"The test pressure seems too high." It was an honest criticism. We had based our line list on the design pressure of the vessel connected to the drain line and come up with 450 psi, with a testing pressure of 680 psi. However, previous hydrostatic tests of the nearby piping connected to that drain line only had been at 400 psi. Such are the challenges of developing a good line list. So, let's review what to consider when putting together a line list and some of the pitfalls.
The line list should define operating, average and maximum temperature and pressure, design temperature and pressure, pipe specification, special requirements (such as post-weld treatment for caustic and sulfur exposure), and hydrostatic test pressure. Most importantly, it must justify the values selected. A history is useful; to document design history, I make comments in a spreadsheet so they can be viewed later.
The line list should consist of two sections: mechanical and process. The mechanical section often only contains line numbers, connections and tie-points but sometimes includes the pipe specification; if the line list requires mating a 300-psi flange to a 150-psi one, it's mechanical's fault. The process portion covers corrosion, safety and pressure relief.
To develop the line list, you'll need: 1) an approved marked-up piping and instrumentation diagram (I never begin a line list without one!); 2) isometric drawings of normally open connecting pipe; 3) pressure safety valve (PSV) set pressures as well as design pressures for equipment and vessels — these may not agree; 4) field or distributed-control-system readings of average and maximum operating temperature and pressure, preferably over at least two to three years; 5) a process flow diagram (PFD) to define the stream composition and risk should a leak occur; 6) equipment curves, e.g., for pumps; 7) a unit operating manual — this is useful for understanding the unit operation, if the PFD isn't current, which is often the case; and, ideally, 8) a discussion with operations to confirm operating conditions — these frequently get lost in translation between project engineering and operation. If pressure readings are too remote, you may need a hydraulic analysis to estimate the true pressure in a pipe.
List the discrepancies among the collected data. A vessel design pressure may exceed the PSV set pressure. A pipe connected to one on the line list may not match the pressure desirable — usually, the highest pressure is chosen for the design pressure unless it conflicts with normally open connecting pipe. This is an important concept: base line list values on normal conditions, not worst case ones.
Design pressure usually is the vessel design pressure (especially if the pipe connects directly), the deadhead pressure of a pump or compressor, or the PSV set pressure. No hard and fast rules exist. If nothing else is available, refer to the plant standards or add, say, 50 psi to the maximum operating pressure.
Because the design temperature really represents fire conditions where metal softens and withstands less pressure, choose it with care. Normally, opt for the isometric temperatures, not the vessel design temperature. If possible, use the maximum operating temperature plus a safety factor of, say, 50°F. Tracing poses an additional problem. For steam tracing, select the normal saturation temperature of the steam supply. For electric tracing, I have seen 250°F used; constant wattage tracing can have an exposure temperature of 500°F.
Now, let's consider the pressure test. Hydrostatic testing is commonly chosen due to familiarity; some companies set a limit of 100 psig on pneumatic testing for "safety reasons," because of a lack of understanding of dynamics. (See: "Choose the Right Fluid for Your Pressure Test.")
It's good practice to test all utility pipes connected to the process at process pressures. Conduct the test at whichever is higher — the flange rating or the design pressure corrected for temperature — plus a 50% margin. For a design pressure of 300 psig, test pressure = 1.5×Ct×300, where Ct is the temperature correction factor for the pipe material. Here's a useful equation for the design-based test with carbon steel: test pressure = 1.5×(5.293×10-11×T4 – 8.710×10-8×T3 + 4.797×10-5×T2 – 8.656×10-3×T + 1)×design pressure, where T is in °F. The pressure mustn't exceed the flange limit set by the metallurgist — if this is a problem, change the pipe specification and probably the tie-point.
Lastly, most line lists require an inspection specification. For carbon steel, checking 10% of the welds is adequate; X-ray 100% of alloy welds.
DIRK WILLARD is a Chemical Processing Contributing Editor. You can e-mail him at firstname.lastname@example.org