated bolts, and every other one was removed to minimize the amount of mechanical work necessary while the pipe was frozen. Scaffolding was erected to support a chain hoist to ensure gentle installation of the new valve, and the written plan for the job included carefully lifting out the piping above the broken valve and the valve itself using a crane, which would immediately remove those components from the job site.
To minimize the duration of the mechanical work, all materials and tools required for the work were brought to the site and organized prior to introducing nitrogen to the jacket.
In preparation for the unlikely event that the piping was fractured during the freeze operation, operators of all potentially affected plants were notified so they could review their emergency procedures.
The section of piping to be isolated contained a branch connection available for a pressure indicator and drain connection, so pressure could be monitored and bled off as the ice plug expanded into the trapped volume. This pressure rise is one of the indicators that a plug has fully formed.
The jacket was installed on a vertical pipe 16 inches from the upstream flange of the broken valve. Thermocouples above and below the jacket monitored the pipe wall temperature, which correlates with plug formation.
Upon completion of setup activities, the job was ready to begin. The nitrogen trailer was pressured up to 35 psig--enough to ensure that the freeze plug jacket remained full of liquid nitrogen, and not the warmer nitrogen vapors. The trailer was sized to contain three or more times the required volume for the work, to mitigate against unforeseen events.
The nitrogen was delivered to the jacket through a 3/4-inch diameter nitrogen hose and nitrogen gas vented from the jacket through two 1-inch diameter vent lines. The lines vented downwind of all work areas in the vicinity.
It took 18 minutes for the liquid nitrogen to reach the jacket, and just under two hours later, temperature and pressure readings indicated the plug was fully formed. This was verified using the bleed connection, and workers were given the green-light to drain the pipe and begin the valve replacement work. During this work, the contractor continued to monitor the temperature in the jacket to ensure the plug integrity was properly maintained.
In less than 20 minutes the upper pipe section and the broken valve were removed and lifted out of the way.
Once the old gasket-which had sealed the old valve to the pipe--had been successfully scraped off the flange, the new gasket and valve were carefully set in place using the chain hoist and gently bolted down. The space inside the pipe between the ice plug and the valve was filled with water to eliminate the possibility of the ice plug violently dislodging during the thaw and damaging the new valve. The valve was then closed, and the pipe was left to thaw overnight.
Upon completion of these activities, the planned work to replace the heat exchanger was able to proceed immediately.
The following day, after the pipe thawed, the new bolts on the valve were tightened to a final value and the freeze jacket was removed. The new valve was ready for permanent use.
Although none of the engineers on the team had worked with freeze plugs before, after establishing that in this case it was the safest alternative, they were able to implement one successfully on short notice, as part of a discovery job within a planned maintenance window. Employing a freeze plug proved more efficient both in terms of cost and schedule than the other repair alternatives, and was executed safely and with confidence due to the guidance provided in PCC-2.
It is clear that the ASME has once again delivered a standard that provides great business value, meeting a recognized need and enhancing the safe and reliable operation of existing process plants.
The authors would like to acknowledge management at the Chevron Richmond Refinery and at Chevron Energy Technology Co. for their support of this effort.
Jaan Yaagepera is technical team leader of the engineering analysis team at Chevron Energy Technology Co. and vice-chair of the ASME Post Construction Committee's Subcommittee for Repair and Testing. Nathan Tyson is a design engineer at Chevron Global Manufacturing. They are based in Richmond, Calif.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||FOCUS ON: PLANT ENGINEERING AND MAINTENANCE|
|Author:||Taagepera, Jaan; Tyson, Nathan|
|Date:||Aug 1, 2011|
|Previous Article:||Alternative measures: when instruments can't reach the pump, there's another way to go with the flow.|
|Next Article:||A pump war story: back to basics.|