Safety Alerts 17

Refinery

• HDT: Vibrations by a recip compressor + piping expansion/contraction during hot and cold cycling, caused 1" globe valve plug to fall off. As the valve was not fully closed, H2 leaked but quickly dispersed

• HDT: Severe failure of Naphtha HDT HX shell. 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

• DHT: Recip type Recycle Gas Compressor. Each loading/ unloading valve had Instrument Air tubing for valve actuation and a gland leak collection tubing to flare. Identical tubing. During maintenance one of the IA and flare tubing got interchanged. On starting the compressor, gland leak stream with 95% H2 and 0.05% H2S reached the nearby Analyser House via the IA tubing. Explosion. Different types of tubes, colour coding and different directions of rotation for IA and flare tubing 

• HDS: Restart. HX additional bolt tightening being done due to thermal expansion. Leak and explosion. Channel cover and parts hurled over 200m

• Fuel Oil HDS: During shutdown, reactor was neutralized and washed with a polythionic acid solution. It accumulated at the drain nozzle in reactor inlet / furnace outlet piping. On startup, hot HDS gas from furnace evaporated the water concentrating the chlorine. Industrial water was used to dilute polythionic acid had high chlorides that was further concentrated. SCC (Stress Corrosion Cracking). H2 leak and local fire. The dilution water was changed to pure water to lower the chlorine level

• Fuel oil HDS: Leak from a PG piping caused jet fire. Minimize additional tapping. Add depressurisation

• Benzene: During start-up, hydrogen gas leaked from HX flange and fire. Insufficient tightening / hot-bolting

• Catalyst Dewaxing: During hot hydrogen regeneration of catalyst, an explosion occurred. Reactor outlet pipe failed at an elbow. Possible cause sulphidation corrosion from catalyst activation years ago. Wall thinning from corrosion + turbulent flow at the elbow

• HF Alkylation: Piping elbow corroded. Had a high nickel and copper content. Carbon steel with a higher % of nickel and copper corrodes at a faster rate than carbon steel with a lower % in HF service. VCE. No SDV installed to remotely stop the leak

• Isomerisation: Faulty level instruments. Column overfilled and spewed out via vent HC liquid and gas. Explosion/Fire

• MEA column ruptured at a weld joint made 10 years back. No PWHT. C3 at 200 psig propelled the vessel. Resultant fire and BLEVE burnt other units

• Sulphur Plant: Accelerated corrosion of Claus Sulphur equipment when opened for maintenance - due to humid air. SOP to address 100% drainage of contents + plugged piping + molten sulphur issues while operations hands over to maintenance contractors (LinkedIn)

Petrochemicals

• Ethylene is unstable. Decomposes to C + H2 + Heat. Reaction front along tops of pipes/ equipment. Tell-tale: Scorched paint, top of piping. Heat release can cause rupture, release and fire. Past incidents at: Ethylene Dryers, Mol Sieves, as regen gas, Compression with O2 or N2 (diatomic gases heat up more in compression). Avoid initiators like O2, Metal powder/dust, Peroxides and Mol Sieve. Purge air/N2 from vessels slowly with C2=. Monitor temperature and cool sieve beds before introducing C2=

• Ethylene leaked into steam line and reached space heater below control room and accumulated in the condensate drain inside the building. Building was destroyed

• Ethylene: Regeneration line. Temperature range -17°C to 220°C.  Atmospheric moisture condensed and evaporated cyclically. Led to CUI and leak of 40% C2=. Foot traffic or stepping on insulated line damages external jacket

• Ethylene: Start-up. One of the EVAL reactors had a lower pressure than the other 3. Circulation of refrigerant to the HX inside the reactor caused C2= condensation and low temperature. Steam was introduced in reactor vessel jacket to warm up. C2= vaporised increasing pressure to PAH. Operator manually routed gas to flare; not enough and the PSV popped - not routed to flare but vented to atmosphere. Ignited by a welding machine. Fire

• C2= Plant Refrigeration Compressors: Low mass rotating assembly. Discharge check valve failure resulted in high reverse rotation speeds. Multiple incidents. A few resulted in gas release and fire

• C2= Plant: Exothermic runaway reactions are known in Acetylene converters, MAPD+H2, C4 Hydrogenation, PGH, Methanators, MTBE reactors, Adsorption beds etc. Number of incidents

• C2= Plant: Pneumatically assisted NRV in Cracked Gas Compressor line failed. VCE. Similar, C3 Refrigeration system compressor discharge in another C2= Plant

• C2= Plant: Fire at the base of Deethanizer column, due to flange leak at reboiler or relief header. Leaking C3= ignited by steam piping. Fire engulfed C2= and C3= columns. A vertical C3 storage tank exploded and 2 C3= tanks toppled. 5 C2=/C3= tanks collapsed and exploded. C2= column released its inventory

• C2= Plant. Quench column PSV opened on Cooling Water failure. PSV vibrations led to failure of bolted flanges. Local release and explosion, shutting steam to cracker furnace tubes. Loss of steam resulted in cracker furnace tubes rupture and release of quench oil. Quench oil pool fire below a cracker damaged 4 cracker furnaces

• C2= Plant. Lightning strike caused black screen in Cracker Control Room. Poor grounding of cable shield wires. Compressor controls damaged by over-voltage. PSV release. PSV chattering cracked 2 compressor vent lines. Plant ran on last Set Points of controllers

• C2= Plant. Cold Box Failure. Hole in an elbow downstream of JT valve - localized erosion due to impingement + cavitation damage

• Ethylene Polymerization: HDPE made by polymerizing C2= dissolved in isobutane. Settling polymer particles dropped into legs at the bottom of reactor pipes. They were periodically removed by closing an upper DEMCO (butterfly) valve and a bottom valve in a leg. Frequently the legs used to get clogged. After closing the DEMCO valve, the leg was removed and cleaned out. Pneumatic connections for opening and closing the valve were identical and were improperly reversed when last re-connected. As a result, when the switch in the control room was in closed position, a leg was removed without closing the DEMCO valve. Flammable gas release and explosion Interlock or DBB? Different hose connections?