Topics
Heat Exchangers
To detect leak, after removing head, 8 bar gas was applied on shell side. Tube bundle moved out killing mechanics. Liquid instead of gas could have avoided fatalities
Similar, to push out a stuck tube bundle, steam was supplied to shell side and bundle blew out. WWW
Tube plug ejected like a bullet under N2 gas leak test pressure. Hydrotest is better. BP
Tubes cleaning: With water at 1,000 bar. Foot operated Deadman switch was disabled to clean lower tubes. Cleaner tripped and the nozzle came out of the bundle and killed him. Use automated cleaning
Similar, the cleaner stumbled, when the jet was suddenly shut-off and restarted. Injury. BP
During cleaning, the threaded hose connection came apart and lashed. Injury. BP
Similar, 2 of the 3 jetting nozzles were blocked. Bursting disc opened. Injury. BP
Cleaning: Hot butane, a reactant in polymerization reactor, was circulated. Catalyst traces on the walls caused a runaway reaction rupturing a plastic blowdown drum. Pure butane, instead of a mixture with other components, is very reactive
Cleaning: Shell side isolated to replace a rupture disk. Forgot to open the isolation valves. Tube side was cleaned with steam. Shell side overheated and burst
During start-up, a bleed valve on top of crude unit HX remained locked open. Hot oil at 295°C (auto ignition temperature 250°C) flowed out. Ignition. CSB
Wrong/ lower torque used to tighten shell and channel bolts. After 43 thermal cycles the bolts relaxed. H2 release and fire. CSB
Welding: Trapped fluids in plugged tubes may explode or emit obnoxious fumes when heated. Titanium tubes may melt; molten titanium readily burns in air. WWW
CDU/ VDU: Temporary feed loss to VDU column. VDU feed/ overhead HX thermally shocked. Leaked long residue ignited. BP
VDU: Internal leak within Crude Oil/ HVGO HX. VDU Column RV lifted. BP
Low Temperature: During a trip, C3= on the shell side hilled down. Pressure fell. Water in tubes froze and tubes broke. Blew out a section of 16” water line. Ignited by a furnace 40m away. WWW
C3=: HX reached (-) 40°C after depressurizing. Was restarted (pressurized) and it ruptured. Brittle failure. Fire and explosion. Train operators on auto-refrigeration and MDMT issues
C3=: Reboiler isolated from RV on column. On introduction of hot quench water to tube side, thermal expansion of trapped C3/C3= in shell. Release and fire
Hydrotest: Tube side hydrotested. Without draining the water, steam was introduced on the shell side. Thermal expansion ruptured the tube side. BP
BAHX (Brazed Aluminum Heat Exchangers). Accumulated thermal fatigue caused rupture. Don’t go by ‘leak before failure’ but by service life to replace HX
In another Air Separation Unit, burning of Aluminum heat exchanger elements in the presence of liquid oxygen caused explosive rupture
Mercury liquid metal embrittlement. Aluminum cold box nozzle failed leading to explosion
Water Hammer: Condensate in steam line damaged the impingement plate and broke HX tubes
Condensate build-up can reduce heat transfer. Supply valve may have to fully to expel the condensate and then throttled allowing condensate build up. Vicious cycle. Water hammer and corrosion. WWW
Erosion: Thinning/ hole observed in Cold Box piping bend, downstream of a liquid Methane LV. Flashing and cavitation
Closed blowdown valve resulted over-pressurization when dry methane (defrosting gas) was introduced into the shell side of a cryogenic HX during start-up of an LNG plant. Explosion. BP
Similar, closed last valve of a HX train resulted in high pressure on the last LP HX. Burst and fire. BP
HX with NH3 in shell taken out on maintenance. RV inlet closed. HX steamed out. Liquid NH3 vaporized and HX burst
HTHA: Severe failure of naphtha HDT shell. 38-year-old unit designed when HTHA was not understood. HTHA can occur in the so-called safe region of Nelson curve in API 941. Operating conditions can get severe than design. Avoid CS in HTHA service and go for Cr-Mo steel
Graphite HX cracked, due to higher than allowed torque while hot-torquing bolts during a restart. HCl gas released. Torque for HX to piping was lower than adjacent piping to piping bolts. CSB
Fired Equipment
Major Causes: Flame lick, poor draft control, not regularly checking tube + support for corrosion/ thinning, failure to purge, condensate in fuel gas and water in fuel oil. Establish draft first during a cold light off
HCU: Feed heater tube ruptured due to flame lick/ hot spot. 1.4m longitudinal split. Hot spots were wrongly assumed as burning scale. Early warning via smoky plume in stack tip + reducing O2 in stack, due to additional fuel from leak. Failed to recognize that scales in SS347H tubes form at higher temperature (700-815°C) c.f. ferrous tubes in other heaters. Optical pyrometer scan not used
HDT Feed Heater: Naphtha + H2 feed. Leak in convection tubes ignited. 0.6m long flame. Naphtha feed and fuel were cut. H2 flow maintained to cool the tubes. Invisible H2 flame worsened the damage until molten metal drips started. H2 flow was stopped and steam injected. 32 tubes damaged
CDU: Process gas leak into firebox from tube header plugs, during startup. Explosion. BP
CDU: Coil steamed and N2 purged. Checks done on tube header plugs for leaks. One plug was dislodged and released N2 with HC. Ignited. BP
CDU: Thinned heater tubes failed on water hammer while steaming through to the column. BP
CDU: Major process upset resulted in liquid into Fuel Gas system and firebox explosion. Check regularly Fuel Gas KODs. BP
CDU: Fired with FCCU Regenerator off-gas. FCCU upset resulted in unburnt hydrocarbons, excessive tube metal temperature, 810°C, and failure. BP
CDU: Hygroscopic combustion deposits entered header boxes during operation. Severe corrosion of the headers when deposits hydrated during shutdown. Failure when the unit was recommissioned. BP
CDU: Tube plug ejected from the header box when coil was under pressure during maintenance. BP
CDU: FD fan was used to purge a firebox, but its louver was stuck closed and PAL was disconnected. Explosion. BP
CDU/ VDU: Fired reboiler faulty pass flow controller slowly moved to closed position while showing constant output. On excessive skin temperature and red hot tube, heater was shutdown. But coking had reduced flow by 75%. Pass outlet TI may show a wrong value, influenced by conduction from combined flow from other passes. Maintain flow variations within 10% and outlet temperature variation within 3-5°C. BP
VDU: Firebox explosion with high waste gas flow coincident with low fuel gas flow during startup. BP
VDU: Liquid HC entered with waste gas fuel when fuel gas KOD bypass was open. External fire. BP
VDU: On power failure, naphtha feed to convection section stopped while radiant section feed flow was maintained by steam turbine driven pumps. Tube failure and fire. BP
HDS: Automatically tripped off 5 times due to false alarms on another part of the unit. Re-started four times. On the fifth attempt exploded. Poor flame monitoring of main and pilot burners. Signal bypass showed pilot burners always in a “good” state and BMS logic allowed the gas valves to open without establishing ignition - unburnt gas accumulation in firebox. BP
C2= Cracker: On low steam drum level, Boiler Feed Water (BFW) pump and furnaces tripped. Tubes ruptured in 2 furnaces. Firebox fire. One firebox had pool fire below as burning quench oil ran out of firebox. Additional fires as gasoline contaminated dilution steam was sent to furnace coils.
(1) Auto-start standby BFW pump + alternative sources for dilution steam (2) Better pyrolysis gasoline-water interface detection in quench tower to avoid sending gasoline contaminated water to steam drum (3) On loss of dilution steam, quench tower interface level fell. Alternative make-up water to quench tower (4) Quicker trip of quench oil valves to furnaces (5) Better and faster response on furnace tube rupture (6) Paved and sloped grade below heaters avoid leak liquid accumulation