This Month’s Puzzler
We are investigating whether we can increase our refinery capacity by about 15%. I’ve been told not to worry too much about the vent system. However, I have some concerns, especially because all the pressure safety valves (PSVs) are conventional and some of the scenarios I am looking at involve more mixed-phase flow (see Figure 1). A model of the vent shows that all PSVs have backpressure ratios less than 10%, although valve 2 is close at about 9%. Also, some of the steam heat tracing and the insulation on the vent to the knockout drums haven’t been inspected in some time. Should we be concerned?
Hunt For Choke Points
Absolutely, check the vent system! I have often run into incorrectly sized vent systems due to unanticipated choke points at pipe transitions and higher-than-expected pressure drops in piping runs. Two-phase flow has enormous impact on the pressure drop calculations and can create additional choke points as the liquid vaporizes. It is highly likely that you will find the resulting backpressure at the exit of the PSVs to be much higher than allowed for a conventional relief device and the resulting flow reduction and chattering could have catastrophic effects for your plant.
P. Hunter Vegas, project engineering manager
Wunderlich Malec, Winston-Salem, N.C.
Check The Model
You can’t ignore the vents. If PSV 2 is near 10% now, it is likely to be well above the 10% limit for a conventional relief valve. Putting in a balanced bellows valve will get you up to 35%; by the textbook, a balanced bellows is good up to 50% but, beyond 35%, you should consider a pilot-operated PSV. Before you can select options for upgrade you must create a working model based on the changes planned during the expansion.
Let’s start with physical properties. The physical properties, like boiling point and vapor pressure, control the composition of the vapor to a PSV — but these properties can change if the PSV is moved to a new location. Relocating a PSV to a different tray could make a big difference. Property changes could affect the entire vent system and even the flare.
As far as the whole system, you should check the capacity and performance of the knockout drums. A 15% increase in the throughput of a refinery or single unit could reduce the efficiency of separation. In addition, check the pump sizes and, if the pumps are on timers, perhaps re-calibrate the settings.
As for the heat tracing, ensure that all relief lines are traced up to the knockout drums. Trace all small gas lines if the stream could contain liquid. The larger vents after the drums should be okay. A model should confirm how much liquid is present and whether a vent should be heat-traced and insulated.
Remember tracing without proper bonding to the pipe and insulation is useless. So, inspect periodically and, just as important, test the heat load. With electric tracing, this is fairly simple: too much load means you’re heating air between the insulation and the pipe; if the tracing leg stays on constantly that’s another sign that you’re heating air. A dead tracing leg could indicate a burned-out heating element. With steam tracing, checking becomes complicated. Use an infrared gun to verify the traps aren’t cold and periodically strip off the insulation to inspect the bonding of the coil to the pipe.
Returning to the relief valves for a moment, consider the PSV inlet pipes. These are a potential bottleneck to expansion. Only 4% of set pressure is allowed under API 520. This likely will be a check-box on your EPA Permit to Install, which is required anytime your plant capacity is increased. It would be best to clear this roadblock early.
Another potential roadblock is the flare itself. A 15% increase in the flare requires more fuel gas and higher pressure drops through the flare nozzle, separator water seal and even the stack. Additional pressure drops could necessitate bumping up to balanced bellows relief valves. You may even need a larger flare nozzle. If that’s the case, you’ll have to re-run your flare stack calculation. According to Crowl and Louvar’s “Chemical Process Safety: Fundamentals with Applications,” 1st ed., p. 259: 2,000 BTU/hr-ft2 causes blisters in 20 seconds. Note that 350 BTU/hr-ft2 is typical for solar radiation; 450 BTU/hr-ft2 is the limit set by 24 CFR 51.203 (see Geyer’s “Handbook of Storage Tank Systems,” p. 230). If you’re beyond 450 BTU/hr-ft2, you’ll need a taller stack or to relocate one of the relief valves to another flare.
Dirk Willard, consultant
I’ve been tasked with increasing the capacity of a screening and pressing operation at a pulp operation that uses switchback grass straw. Our chemical process relies on atmospheric cooking. The final product is for food service, so it must meet basic cGMP cleanliness requirements.
The screening and pressing operation consists of a spray screen followed by a belt press. The lights go to a hydrocyclone for cleaning before passing to our wastewater plant. The heavies collected in front of the screen go to the press.
We bought used equipment and are suffering ongoing problems in maintenance and cleaning. The belt press arrived contaminated with cockroaches, which continue to survive despite our best efforts. In addition, the belt frequently tears or spins off the rollers; solids cake to the blade and slide ramp; and water drips off the press in the wrong places. We’ve cut back our pulp feed rate and diluted the pulp extensively with city water to improve the separation at the screen. The resultant higher flow rate through the screen has forced us to shut off the outer screen valves, so we operate at about 8 psig at 200 gal/min at the inlet to the spray nozzles. The belt press dewatering pump is limited to 100 gal/min, according to a hydraulic analysis. Our reject flow to the hydrocyclone is 120 gal/min, according to the operating foreman. The belt press is designed for a maximum speed of 10 ft/min but we’re operating at 12 ft/min.
What should I include in my design report?
Send us your comments, suggestions or solutions for this question by February 9, 2018. We’ll include as many of them as possible in the March 2018 issue and all on ChemicalProcessing.com. Send visuals — a sketch is fine. E-mail us at ProcessPuzzler@putman.net or mail to Process Puzzler, Chemical Processing, 1501 E. Woodfield Rd., Suite 400N, Schaumburg, IL 60173. Fax: (630) 467-1120. Please include your name, title, location and company affiliation in the response.
And, of course, if you have a process problem you’d like to pose to our readers, send it along and we’ll be pleased to consider it for publication.