What would it take to avoid sinking the platform? 2

[docellen]

Seems like just a little extra steel between the underwater pontoons would do it. Then, of course, we would need a tugboat nearby to keep the rig from drifting off and snapping the riser.

 

[docellen]

JR, excellent answer. It raises another question, however. Why would either the deck box or columns not simply drain whatever water came in? Are the pontoons close enough to the water surface that they could be ruptured by an explosion?

I think your guess about the ballast valves not being a fail-safe design is the most likely reason.

We need an emergency connector, accessible from a boat even with the rig on fire, connected to ballast valves, underwater thrusters, and whatever else is needed to maintain minimum control in an emergency.

 

[JRMacGregor]

Normally the deck box will have scuppers etc to drain any water over the side. But as I said, this integrity was probably breached by the explosions, allowing liquid into spaces normally dry. Allowing 10,000 tonnes of liquid into the deck box would be more dangerous for stability than allowing 10,000t into the pontoons. It would flow to one side, causing a heel angle. Then when the heel angle gets bit enough, the sea starts to flood in through external openings, and that is the end of the story.

The pontoons are not so close to the water surface, and if they had been breached by the original explosions, then the listing and sinking would have been quicker than seemed to be the case. In addition, the pontoons consist mainly of ballast tanks, which are mostly already full when at drilling draught.

Breach of a column at/below the waterline and its flooding would be more easy to imagine, and this would lead to a heel angle problem. The columns are what gives the vessel its "stability" on the surface, whereas the pontoons main purpose is to give weight carrying apacity (Archimedes principle).

As for the ballast valves, this was a modern rig and I would expect the ship side valves to be fail-to-closed rather than fail-to-open. This is certainly a requirement in the UK since the 1980s. The loss of OCEAN RANGER offshore Canada was one reason for this requirement.

 

[JRMacGregor]

Docellen

Re your question/suggestion about an emergency connection which would allow thrusters etc to be maintained following such an event..........

You are certainly right in wishing for some means of keeping the rig in position after such an event. But the power generation, distribution and control systems on these DP rigs are so complex, attempting to do it via the existing thrusters is impossible. To take one simple example the day tanks (service tanks) in the fuel supply system would need topping up from the storage tanks within a day or so.

It would be easier to ensure there are some deployable (but how ?) towing bridles that would allow support tugs to hook up and hold the rig without men having to go onboard.

While on this subject, I read on this forum some folk making remarks about the fact that the generators were breathing methane and should have had shut down flaps to prevent this.

Since this was a large modern DP rig, dependent on engine power for positioning (a.k.a avoidance of riser rupture) I would expect the following to have been present in her design;

1) automatically actuated fire flaps on the engine room air intakes, linked to the fire and gas, emergency shutdown matrix, to avoid gas ingress to the ignition sources in the engine room in case of gas detection

2) engine room air intake DIRECT DUCTED from the outside into the turbo intake on the generator - specifically to ensure that the engines WOULD continue to run and the rig not drift off (risking the riser) in the event of a gas leak (when engine room intakes closed, hopefully)

And finally, whatever the failings of BP on this rig or in their former ARCO/AMOCO refineries ashore, the responsibility for properly maintaining the rig's explosion proof equipment and the ESD fuctions on air intakes etc was not theirs (as some seem to think).

 

[JRMacGregor]

Sorry Docellen - I realised that I did not properly answer your question about why the deck box would not drain any water/oil that came INTO it.

The scuppers would drain off oil/water that came ONTO it.

The deck box would not let water/oil out that came IN because it was designed to be watertight (from the outside). The buoyancy in an intact deck box is one of the things that stops the rig heeling too far in a damage incident (like loss of a column).

The underside of the deck box would have been totally watertight. When you get to the top of the deckbox you cannot avoid all sorts of penetrations (doors, vents etc) that may or not be watertight or resistant against flooding when the rig reaches an extreme angle and the water comes over them.

 

[CH5OH]

JRMacGregor,
consider the amount of power a platform uses
consider the volume of air required to power the diesels
shut down flaps with gasdetectors on the engine room air intake will pull the engine room in vacuum and black out the rig, so no more thrusters, just the emergency fire pump
running on the emergency diesel, maybe
consider a blowout of methane,the rig completely surrendered by a cloud of the right mixture of methane and air.by the time the gas detetors on the air intake work, the engine room is filled with methane/air
direct ductig of turbos is not done, you would pull to much salt in your engine, in case of gas leak, you still overspeed them on the methane/air mixture
remember the BOP was the last line of defence, a triple redundant device (on paper)
1 annular device, shred to pieces while performing normal operations
2&3 failed to operate (did they?)
the maximum pressure they have to withstand is the well pressure plus a safety margin (really?)
static force=well pressure (this you could simulate and therefore test
dynamic force=bringing a 15km long liquid flow to a sudden stop.its like trying to stop a runaway train by placing a truck in front of it.

 

[JRMacGregor]

No - that is exactly my point.

The engine combustion "air" has a separate duct from the outside. If the engine ROOM fire flaps are closed the engine can continue to run.

Direct ducting of engines IS done - I have built DP drilling rigs.

 

[docellen]

JR, your knowledge and experience is greatly appreciated. If I understand correctly, a little extra steel between the pontoons won't help. That part of the structure is already strong enough. The problem is in the columns (which must be buoyant to keep the platform level), and the deck box (which must be buoyant to avoid rollover if the columns are damaged).

At this point, I can only state a general requirement, not a specific solution. The rig must not sink after a blowout and explosion. Specific solutions might include having separate compartments in the deck box, so they don't all flood when there is limited damage, or maybe having inflatable pontoons in a strong box under the deck.

Another solution might be to let the rig sink, but have separate flotation for the top of the riser. Maybe this could be a big collar with rings that could be hooked to winches on a nearby ship.

There must be a simple solution that doesn't require readiness and planning that may not be there 20 years from now.

 

[waross]

The emergency shutdown flaps in a diesel are installed at the air intake intake of the engine. The original application that I know of was in the two cycle diesel engines built by General Motors, Detroit Diesel Division. Probably the best known of these was the 6-71 model Detroit Diesel. When these engines were first built back in the 30's oil seals were made of leather. Failures were common. A failure in a shaft seal in the blower would allow engine lube oil to enter the blower and intake manifold. The engine would start to run away and could fail catastrophically due to high speed. Today in Canada shut down flaps are common in diesel vehicles used in petroleum facilities from rigs to refineries and upgraders. This is not new technology. Intake flaps on diesels have been around for over 70 years.
Regarding dampers in the ducting to the engine room. Maybe not a good idea. There have been failures in the dampers of HVAC mixing systems that closed both the return air and makeup air ducts with the fans running with the result that the plenums collapsed due to reduced internal pressure.
And a friend once told of being in a standby generator room for a startup where the air intake dampers to the room had been wrongly installed. Common practice at that time was to use a spring return damper motor to hold the air intake flaps clopsed. When the ATS sent a start signal to the gen set, it also interupted the power to the damper motor and the spring return would open the air intake damper. In this instance the damper motor was wired normally closed with power to open. My friend was with a group in the generator room when the generator was started for the first time. He described dust and paper flying around in the room and a futile effort to open the door against the partial vacuum. Eventually the engine came up to speed, the voltage stabilized, the ATS transferred and the damper motor got power and opened the air intake flaps. It generally takes less than half a minute for a good stabndby set to come on line, but I imagine that that may seem like a very long time if you are in an area of reduced pressure being pulled by the engine.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[JRMacGregor]

Hi Docellen

The semisub rigs already have to meet international (and some national) damaged stability codes demanding they withstand serious flooding and loss of buoyancy. Modern semisub rigs are subdivided internally and designed to withstand loss of any 2 adjacent compartments in the columns or pontoons.

It should be noted that this was a large modern DP rig. Many older semisubs continue to operate (albeit rarely beyond 3000ft water depth) and most of them have no buoyant deckbox, just an array of girders and a plated deck. These are mainly anchored, hence no need for such a huge power plant (or direct ducted engine air intake) as found on the DP rigs like Deepwater Horizon.

As for the riser, in very deep water it requires up to 2000 tonnes of uplift at the rig to keep the riser properly tensioned. That rig uplift is what must be provided after the foam buoyancy which clads the outside of the riser itself is taken into account. It might not be impossible to add more buoyancy on the riser to improve the situation as you describe, but would be difficult.

The underwater transverse structural bracings between the pontoons of the semi would obstruct the ability to pull the rig away from the riser (even if the riser could be remotely disconnected and lowered somewhat at deck level).

I would say that somehow improving the security of the protection at the seabed end of the riser would be the best focus - although that has always been the objective with FLOATING drilling rigs - and perhaps we suffer from too much "conventional" thinking.

I think others on some other threads have made some plausible suggestions there - like "fail to closed" actuation.

 

[docellen]

JR, good discussion. If the riser requires 2000 tonnes of tension to hold it up, how do they disconnect temporarily during a hurricane, as they are planning now with the rigs drilling the relief wells?

If we assume that the pontoons, columns, and deck box were all designed to modern standards, and had separate compartments, then perhaps the damages from the explosions were severe enough to open up large numbers of these compartments.

Maybe they should have perforated the bottom of the deck box, as soon as they saw the rig heeling over. Maybe we need connections on each compartment of the pontoons, so we can suck out the water and provide additional buoyancy. Just trying to think of simple, low-cost solutions.

 

[waross]

The simplest, lowest cost solution may have been to take the time to shut down long enough to repair the BOP before all this started. The rig sinking was a result or symptom, not a cause.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[JRMacGregor]

Waross is right - the sinking was a SYMPTOM - not a cause. The real defence was to be found earlier in time and further upstream in the process. Once a fire of that intensity takes hold, it is hard to design any quantifiable/reliable defences against it.

Nevertheless, it would be interesting - although perhaps irrelevant - to know if/when the rig would have sunk if the fireboats had not fed her with water. Unburned oil flowing into the spaces would have just as bad an effect on stability as the firewater.

As for the riser when it is disconnected, it disconnects at the seabed and then hangs below the rig. It is then in TENSION due to its self weight. Tension (within limits) is a good thing for the riser as it avoids buckling in compression.

If an unsupported riser was connected to the BOP on the seabed, it would be "standing up" in compression and in danger of buckling due to it being incredibly long and slender. To change that compression to tension is why the buoyancy is added on the riser and tension is applied at the top (on the rig).

Compared to the connected riser, the only extra force the disconnected riser applies to the rig is the force which was previously acting downwards on the BOP at the seabed. And this load is limited for reasons of BOP integrity.

When the rig sank, the uplift on the riser provided by the rig and her riser tensioners disappeared, and this would have increased the downwards load on the BOP and LMRP assembly at the seabed. I don't know if this would have contributed to the post-blowout failure of the BOP to operate.

Incidentally, one area where deepwater drilling is safer/easier than shallow water drilling is in terms of riser angle. The connection of the riser to the BOP can tolerate only a limited riser angle. This angle is a function of the water depth and the horizontal offset of the rig at the surface.

The horizontal offset is caused by the wind/waves (and perhaps DP system failure). And 100m offset in 140m of water (e.g. North Sea) is far, far harder for the riser to stand than 100m offset in 2000m water depth.

This water depth thing seems to be over played in the media. On any well beyond diving depth (say 300m at a stretch) the recovery problems would have been very similar (totally dependent on ROV).

In fact, even had the well been within diving depth, it is difficult to see what saturation divers could realistically have achieved that has not already been done at Macondo.

 

[docellen]

The simplest, lowest cost solution is to monitor the mud flow, and detect a kick at the earliest possible moment, when it can be easily controlled. What happened was not *caused* by bad design, but it could have been *avoided* by better design. It bother me that engineers are not more interested in finding these better designs. It's like having air bags fail because someone forgot to change a battery. We can't just say "not our job".

 

[waross]

The simplest, lowest cost solution is to monitor the mud flow, and detect a kick at the earliest possible moment, when it can be easily controlled

Monitor the mud flow. Controversial decision to replace the mud with sea water. Mud? What Mud?
Detect a kick at the earliest possible moment. I understand that the kick was detected early enough and the annular activated early enough. Annular? What annular?
As for the design of the BOP; I have wondered why the shears are not designed to be capable of shearing a joint. Other than that, this BOP had failed operations on about four or five devices. (Two annulars, two pipe blinds, and an inverted pipe blind.) It doesn't matter how good the design is if someone turns it off, (as may have happened to one of the annulars) and/or doesn't repair it in a timely manner.
I don't say that there is no need for modification, but I think that changes may be more appropriate if they are such as to facilitate easier and quicker on site repair by ROVs and make provision for ROVs to quickly and easily connect to the hydraulic systems. Another possibility may be to use edible oils in the hydraulic system so that repair operations that involve the loss of hydraulic oil are not so damaging environmentally.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[BigInch]

Hey there's an idea. What flavors?

Look at all the stuff on a Dreamliner. Still won't stop somebody with poor judgment from flying it into a building. Human brains are part of the control of these things too. Let's be sure they function like they're supposed to before we turn them lose on the world with WM-Environmental-D.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv

 

[waross]

Hi BigInch. The food processing industry uses edible oils in gear boxes in food handling areas. I'm sorry, I don't know what flavors are available.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[DrillerNic]

Waross- I'm tired of saying this, but the decision to partially displace the top of the casing and part of the riser to water ISN'T controversial- it's the only way to do an inflow pressure test on the casing....

Having done the inflow test and convinced themsleves it was ok (lets ignore the 15bbls additional bleed off, and some of the weird pressure build ups, eh?), they then continue displacing the riser to sea water as part of the suspension operations- they have a pressure tested casing (both ways!) and a BOP, so it's a perfectly normal operation.

Also , looking at the BP presentation, the kick was detected early enough at about 21:00 (almost an hour before the explosion) while they were displacing the riser (but no action taken.... why?) and then with spacer back at surface, stopping for a sheen test before diverting returns overboard. But they continued displacing!! Why??? And then at 21:31, they stop displacing again, and I think by then they'd clearly identified something was wrong... the driller calls the Toolpusher 4 times but doesn't shut the well in!!!

One of the annulars was closed during the inflow test, (and then opened again for the displacement), but the well was never shut in despite all the indications, until 7 minutes AFTER the first explosion.

Now all of this is obvious with hindsight (and hindsignt is always 20:20.... I lost a $2million BHA once, and with hindsight it was obvious I'd nearly got stuck twice before it did get stuck), but I doubt the reported problems with one of the annulars have much bearing on the BOP failure; the issues with the hydraulic leak might explain why the susbsea acculumators didn't have the power to close the well, and certainly caused problems when they were trying to actuate the BOP with an ROV. But we'll wait until the BOP is recovered to surface to find out what happened with the BOP, and if we have to alter all our BOPs or if this was just a cock up (like the Ekofisk blowout, where part of the BOP was upside down!)

The last thing is that as a subsea BOP control system vents to sea, I'd expect the hydraulic oils to be non damaging....

 

[BigInch]

Don't count on hydraulic oils being safe to drink. Well ... I certainly wouldn't do it.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv

 

[DrillerNic]

well BigInch, glycol isn't safe to drink (outside of Russia....) but it's on the OSPAR PLONOR list of things not damaging to the marine environment!

 

[BigInch]

Ug! That's why there's a zero discharge policy in US waters. No confidence in the listed items.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv