MEC&F Expert Engineers : ACCIDENT INVESTIGATION REPORT- SHELL OFFSHORE DRILLING – DISCHARGE OF DRILLING MUD IN THE GULF OF MEXICO

Tuesday, December 2, 2014

ACCIDENT INVESTIGATION REPORT- SHELL OFFSHORE DRILLING – DISCHARGE OF DRILLING MUD IN THE GULF OF MEXICO



ACCIDENT INVESTIGATION REPORT- SHELL OFFSHORE DRILLING – DISCHARGE OF DRILLING MUD IN THE GULF OF MEXICO



INVESTIGATION FINDINGS
Incident occurred on 11 February 2014 at AC 857 operated by Shell. Estimated spill was 144 BBLS.  The JSA dated February 12, 2014, titled Transferring mud from reserve pit to active pits utilized for the mud transferring operation during the time of incident occurring on February 11, 2014, does not make any mention for a procedure assuring the proper alignment of all dump valves on the mud pits after they were cleaned and before they were refilled.
Not all persons responsible for opening and closing the mud pit valves had signed the JSA for the mud transferring operation.
The JSA was not properly completed or executed due to incorrect date when the document was generated.  Date Generated indicates 2-12-2014; the incident occurred 2- 11-2014 during the JSA applicable operation.
The JSA was not properly completed or executed due to all personnel performing the job tasks during the time of incident failed to review the proper Lock Out/ Tag Out safety steps to be taken in order to properly identify the #3 Dump Valve was still in the open position as also documented in the H&P LO/TO/TRY Record Sheet.
The Lessee failed to properly gather a sample for the Static Sheen Test at the correct sample location prior to releasing pit fluids to gulf waters during the pit cleaning operation.  Samples were gathered from the #4 Auxiliary Tank and the Trip Tank, but not the #3 Active Pit.  The Dump Valve on the #3 Active Pit was utilized to discharge the pit fluids into the gulf waters.  The #4 Auxiliary Tank was utilized in order to pre-stage the washing fluid for cleaning purposes.

LIST THE PROBABLE CAUSE(S) OF ACCIDENT:
Contractor did not follow JSA in place for opening and closing overboard discharge valves.

LIST THE CONTRIBUTING CAUSE(S) OF ACCIDENT:
Piping layout design did have an isolation valve which would have provided another barrier against accidental discharge of mud into the Gulf of Mexico. See Attachment for Diagram.
LIST THE ADDITIONAL INFORMATION:
Witness statements detailing the events of the Synthetic Base Mud spill are contradictory in the date of the incident given in the reports.  The Alleged Incident Report given by the Toolpusher states the incident occurred on February 11, 2014.  The Alleged Incident Report given by the Tank Hand states the incident occurred on February 12, 2014.  The Alleged Incident Report given by the Floorman states the incident occurred on February 12, 2014.  In the official Timeline given by the lessee, the date of the incident is February 11, 2014.  The Driller's Report states the date of the incident occurred on February 12, 2014.
Shell Offshore INC. has changed their operational procedures taken by H&P personnel for moving fluids between the mud tanks along with the planned installation of an additional valve downstream to eliminate the possibility of the issue happening again.
Also Shell Offshore INC. has added procedures for reporting incidents to BSEE to prevent the incident from happening again.


SPECIFY VIOLATIONS DIRECTLY OR INDIRECTLY CONTRIBUTING. NARRATIVE:
G-110
AT THE TIME OF THE INVESTIGATION, THE LESSEE FAILED TO PERFORM ALL OPERATIONS IN A SAFE AND WORKMANLIKE MANNER AND PROVIDE FOR THE PRESERVATION AND CONSERVATION OF PROPERTY AND THE ENVIRONMENT.
E-841
AT THE TIME OF THE INVESTIGATION THE OPERATOR FAILED TO ORALLY REPORT WITHOUT DELAY TO BSEE.
E-100
During the investigation it was discovered the operator failed to prevent the unauthorized discharge of pollutants into offshore waters.
_______________________________________________
MUD STORAGE TANKS AND RESERVE PIT


Mud is made up at the Rig location.  Most Rigs have several steel Mud Tanks. Muds and additives are mixed and held in the tanks. Some land Rigs also have a Reserved Pit dug out of the ground. Mud tanks are also called Mud Pits, a carry-over from the days of earthen pits. Mud Tank is the preferred term. The Rig does not necessarily use all the Mud Tanks at once although it does use several. The Active Tanks hold mud the pump actively circulates. Often mud components come to the Rig in sacs.
Normally the crew stores the sacs in a special component called the Mud House or Sac Room. The house or room keeps the sacs dry and allows them to be stored with care. These silo like tanks are bulk tanks or P tanks. They hold mud additive like Barite and Bentonite. Crew members use some additives in such large quantities that suppliers load them into the bulk tanks to save time and money. Bulk tanks usually have their own Hopper or Pneumatic system for transferring the additives to the mud system. The pump takes the mud out of the active mud tanks and circulates it through the system. Crew members connect the mud tanks with the piping and manifolds.
The number of active mud tanks depends on the amount of mud needed to keep the hole full. And the volume required on the surface to keep the mud in good condition for circulating.  The Sand Trap is the tank directly below the Shale Shaker.  The Shale Shaker removes most of the cuttings from the mud. But some are so small the Shaker cannot trap them, these fall into the Sand Trap. The Sand Trap is the first Settling Tank. Crew members have to clean it regularly to remove the built-up solids. Some small or old Rigs may have two or more Settling Tanks in the tank system. They allow solids in the mud to settle out but settling tanks do not do a very good job as compared with newer generation solids removal equipment so today most Rigs use a de-sander and de-silter.  Reserve Tanks are not a part of the active mud tank system. Instead the crew uses them to hold excess mud. Or they may use them to hold a different type of mud than the pump is currently circulating.  They may also store heavy mud for emergency well control operations.
A Slug Tank is a relatively small separate tank or it may be small separate part of a larger tank. The crew uses the Slug Tank to mix the slug. A slug is a small amount of a heavy mud that is pumped down the String. Crew members may also use a slug tank to mix a small amount of mud for a special purpose. For example the driller may need place or spot a small quantity of high viscosity mud also called a Pill at some point downhole. The Suction Tank is where the mud pump picks up mud ready to circulate downhole. Mud in the Suction Tank should be clean, free of solids and gas and be properly formulated or conditioned. Crew members use the Chemical Tank to mix special chemicals such as Caustic that they will put into the Active Mud Tank.
On some land Rigs, the Rig owner digs a large pit next to the Rig. This pit is called the Reserve Pit. The crew puts waste mud and the run-off from the work site in the Reserve Pit. In an emergency they can also use it as a place to put more mud than the tanks can hold. Often the Rig operator lines the Reserve Pit with a thick plastic sheet to prevent Liquids from leaching into the soil. And if the Rig is on a migratory bird fly way the operator covers it with netting to keep the water fowl from landing in it. Land Rigs drilling in environmentally sensitive areas will not have a Reserve Pit. Instead waste and run-off or hulled to an approved waste disposal area.

________________________________________________




MUD TANK CLEANING
 The History
 When oil based mud was first developed in the 1970s, the supply vessels of the day were not fitted with purpose built tanks, and initially the product was carried in converted fuel tanks. The result was that little of the product actually reaching the oil rig, the majority being removed by the tank cleaners. Tank cleaning took days and cost a fortune. Hence it was not long before the tanks were improved.
Often the improvement consisted of plating in the internal framing to present smooth sides and bottom, but even this process did not com pletely cure the problem. If the mud was left in the ship for more than a couple of days the baryte began to drift out of the oil and settle on the bottom where it would remain until manually removed.
The oil industry is never at a loss for a knee-jerk response, known by some as "opinioneering", and some ships were fitted with jetting nozzles not unlike those fitted to the mud pits of oil rigs. They are known on the rigs as "mud guns".
If mud guns were successful on oil rigs the success certainly was not repeated on the support vessels. The mud was of course drawn from the suctions in the bottom of the tanks, using the discharge pumps or possibly specially installed circulating pumps, and then re-injected into the bottom of the tanks through nozzles.
The baryte continued to drop out of suspension so that the liquid at the bottom of the tanks became thicker and thicker. This was circulated by the pumps until they finally gave up or the nozzles blocked up. Even worse, the suctions from the tanks frequently blocked so that ships returned to port with a full cargo and were faced with the problem of how to discharge it. In the 1970s contractors in Lerwick were often employed to transfer mud from one tank to another through the deck manholes using portable pumps.  So a further development understandably was the multi-level suction. Many ships were fitted with a lower and an upper suction and some even had three levels. The theory was that even if you had to leave a couple of  feet of the product in there, it was nothing com pared with a full tank.
Of course tank cleaning became more frequent on the theory that no sediment meant no blockages.
Eventually the circulating systems became more practical, taking the liquid from the bottom of the tank and returning it to the top although designers remained obsessed with small diameter nozzles. The nozzles were usually removed by the ship's staff as soon as the vessel entered services on the basis that the possibility of blockages would be reduced.
Improvements in formulation of the product and the provision of agitators, together with the acceptance that frequent tank cleaning must take place more or less removed the requirement for the upper suction. However the main suctions were generally placed well above the tank bottoms and were of large diameter, both factors which reduced the possibility of a com plete discharge.
Alternatives were tried, and hopper suctions usually worked though they were com plex to install. Suctions in hatboxes were generally less successful due to the tendency for the mud residues to drop into the hatbox after discharge had been com pleted, and then solidifying.
It appears to be purely an accident of design that many tanks were, and probably still are, filled through the suctions and this ensures that they will be kept clear.
Many early systems used conventional centrifugal pumps, and indeed some are still so fitted where economy is considered to be of greater importance than efficiency. Centrifugal pumps suffer from a tendency to lose suction if the pump is higher than the suction or if the lines to them are at any point higher than the pump. These inherent deficiencies may well result in high levels of residues remaining in the tanks.
The increasing use of constant displacement pumps has considerably reduced this problem, and hydraulically powered pumps which offer variable pumping rates may well provide the means of reducing residues by avoiding cavitation at the suctions.
Recent Developments
During the 1990s the major Norwegian designers addressed the whole problem of the carriage of mud, and almost universally adopted cylindrical tanks with hopper bottoms. The redesign of the tank on its own has vastly improved the ability of the vessels to discharge com plete cargoes.
These cylindrical tanks are also now often fitted with a single tank cleaning machine somewhere in the upper part and an agitator in the lower part.
The agitator replaces the circulating system.
It is certain that the tank designs and coating has reduced the level of residues remaining in the tanks and that the tanks may now be cleaned to what ever standard is required using tank cleaning machines. Here there is a difference in view as to the manner in which the tank cleaning should take place between the British machine suppliers and the Scandinavian suppliers.
The British, represented by Marex and Dasic believe that the tanks should be cleaned with the product during discharge, reducing and maintaining the residues to a minimal level, and that very occasionally a water wash should be undertaken. The Scandinavians represented by Gunclean Toftejorg and Scanjet think that a water wash should be carried out every time.
The Brits are of the view that their continental rivals have taken this stance because of limitations in the ability of their machine to circulate mud.  The Dasic manufactured machine can successfully circulate mud with its attendant impurities which gives it a number of advantages. On the other hand the continental manufacturers are always looking to reduce throughput in order to limit the amount of water which is required to be heated and the volume of residue produced. In the tank cleaning business this aim can be counter productive. If too little volume is applied to the surface to be cleaned then more cleaning cycles have to be carried out. Result - same amount of contaminated water.
 The British suppliers also claim that their process of cleaning the tanks with the product is more beneficial for retro-fitting to vessels constructed before the latest designs ensured easy-to-clean tanks.
Even where there is substantial internal framing, the installation of one or two machines can reduce residues to a minimal level when the tanks are cleaned with the product and no heaters or chemical injectors are required. The result will be a substantial reduction in or the removal altogether of tank cleaning costs, and most importantly today, a vast reduction in the volumes of contaminated water which have to be disposed of somewhere.
Tank Cleaning on Oil Rigs
Only a few oil rigs and platforms have been fitted with tank cleaning systems. Tofjorg have supplied some platforms in the Norwegian sector and Marex has supplied a full set of machines to the semi-submersible Stena Don, and a number of individual machines to other rig owners.
Drilling fluids are as difficult to deal with on board rigs as they are on support vessels. The solids still settle out even in pits which have been provided with mud guns and agitators. This is probably because the mud guns were never going to work anyway and the agitators are mainly too small and of unscientific design.
Mariners who visit oil rigs are usually surprised that these floating objects are provided with mud pits at all. Virtually none of the pit systems are totally enclosed. The best are decked in but still have holes through which pipes may pass. Some have manhole coamings but no hatches. The worst are decked out only with gratings. They are therefore not tanks in any sense of the word except in the way that an ordinary domestic bath is a tank.
In the past this was not much of a problem. When the pits were empty some-one would open what is known as a "Dump Valve" or to continue with the bath tub analogy – pull out the plug, and then some minions would climb into the pit and wash it out with water. The result would then drain out through the dump valve into the sea.
Traditionally the only place which held mud was the mud pit area and the total quantity held was about 2000 barrels in oilfield figures or about 300 tonnes.
Now with the great increase in the expected water depths at which mobile rigs will operate, and the extra depth in the substrata to which they are expected to drill, it has be com e impractical to hold all the mud in the pits and so older rigs are modified so that some can be held in tanks in the legs or pontoons. New rigs are provided with purpose built tanks. Of course offshore oil industry, remaining oblivious to the technology available in the marine industry, have also fitted these new tanks with agitators and mud guns and the lack of effectiveness is only apparent when the rigs return to port for maintenance and the tanks are found to be half full of solids. The solids can cost a five figure sum to remove.
In addition to the unwanted costs to the rig owners, their clients are now be com ing more interested in the environmental effects of the drilling process. Today drill cuttings are being transported back to the shore for processing solely because they are contaminated with drilling fluid. However the residues within the mud pits and the mud tanks remain a problem.
Even knowing the extent of the problem does not seem to provide the industry experts with a solution. The manifestation of their difficulty is the depth of sediment in the bottom of the pit and so the approach up to today has been to clean out the pits with water and then to send the resulting contaminated fluid ashore in the tanks of supply vessels. This process transfers both the sediment and the problem to the support vessel and still necessitates the processing of the contaminated water. When the water is discharged, all the sediment from the pits is left in the bottom of the supply vessel tanks which then have to be cleaned. This produces even more contaminated water which has to be sent for processing. It is a total lose, lose, situation.
In an effort to reduce the levels of residues some Operators in the North Sea have employed contractors who use tank cleaning machines, detergent and water, and then suck out the fluid for transport to the shore in one way or another. This technique appears to have cut down the level of residues, although claims of a maximum of 60 bbls (For those who deal in cubic metres there are 6.3 bbls to a cubic metre) per tank seem to be a trifle optimistic. Retired mariners may remember that Dasic and other tank cleaning machine manufacturers actually provided experts who carried out exactly the same task on oil tankers in the 1950s.
Attempts to install fixed nozzle mud cleaning machines in the mud pits of rigs have been frustrated by the tendency for the mud spray all over the pit room through holes in the top of the tank, and programmable single nozzle machines so far have been unable to cycle the product. However Marex are now marketing an ingeniously modified fixed nozzle machine which only jets downwards, and which can therefore be installed in open mud pits. Hence the operators of rigs can now enjoy the same advantage as the operators of ships. They can clean the pits with the product, removing the sediment without producing any residues for disposal.
Therefore – by cleaning the pits with the product every time they are emptied there will be a minimum of sediment left, and by making a special effort during the return of the mud to the supply ship there should be no sediment left. Similarly if the supply vessels clean the tanks with the product at every discharge the residues will also be minimised. Production of contaminated water is negligible. Operators and ship-owners can conform to their environmental policies and avoid needless expenditure. This is surely a win, win situation.      

Metropolitan Engineering, Consulting & Forensics (MECF)
Providing Competent, Expert and Objective Investigative Engineering and Consulting Services
P.O. Box 520
Tenafly, NJ 07670-0520
Tel.: (973) 897-8162
Fax: (973) 810-0440
We are happy to announce the launch of our twitter account. Please make sure to follow us at @MetropForensics or @metroforensics

Metropolitan appreciates your business.
Feel free to recommend our services to your friends and colleagues.