Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

The past --LNG - BOILER PLANT EXPLOSION -2004 - SKIKDA 1

Status
Not open for further replies.

MJCronin

Mechanical
Apr 9, 2001
5,087
An oldie, but a goodie .....

SKIKDA, ALGERIA -- LNG PLANT BOILER EXPLOSION - JANUARY 19, 2004

___________________________________________________________________________________________________________

What Went Wrong In Skikda LNG Explosion (27 employees died, 56 injured and cost nearly $1Bilion

The Skikda LNG complex comprises 6 LNG liquefaction trains. Units 10, 20, 30 and 40 are located parallel to each other on the west side of the LNG storage tank area. Units 5 and 6 are located remotely on the east side of the LNG storage tanks. The administration, maintenance and security buildings are located adjacent to the most westerly unit ( Unit 40 ). Units 10, 20 and 30 ( utilising mixed refrigerant technology ) were brought online in 1971/1973. Units 40, 5 and 6 ( utilising single mixed refrigerant technology ) were brought online in 1981.

On 19th Jan 2004 with Unit 40 operating normally, a steam boiler providing high-pressure steam for the Unit 40 refrigeration compressor turbine driver exploded. The boiler firebox casing was breached, triggering a fireball and a second, much larger, vapour cloud explosion (VCE ) which spread outward, completely destroying Units 40, 30 and 20 ( 43 % of the site's production capacity ). It also destroyed the administration, maintenance and security buildings, trapping workers under the debris. Damage to Units 10, 5 and 6 and the LNG storage tanks was minimal. However, surrounding facilities including a nearby power plant, an LNG loading berth at Skikda harbour and numerous homes and other buildings in the community were also damaged. The neighbouring refinery was shut down as a precaution. Unit 6 of the LNG Complex was restarted in May 2004. Units 5 and 10 were restarted in September 2004. Units 20, 30 and 40 were eventually rebuilt.

The possible cause is the release of mixed refrigerant vapours and LNG (from a cold box heat exchanger leak ingested by the air intake of the forced draft combustion air fan at Unit 40 ) which were steam boilers, creating an explosive mixture in the boiler firebox.

Accident Analysis:

1 ) The Unit 40 steam boiler was located very close to the LNG liquefaction and separation sections of the Unit 40 process plant ( newer LNG plant designs use gas turbines to drive the refrigerant compressor these are more efficient, more robust and eliminate the need for a steam boiler ).
2 ) The loss of primary containment at the cold box released hydrocarbon vapour into a congested space between Unit 40, the control room and the boiler ( exacerbating the impact of the VCE ).
3 ) Still ambient conditions ( no wind to disperse leaking vapours ).

Root causes are:

1 ) Poor plant layout ( proximity of neighbouring LNG liquefaction trains and occupied buildings )
2 ) Inadequate inspection and maintenance (cold box heat exchanger ).

Lesson Learned :
1 ) Escalation impact studies should be carried out to determine the best plant layout and equipment spacing to minimise the risk of a major accident.
2 ) Land use planning regulations specifying minimum separation distances between high hazard facilities and residential buildings should be enforced.


Ref. Ichem free LL library.

#SkikdaLNG #lessonslearned

No alternative text description for this image

MJCronin
Sr. Process Engineer
 
Replies continue below

Recommended for you

Is this that one that showed the idyllic life at the outskirts of an enormous facility and then a huge vapor cloud instantly enveloping the moving trucks, running people, and equipment, that then ignited?

Keith Cress
kcress -
 
I do not know the ultimate cause of the failure in the cold box, but I recall a failure of a new heat exchanger in the cold box of a LNG plant in Australia at the same time was ascribed to corrosion of the aluminum HX due to mercury contamination of the process fluid.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick
 
davefitz, Latex, itsmoked, et al ....

I have no doubt that the latest crop of STEM or STEAM engineers entering the LNG field will be able to use the lessons of the past dealing with the complex and subtle issues of plant design.

The new education system guarantees that these brand-new engineers and designers will guard and protect human life from billon dollar disasters !!!







MJCronin
Sr. Process Engineer
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor