Gas Leak at North Sea Elgin Platform

A crisis situation has developed at a gas and condensate production platform in the Elgin field in the North Sea. Gas is leaking out of a well near a offshore platform at a rate of approximately 2 kilograms per second (12 MMCF/day if gas), and a large sheen (assumed to be condensate) has been observed on the water. All workers on Total's Elgin PUQ (production-utilities-quarters) Platform plus those on the Rowan Viking drilling rig, which had been working next to it, have been evacuated. On Monday, workers on a platform and drilling rig at the Shell-operated Shearwater field (4 miles / 6.4 km away) were also evacuated. There is currently a two-mile vessel exclusion zone around the site and a no-fly zone. As current winds are light, the most immediate concern is the potential for explosion both at the PUQ and elsewhere. While it is possible that the leak rate will lessen over time, the Rowan Gorilla V jack-up drilling rig is being provisioned by Total for a possible relief well that could take months to drill.

The Where

Elgin is a high temperature/high pressure (HT/HP) gas field in the middle of the North Sea between Scotland and Norway at an ocean depth of about 90 meters. The reservoir is about 5500 meters deep with temperatures near 190°C and 1100 bar pressure. It was discovered in 1991 by Elf Exploration UK Ltd. It is being produced along with the nearby Franklin field using one unmanned wellhead platform in each field. The PUQ platform, connected to the Elgin wellhead platform by a 90m bridge, processes the sour gas and condensate for pipeline transit to shore. Production from the combined fields currently averages 140,000 boe/d (though a report on Rigzone said 230,000 boe/d), and accounts for 10% of UK gas production. Total recently upped its stake to 46.2% in the project, where the focus is shifting to new prospects in the West Franklin field. From SubseaIQ:

This phase of development is aimed at producing estimated reserves of 85 million barrels of oil equivalent, and will consist of drilling three wells and installing a new platform tied-back to the Elgin/Franklin facilities.

The map below shows the location of the fields and the platforms within.

The picture below is probably representative of the location of the
Rowan Viking relative to the Elgin PUQ platform.

Rowan Viking rig next to Elgin PUQ platform

The Upstreamonline report provided the following on the sequence of events:

Hainsworth said the problems had occurred during operations to plug and abandon the field’s G4 well.

“On Sunday there was quite a sudden rise in pressure and they were finding it difficult to bleed off this rise in pressure,” he said.

“That initiated what followed. Witnesses who were working on the well saw a release of what we expect is mud from just below the wellhead at the top of the casing followed by gas.”

Hainsworth said that gas is being released on the Elgin wellhead platform at low pressure – about five bar - and that it is coming from a non-producing reservoir above the Elgin formation.

The Telegraph has something similar:

The exact cause of the leak has still to be confirmed but Mr Hainsworth said workers at Elgin on Sunday had reported "a spray of liquid followed by a gas release" from a well casing.

The well had ceased production a year ago when it was plugged at its reservoir source, 6km below ground. Workers spotted changes to the pressure in its outer casing weeks ago and had been in the process of pumping in heavy mud to "kill" the casing on Sunday.

Gas is thought to be entering the casing from another, non-producing reservoir 4km underground, Mr Hainsworth said. Total did not yet know the capacity of this reservoir but in a "dream" scenario it could simply "run itself out".

From HazardEx comes this third-hand report:

"Engineers have told me that it is almost certain that gas is leaking directly from the reservoir through the pipe casing."

The Well

The Elgin 22/30c-G4 well was the first development well drilled there. A paper in SPE Drilling and Completion, Field Hydraulic Tests Improve HPHT Drilling Safety and Performance discusses the drilling of this well, specifically on some testing that was conducted above the HT/HP regime (but at greater depths than suggested by Mr. Hainsworth's assessment above of where the gas is coming from).

One problematic zone (the Kimmeridge clays) is situated at a vertical depth between 5130 and 5370 m. The problem of the Kimmeridge clays is the uncertainty on the native pore pressure gradient (between 2150 and 2200 kg/m 3), and the possibility of a ballooning effect.

According to this source, the well was producing 20,000 barrels of oil equivalent per day and gas at a rate of 2.6 million standard cubic meters per day in 2008.

The Why

The well was in the process of being shut. But why? The HazardEx link provides another quote:

"The well in question had caused Total some problems for some considerable time ... a decision was taken weeks ago to try to kill the well, but then an incident began to develop over the weekend," he said.

This source says it has been shut off for a year. A very recent paper in the Journal of Petroleum Technology as some very useful information:

Advanced Drilling in HP/HT: Total’s Experience on Elgin/Franklin (UK North Sea)

The paper discusses infill drilling in the Elgin and Franklin fields. After over 15 years of production, these fields are rather pressure depleted (over 800 bar!), which causes sand infiltration problems and liner deformation due to compaction of the sandstone reservoir. It is possible that well G4 had several problems and they eventually gave up on it.

Given the pressure depletion and the fact that both fields seem to be well past their prime, can they possibly still be responsible for 10% of UK gas production? Perhaps not. From HazardEx:

Peak production capacity for the Elgin/Franklin field is 280,000 barrels of oil equivalent per day (around 7% of UK production), 175,000 barrels per day of condensates and 15.5 million cubic metres of gas per day (mcm/d), according to Total.

7% of UK production is less than 10%, but then this:

The Elgin-Franklin fields produced a daily average of 61,386 barrels of condensate in November, according to the most recent government data.

Condensate production is only about a third of original capacity, so gas production might be proportionally lower. Of course, UK production is down overall as well...

Here is more information on the development of the fields:

Elgin/Franklin: What Could Have Been Done Differently?

(note: article starts on page 54 of the above)

The What Next

Given that there is nobody within two miles of the platform at present, and the story is still unfolding, it is prudent to be cautiously skeptical of Total's analysis of the situation. Some questions:

  1. 1 kg/sec (reported by several sources) is a very high leak rate from a formation they drilled through to get to the one they were going to produce. In any case, if the report is true, how was this estimate made?
  2. BBC has reported that the flare is still burning high above the platform. If the wells have been shut off, why is this?

The gas is sour, with hydrogen sulphide. The gas is bubbling up from the sea bottom through about ?50 metres of sea, and the HS will be partially absorbed into the water, turning it toxic. The North Sea as a very degraded marine environment already, but it is still a major (and heavily overfished) fishery, in danger of total collapse. This is not going to help.

The gas cloud is spreading over the surface over many square miles. The current weather is a stable high pressure area producing a moderate North wind in the area. This brings the air mass ashore, around about here (Eastern England) around about now. I wonder how long before atmospheric methane concentrations start rising (or if anyone is monitoring them).

The UK media is slowly waking up to this emergency, but clearly have no idea what the important questions are.

The gas is sour, with hydrogen sulphide.

So much for being thankful it is gas and not oil. How will that gas in solution effect the heat transfer and therefore the 'flow' of the other side of the Gulf Stream?

At least it's not yet another failed Fission reactor.

According to the "Advanced Drilling in HP/HT" linked above (and other sources), the H2S content is 30-40 ppm. Compared to most sour gas fields around, this is really minor. Ignition is the bigger problem.

What would likely happen upon ignition?

A big boom. Most likely lose the rig, with the fire only extinguished when either 1. someone can cut off the flow of gas or 2. the riser falls below the surface of the sea.

Damage to the facilities. There is a lot of money tied up with a couple platforms plus the drill rig. There might also be some unprocessed condensate in a tank somewhere.

I found this link

http://www.guardian.co.uk/business/2012/mar/28/flare-burning-north-sea-g...

A flare is still burning on the North Sea platform that has been leaking gas for the last four days.

Total, the French oil company that operates the platform, first disclosed that the flare was still burning late on Tuesday night. The wind is blowing the gas in the opposite direction but, if it should come into contact with the flare, there is a high risk of explosion.

Thanks for the post.

On 1. You can make a rough estimate of the flow rate using knowns such as choke size and reservoir pressure and then use an estimated height of gas flare to give you a volume of gas per unit of time. You could also take the condensate sheen volume areal extent estimate and then calculate the gas rate using the gas/condensate ratio which I expect Total would know.

On 2. I am not a facilities engineer but as I understand a flare is always lit for safety reasons. Kind of like a pilot light. Can you imagine trying to light the flare once the gas has been diverted to the flare? Once on an onshore rig I watched a leasehand light a bucket full of oily rags and throw it into the flare pit to ignite the gas we were testing. It eventually worked after a few attempts but no-one likes having unignited gas floating around a lease.

... calculate the gas rate using the gas/condensate ratio

Initially, 320 bc/mmcf, a really rich gas/condensate.

...175,000 barrels per day of condensates and 15.5 million cubic metres of gas per day (mcm/d), according to Total.

But it is uncertain where the gas and condensate is coming from, I think.

1. Is the lack of media attention due to the fact that this incident has not yet resulted in a massive ball of fire and loss of life like the Macondo disaster in the Gulf of Mexico?

2. Will they be using submersibles with live video feeds of what is happening at the sea floor??

1. Yes.

2. No idea. Not unless there is a lot of political/media pressure to provide it.

The well that has blown out was just one horizontal well that has been drilled at this production rig. The remaining wells have been shut in, reducing gas flow to the UK, and raising the spot price back up to about 60p therm, up about 20%. This is about what the price was through most of the (relatively mild) winter, and current demand is low because of our local mini heat wave. Still, we are failing to refill storage which we would otherwise be doing.

I think the well was only recently drilled, as part of the re-working of this mature field. but had major production problems, and was in the process of being killed and shut in.

Does anyone one know if the well was fracked as part of the managment process? Given the local opposition to on-shore shale gas drilling, this would be a major PR blow to the industry if the blowout could be linked back to the fracking process.

2. Will they be using submersibles with live video feeds of what is happening at the sea floor??

Given that the water depth below surface is only 90 meters (or ~295 feet) I would think that human divers could be a possibility (something that was not possible at the 5,000 foot depth of BP's Macondo well in the GoM) - or maybe a combination of submersibles and human divers?

I don't imagine they would want human divers close enough to the leak to do anything useful. Send in the drones!

It is shallow enough to let humans dive in it (I think the world record is below 300 meter, the dive took a full day under water with all the decompression stops) but it is not easy. I am alowed to dive down at 40 meter with my SCUBA certificate. Below that we get into technical diving wich is a completely different animal. The big issue will be all the man power above the surface, the limited down-times and the plan-ahead time involved.

Those robots are more likely to get involved.

CNN has some commentary this morning.

Energy giant struggles to contain offshore gas leak

Apparently the risk of explosion is viewed as lower now, because there is still a flare burning on the rig, and has not ignited any gas. At least, thus far.

I would still regard the risk of explosion as a matter of when, not if, as long as the flare is alight, and the sea level gas release is at it's current rate. How long before the wind turns light and variable, leading to local eddies that twist the gas plume in the direction of the flare? 90 metres is too narrow a margin for error.

It did occur to me that the US media might be publishing commentary that downplays the risk, due to the recent interest in opening up more offshore areas to oil and gas drilling.

I saw an earlier report indicating a flow rate of 2 kg per second (with a big margin of error), which if it is methane would be about 9 million cubic feet per day. This would fill a hemisphere of about 300 ft diameter in a day (calculated after supper including accompanying libations). Lower explosive limit for methane is 4 to 5%. So by a very rough and ready calculation if there is about one air change per hour around the platform the concentration shouldn't get above the LEL at the flare - I think you get much more diffusion than this outside even on a still day. Of course rough and ready isn't a very good basis on which to determine the fate of a multi million pound facility. I would guess there are some high powered brains conducting a lot of CFD dispersion studies to see what concentrations are likely to be seen at the flare. I think the flare would have needed to be left on during evacuation to ensure all the facilities could be blown down - it would have been worse to let the high pressure gas from the facilities disperse out of the flare tip without being burnt, there would have been higher probability of it finding an ignition source then, and a higher consequence from the explosion, than for the leak gas now.

Methane is lighter than air too, so it will tend to dilute quickly.

A layer of propane hanging just above the surface of the water would be be very bad though.

Where does the gas in the flare come from? Another well that can be turned off with the push of a button, or from the leaking pipe? In the latter case, we need to let it burn to reduce the emissions and thus the risks.

My thinking behind the leak if it is coming directly from the reservoir rather than up the well that it would be a result of the depletion of the reservoir, in particular because the report states that they were planning on abandoning the well. In other HPHT fields there have been difficulties in re- entering older wells to carry out work over due to reservoir depletion allowing the formation to expand and causing casing/cement failures. I believe in one other field it was actually so bad that the casing sheared.

So all I can imagine at the moment is somehow it has migrated up into the ntermediate or surface casing through the sides of the products casing and there is some integrity issues with the surface/intermediate casing that is allowing it leak. ESP if the well heads are wet trees I can see how this will be a problem to kill it.

As I understand it gas bubbling up through a liquid lowers the density of the liquid. How much gas needs to leak up through the water to lessen the buoyancy of the rig to the point that it sinks?

Unlike the situation with the Deepwater Horizon, this thing is up on stilts sitting on the seafloor.

Team,

It appears, as new info comes in, that the leak is at the well head, which in on the wellhead production platform, high and dry.

The alarm was raised by workers just after noon on Sunday after mud sprayed out onto the wellhead platform, followed by escaped gas.

http://www.dailymail.co.uk/news/article-2121457/North-Sea-gas-leak-threa...

(Something is wrong when you have to rely on the Daily Mail as an information source)

Don't take too much notice of all the pretty pictures in the press as I don't believe any are current. The bubbling up around the leg of the processing platform form will just be cooling water, with the platform under normal operation. My first thoughts were this was the leak, so much taking notice of the press.

Therefore at the moment, to be corrected as new info comes online, I don't believe there is any gas in the water, only in the air and just waiting for a spark to let us know of its presence.

Edit, I stand corrected already, same article,

‘They saw the sea bubbling with gas under the platform,’ he said. ‘This is the well from hell. This problem is out of control ... and is only going to get bigger and bigger.’

we wait on more details, but sounds like some major casing failures.

Photobucket

David Hainsworth, health, safety and environment manager for Total, told the BBC that earlier information provided by the company about the flare was wrong.

Mr Hainsworth said that the flare was "still alight", adding "we don't believe it has been reduced in size".

He said he could not put a timescale on the flare being extinguished.

Mr Hainsworth said it was not possible to say whether that would be "an hour, or 24 hours or two days" - or even longer.

He said there had not been time to extinguish the flare when the Elgin platform was evacuated, as the safety of staff had been the priority.

But had there been time they would have considered putting it out, he said.

It was not possible to do that remotely.

"When you have a significant amount of gas escaping, there is a case for trying to burn some of it off to get rid of it rather than leaving it as a hazard elsewhere”

Jake Molloy, from the RMT union, said it was "beyond comprehension" that the flare was still burning.

http://www.bbc.co.uk/news/uk-scotland-north-east-orkney-shetland-17522086

"Jake Molloy, from the RMT union, said it was "beyond comprehension" that the flare was still burning."
R most components except H2, Heaver than Air?? Anyone have a composition breakout from this field?
Monty Python's quote "run away" not applicable here, Perhaps "swim away, way away".

I expect that remote shutdown capability of the flaring system will be made standard in the future. That it isn't required now is curious; seems like a no-brainer. It can't be as complex and expensive as the BOPs and their panic button.

When will we learn to stop creating energy producing systems that can't just be turned off?

second guessing the engineering on these things is out of my purview... and I expect most people even here.

sometimes automated or remote safety systems are left out of systems because premature activation can lead to catastrophic failure?

eg automatic altitude sensitive deployment for parachutes are removed from skydivers jumping in a group as opposed to a solo student skydiver

"shrug"

According to Total,

The flare is an integral part of the platform's safety system. In an emergency it is used to safely evacuate all the gas from the platform. During the incident it performed this task perfectly, allowing everyone to evacuate safely.

When the emergency de-pressurisation is initiated, all hydrocarbons feeds are closed and valves are opened on installations vessels to de-pressurise gas to the flare. These valves remain open as they are designed to in such circumstances.

The flare is still lit because when the platform is shut down and de-pressurised in an emergency, it cannot be fully purged as done in a controlled shutdown. This is perfectly normal. Some liquids do remain in the system and these liquids are now evaporating. As these liquids evaporate the flow of hydrocarbons to the flare will exhaust itself and the flare should burn out.

At present the flare does not pose any immediate risk as the layout is designed to take into account the prevailing wind direction, ensuring that these winds are taking any gas from a potential leak in the wellhead area in the opposite direction to the flare. This is in fact exactly what is happening. The wind is forecast to remain in its current direction for the coming days. You can be assured that this is being reviewed on a constant basis and should this change any impact is being assessed. In parallel we are investigating solutions to extinguish the flare if it does not burn out by itself.

http://www.totalepmediacentre.com/index.cfm?encstrg=8%3E%2BJ23%21%3A%3DT...

Saved by the wind, again. Tokyo, now Franklin.

If the rig has been fully shut down to avoid sources of ignition how would remote work? If remote is working then the rig wouldn't be totally safe and ignition source free.

NAOM

Ghung,

The idea of a flare stack is to have an uninterrupted controlled passage to a safe area. These lines are designed with minimal bends, valves and restrictions, and a constant pilot light, either run on produced gas or LPG bottles. Once you put in place control valves to extinguish the flare, you increase complexity and the risk of failure.

Think back to the DWH, someone thought it was a good idea to avoid some minor pollution problems, to plumb the diverter line into the mud gas separator, and then had it pre-selected for this route on the panel. So when the panic button was hit, the mud and gas instead of diverting overboard, ended up on the main deck and the pit room with deadly results.

KISS, is a wonderful concept, Keep It Simple Stupid, especially in emergencies, but every solution always seems to make systems more complicate and susceptible to failure.

As to the flare still being a light, Total says it is just some condensate evaporating. I hope so, because the alternative is leaking SSSV and Christmas tree valves. I don't think I would be volunteering to rush in with a bucket of 601 Cameron grease to get them to seal.

I would feel a lot safer if the leak did catch fire, hopefully in a controlled fashion of course, maybe on a windy day to avoid an explosion. The well head platform would be a right off, but the area would safe to operate in.

I am sure the Total execs are currently finding religion, and preying that it is only a small pocket of gas. Time will tell.

I was thinking along the lines of a simple pneumatic or spring-loaded valve, something like the emergency main steam stops that many ships have. Reading ROCKMAN's post below, there seem to be situations when it is prudent to eliminate all sources of ignition, including the flare. This wait-and-pray-only-option- scenario goes against my engineering sensibilities. The articles indicate that there is a mechanism for shutting down the flare which is now out of reach.

Ghung,

The idea of the flare is a vent line on fire. The ignition of this vented gas maybe controlled, but once vented gas is alight then it will burn, What you are suggesting is to close the vent system to this gas, which could build pressure some where in the system which is not designed for it. The flare/vent line is the circuit breaker in the electrical system, it is where the gas / fumes go when you have nowhere else for them to go. Start putting valves in this system then you will start having situations like the DWH, where in emergencies gas/fumes don't get the release they need.

It's just a pressure relief system that happens to be on fire. The steam stops I mentioned above have a relief valve that shunts excess pressure to the stack (atmosphere), rather than to an engineering space. Flaring systems could shunt gas into an underwater outlet. It now seems that there is at least one scenario when this would be preferable.

Again: See ROCKMAN's posts, below. From his latest:

"But if they were spewing 12 mmcfpd at the well head and the flare didn't ignite it...someone got very lucky."

He seems to feel that removing the flare as a source of ignition is SOP in a case like this; no longer an option it seems.

Ghung,

The gas going through the flare is the blow down from the process equipment, and nothing to do with the leak at the well head platform. As per Total, it is still burning due evaporating condensate in the lines or vessels. Or it could be leaking well head valves, but a minimum four valves in series would need to be leaking for this to happen. Lets go for evaporating condensate.
The leak at the well head platform is an extraordinary event. The well has lost containment. It is not going to the flare.

"It is not going to the flare."

Jeez, let's hope not. Sounds like they don't know where it's going and are just hoping a freak wind gust won't connect this huge and highly combustible mixture to the flame before the flare burns itself out. Matters little that the sources of gas are different. The potential for ignition (and, IMO, a worst-case-scenario) exists. It must be my training; whatever can go wrong eventually will. Better to be prepared, have options, than to be caught with pants down. Seems these folks may have tripped over their trousers on the way out of the crapper. Reminds me of an old Navy joke: What do you do when you see boiler techs jumping over the side?...

I know... armchair engineers and all that, but in my experience with high pressure, high temp systems, we left very little to chance ;-/

North Sea gas leak venting from newly disturbed source

"We've got geologists working on the productivity of the horizon [reservoir] the leak is coming out of," Andrew Hogg, a spokesman for Total, told New Scientist. "We must do some modelling to find out the rate."

Although the main reservoir itself at the base of the drill shaft is safely closed off, Hogg says, the gas from the secondary source in chalk above it is escaping by leaking into the shaft containing the drilling tubes that lead down to the main gas reservoir.

Thanks very much for creating this thread. It's as informative as I imagined. The problem seems just as thorny as Macondo, while different in its particulars. The size of the evacuation zone would seem to indicate the size of the problem. But if the secondary reservoir is doing the leaking, the proposed relief well won't solve the problem. Up close assessment of the sea floor is required, and soon.

Regarding the proportion of UK production affected: how much is from the reservoir below the platform and how much is just passing through? How much is controlled from the platform but moves elsewhere?

The mass flow rate could be determined from the size of the immediate gas cloud plus wind data. Specific software exists for this purpose. How you determine the size of the cloud I don't know. Infra red camera perhaps? Haze shimmer?

The explosion hazard range could be determined using the UK gas industry standard SHA1 (or its IGEM equivalent). Care should be taken to ensure the mass flow rate is within the limits of the methodology.

I think the hazard range would be alot smaller if it was on fire. As a person with some nat gas distribution pipeline experience I would say that the most dangerous leaks are the big ones that DON't go on fire.

Following Piper Alpha there should be emergency shut down valves on the sea bed- but how much UK supply will be affected by their operation will be a concern at a time when emergency fuel reserves are being released.

The idea of tackling a platform with a gasoline puddle on the sea and gas in the air sounds quite tricky to me. Ay least at Macondo the relief well drill could take up a position near the leak.

The condensate goes into a spur of the Forties pipeline system, and the gas into the SEAL pipeline:

http://www.offshore-technology.com/projects/elgin/

The pipelines are on the seafloor, so it's not clear if the gas flow from Shearwater has to be stopped.

5500 meters = 18045'
1100 bars ~ 16000 psi
190 deg C = 374 deg F

The reservoir was geo-pressured, a gradient of about 0.89 psi/ft.

After over 15 years of production, these fields are rather pressure depleted (over 800 bar!), which causes sand infiltration problems and liner deformation due to compaction of the sandstone reservoir.

At 800 bars(~11,600 psi), the reservoir is still geo-pressured (0.64 psi/ft) and will probably continue to compact until a normal gradient is reached.

At 800 bars(~11,600 psi)

No, I read that as depressurized by 800 bar. From the conclusion of that paper:

The first infill well was drilled successfully, completed, and
perforated in an HP/HT reservoir after 660 bar of depletion
had occurred.
...
Two additional wells have been
drilled successfully through more-severely-depleted reservoirs
(approximately 800-bar depletion).

Tech Paper here:

http://www.spe.org/jpt/print/archives/2011/10/19TTS.pdf

“the mud weight window closes at 100 bar”.

vs

“1,100-bar virgin pressure and
200°C, respectively”

Seems the mud can’t hold it. Would it would be easier if it was deeper (like Macondo). Operating out of the mud weight window a cement plug option is not possible I suppose.

How would a relief well help if you are outside the mud weight window? Any relief well would blow out too.

And can you drill at all. One article I can’t re-find suggested all production wells were drilled before production started. Implying that follow-on wells would be difficult or impossible

“Another HP/HT operator was unable to
reach final depth of an infill well because of insufficient formation
strength resulting from depletion, which prevented
safe drilling of the reservoir section.”

“Downhole measurements showed that most Elgin/
Franklin production liners have suffered a loss of internal
diameter of up to 60%”

ie the rock is moving so at 1000 bar leakage past a cementing job can’t be so surprising.

“Often in Elgin/Franklin, some thin limestone layers (centimeter-
to-decimeter thickness) are found to be gas bearing in
the caprock. These zones can be drilled underbalanced, but
sometimes require high mud weight to enable trips.”

Perhaps the leak is not from the production reservoir but from gas bearing caprock in which case letting it hiss for a while may be the safe option.

Glossy brochure here:
http://www.uk.total.com/pdf/Library/PUBLICATIONS/Library-ElginFranklinBr...

I discussed that first paper near the end of my post.

Another paper that I linked seems to have what you remember:

Elgin/Franklin: What Could Have Been Done Differently? (scroll down within to article)

At production startup, all development wells were drilled but one; the last Elgin well was drilled with 90-bar depletion (i.e., just before the closing of
the mud-weight window). By the end of the development-drilling phase, there were 12 producing wells as planned, although one additional well was drilled on Franklin and one fewer on Elgin.

Failure in recovering Elgin appraisal wells meant that a total of only six slots remained available for future drilling (three on Elgin and three on Franklin), instead of nine as planned.

Max – My net connection has been spotty so I haven’t been able to get to get all the facts. But here’s the situation as I understand. The potential drilling problem isn’t that it requires a high mud weight but just the opposite. During the normal drilling of a well we’ll increase mud weight gradually. Ideally you want your MW to be a little higher the formation pressure in the rock. Varies but often around 0.5 pounds per gallon. Normal pressured rocks are usually drilled around 9.5 ppg to 11.5 ppg. But there can be sudden increases in pore pressure and the MW needs to be raised. But you can’t do that with normal pressured formations in the hole. The higher MW would force the drilling mud into those porous rocks. We call that “lost circulation”: the mud is pumped down butt doesn’t come up…it gets lost. Very dangerous and can be costly. The drill pipe can get stuck due to that flow.

The problem with drilling through a pressure depleted zone is that you have to lower MW to get through it or you lose circ through the zone. This happened to one of wells just a month ago. We knew there was a small possibility of zone being partially pressure depleted. Turned out it was which we discovered when we lost circ in it. It required setting a string of casing above the LC formation, lowering the MW, drill through the zone, set another string of casing, raise MW and the drill ahead to the deeper target. It can be done but increases the well cost significantly.

Still unclear but if I understand correctly they think the blow out is coming from a shallower fractured chalk formation. One report said the flow was up a casing annulus. An annulus is that gap between a string of csg and the rock or between two strings of csg. The primary reason for pumping cmt into an annulus is specifically to prevent this from happening. As a deep well is drill there may be several string of csg run…some all the way back up to underneath the well head and some just hung from the bottom of the last csg string (we call it a “liner” in that case; “casing” usually implies it’s run back up to the well head). These various annuli are pressured tested from the well head connection. If they don’t test to a sufficient pressure we’ll pump more cmt down and retest.

If it is an annular blow out why did I happen now? I have no idea…not nearly enough info. But if it's flowing up from that shallow chalk zone then the pressure depletion of the deeper reservoir won’t be a problem with a relief well: it won’t be seen in the RW. But there is a safety system on every well to prevent any annulus from blowing out…the BOP. The BOP is supposed to control flow from every annulus in a well. All the annuli come up into the well head. But if I understand the situation correctly (and I don’t think I do) they may not have had a BOP installed. After a well is drilled the BOP is removed and replaced with the well head that can be used to shut off any flow. Often when you’re going to work on a previously producing well you do install a BOP. I can’t tell if they did or not.

Now about all the unbelievable BS about this not being a dangerous situation because the wind is blowing this way or that way, the burning flare stack is blah, blah blah. Let’s try this: you have a kiddie poll filled with 100 gallons of gasoline and you cooking on your charcoal grill 20 feet away. Dangerous? Give me a freaking break. Are you aware that the most powerful non-nuclear bomb in the world is exactly what they have going on out there? It’s called a fuel-air bomb and it essentially releases a cloud of mixed petroleum gas that’s ignited. All it will take is a mixture of the spewing gas to mix with the oxygen in the air in sufficient volume and be ignited by the burning flare or any other ignition source on the rig or from a passing boat. In one report they actually point why they have flare stack: to burn off the gases to prevent a potentially explosive accumulation. The situation out there has the potential to be one the greatest manmade explosions of all times. Notice the 2 mile safety zone? There's a dang good reason for it and the operator knows why. A NG explosion is what scares the crap out of us: oil burns...NG explodes.

Other confusions: some are reporting the gas is coming up from the sea floor at the bases of the well casing. This can happen and has happened. Again, not enough info. There seems to be a serious lack of detailed info coming out so far. They may trying to minimize public concern or maybe they just aren’t positive they know what happened. But for what it’s worth (not much) I don’t think the well will kill itself naturally. That can happen with an open hole blow out (called “bridging over”). But this doesn’t sound like that situation. Just a wag but they will have to drill the relief well, cut through all the casing strings and pump heavy mud down to stop the flow. With that done they can re-enter the well from the top and plug it properly.

I guess this means nuking the well is not a viable option? ;-)

Rockman,

Your insights into this situation have been sorely missed in the past 48 hours.
Therefore, next time you leave the country without express written pemission from the TOD Community, you will be forbidden to purchase or eat Blue Bell ice cream for a period of six months. And that is just the punishment for the first offense.

Stay safe my friend!!

Roger jarhead. I wasn't counting on getting to the net at all so I can't complain when it throws me off now and then. Had some local ice cream today. Not quit BB but not bad for west Arica.

I've been watching BBC but like US MSM they aren't very good on details. Not much else to do...jet lag has me wide awake at 2 AM...27 hour transit. Off to the bush tomorrow so it will be tomorrow night before I'll see any news updates.

A reminder for folks on the conversion since pressures aspects will be popping up: Mud weight (pounds per gallon) X depth (feet) X 0.052 = pressure (psi) and 1 Bar X 14.5 = pressure (psi)

Basically it looks like they've taken a kick they need to circulate out. Sea water in, natural gas and condensate out. Problem solved.

Bruce,

They had pressure on an annulus, which they were bleeding of, got a bubble of gas.
I found this in a news report, but couldn't find it when I wanted

I don't know if they shut it in, I would like to think so, but then it seems to have blown out sub sea.

Workers on a standby ship at the Elgin field reportedly saw vapour clouds forming and gas bubbling on the surface of the water under the platform

http://www.telegraph.co.uk/finance/newsbysector/energy/9166853/North-Sea...

So either the pressure in the annulus burst the 20" casing or blew out the shoe to gain access to the surface formation and blow out sub sea. If flow path this is connected to a major pocket of gas it could be a bitch to control.

Stipulating that the flow is up the annulus between the casing and the production casing, it would seem that hot tapping the casing between the sea surface and the mud line would allow access to the annulus (see the hot tapping video used on the Costa Concordia http://www.boskalis.com/press/media-library/videos.html?videoCategoryId=566 ) which would be a conservative first step.

Then lubricate and bleed the annulus to kill the well. Given that the well is continuously bleeding, you could do a slow pumping rate continuous lubrication. You only need 5 bar or about 73 psig to overcome the leak. Given that the hydrostatic pressure at the mud line is much higher than 5 bar (8.5 x 0.052 x 295 = 130 psi = 9 bar), a simple hole in the casing would let sea water flow in and kill the well.

Got an ROV and a drill?

The above is from a Total presentation in Moscow in 2005. The PowerPoint (mostly Russian text) had several examples of drilling HT/HP wells. One being discussed was the drilling of a well in the Glenelg field, which is next to Elgin (the well is connected to the Elgin platform). Anyway, the well in the above (used as an example) is the Elgin-G4 well in distress. The title of the slide is Градиент давления на забое or "The Pressure Gradient on Bottom", but Google might have that wrong.

The Kimmeridge Clay was also mentioned in the quote I put in the main above:

One problematic zone (the Kimmeridge clays) is situated at a vertical depth between 5130 and 5370 m. The problem of the Kimmeridge clays is the uncertainty on the native pore pressure gradient (between 2150 and 2200 kg/m 3), and the possibility of a ballooning effect.

I believe you're correct with "Градиент давления на забое" and its meaning.
Roughly it's pronounced "Gradient davleneya na zaboa" or basically "Pressure gradient at the bottom" since na can mean "on", "at", "upon" etc, depending on context.

Very interesting website and comments here. I wish I understood it all!

Rockman, you are very good at explaining the complicated procedures you are dealing with when drilling. I really appreciate your comments on this as well as during the DWH disaster. I understand the geology part somewhat but not the drilling and procedures so your input really helps us layman follow along.

ec - You're welcome. The more posts I read the more confused I get. Was the well drilling? Didn't sound like it...sounded like they were trying to plug a cased hole. Then why is Total showing examples of drilling wells in trouble. If they took a kick I'm not sure they can circulate it out and pump a kill pill: no drill pipe in the hole. The initial report was that NG was shooting out of the well head. Now folks report bubbles coming up around the conductor pipe: I've seen such bubbles from wells that weren't having control problems.

I'm having difficulty search the web so I'll continue to depend on TOD for info. I'm starting to smell a rat...maybe. This has been going on way too long to not have clear set of facts out to the public.

Also: methane has a lower density than air. But all the other hydrocarbon gases are heavier than air.

Rockman,

Here is the latest from the wires,

London (CNN) -- Energy giant Total has located the source of a gas leak on an offshore oil platform in the North Sea, the company told CNN on Thursday. The leak is not underwater, but is on the deck level of the well head platform, Total said.

http://edition.cnn.com/2012/03/29/business/north-sea-gas-leak/

An earlier report stated they had pressure on an annulus (most likely 10 3/4 x 13 3/8)which they were bleeding off, mud gave way to gas, and that is all they said, apart from bubbling water, but from the latest report the leak is only on surface, so go figure.

So to me bleed pressure, hit gas, something breaks, or they run (I hope not)but why not just shut the valve you opened to bleed with.

We will have to wait for the next installment.

I hope you enjoy Ghana more than EG

Thanks pusher...makes more sense. But if they were spewing 12 mmcfpd at the well head and the flare didn't ignite it...someone got very lucky. Have to wait for the post mortem but if the valves didn't malfunction then maybe they just ran. As you know takes a heap of nerve stand by when you hear the gas screaming in. About 6 years ago I was 20 miles from a rig when it kicked and they had to shut it in manually. Company man sucked in up and ran for the wheel to shut it in. They never recovered his body. We all like to think we know what we would do. But you can never be sure. Long ago I heard of one driller who held his ground and saved the rig/hands. He worked 6 more months and quit for good. Said he lost his nerve and couldn't handle normal ops let alone he thought of an emergency situation.

Rockman,

We had a well leaking from one of the annulus recently, and were lubricating and bleeding it to bring it under control, but it was all done under small bore pipe and very close control. I can't imagine opening the 2 1/16 annulus valves full bore to bleed pressure. Surely smaller bore gear would have been used to keep full control. But as the gas evacuated the annulus it is quite possible that something gave way below, as the well head valves should have all been 20k gear.

CNBC USA actually mentioned something about it tonight/morning, due to the 7% drop in Totals share price.

They seem to be thinking it will decline in pressure and flow quickly as it is not hooked up to the main formation. So we will wait and see.

First comment, sorry if noise rather than signal. There's a remarkable satellite image making the rounds of rare, almost cloud-free coverage of the United Kingdom: here. This was taken on the 26th of March.

When I superimpose the platform position map (posted above), I couldn't help noticing the cloud formation and what looks like a sort of huge 'contrail'. Any thoughts?

Here's the two images superimposed:

Again, apologies if this is meaningless coincidence / noise.

Hey! You're horning in on my act!

Here is the best I can do (click for larger version) :

Curious. If there weren't any other similar features in the sky, then perhaps. Of course, the stuff leaking out is mostly transparent. So, what is being seen in the reports of "a cloud" around the platform"? Steam generated by hot gas contacting the water?

HazardEx posted this picture, taken on March 27:

Whatever it is, it seems more like an aerosol.

[edit] I yanked this image, because I think it is archive of perhaps Macando.

Whatever it is, TinEye image search can find no previous occurrence on the web of http://www.bellona.org/imagearchive/elgin_flar.jpg

The wind has been consistently blowing from the north, so I would any trail to be pointing to the south. I think you are seeing a random high altitude contrail.

Not according to Bellona:

Bellona President Frederic Hauge, however, noted that the prevailing winds, which are coming from the west, have kept the flare from developing into an explosion.

The wellhead platform is east of the PUQ, where the flare is.

North sea wind graphic ; may bear watching.

I say North, they say west, the map shows north west. We can't all be right ;)

The linked graphic is actually wave direction; probably a close correlation. more here:

The Oil Flare burns for fifth day at stricken North Sea rig

"The situation is the same as yesterday, the wind direction has not changed. It is a westerly wind and there has been no change in the situation," a Total spokeswoman told AFP.

Total said in a statement Wednesday that it was "perfectly normal" for the flare to be still lit after being "de-pressurised in an emergency".

"At present the flare does not pose any immediate risk as the layout is designed to take into account the prevailing wind direction, ensuring that these winds are taking any gas from a potential leak in the wellhead area in the opposite direction to the flare," the statement said.

"This is in fact exactly what is happening. The wind is forecast to remain in its current direction for the coming days."

I've checked the surface wind stream from GFS and yes the wind had mainly been from the west. Rotated a bit now and the wind is currently blowing from the NW to SE and is forecast to be North to South on Saturday.

On the 26th, when the sat image was taken, the surface wind was lightly blowing approximately from west to east.

Edit: Here's the UK Met Office Hi Res NAE model of surface wind at mid-day Monday 26th

joules - Probably not steam. Though the NG may be very hot in the reservoir as it rises the pressure drops significantly. Just the opposite usually happens: the NG causes freezing. On a producing NG well we actually use line heaters to keep the flow lines from freezing up. I've never heard of it happening but the freezing NG might be causing condensation vapor/ice crystals in the air.

With no hands on board I don't see how the situation can esolve itself even if all the equipment were functional. I'm guessing they're hoping the flow kills itself so they can reman the rig and kill it permanently. A relief well may be the only option if that doens't happen. But I'm not certain they know where the NG is coming from.

Rockman,

Looks like they are calling for your volunteers,

http://english.capital.gr/News.asp?id=1457369

Firefighters and engineers from the Houston-based firm are experts at managing disasters like oil rig explosions and are popularly known as "Hellfighters," thanks to a Hollywood film of the same name. Once aboard, it is expected they will secure the site and assess how best to close off the damaged well thousands of meters below the sea floor believed to be responsible for the ongoing natural gas and condensate leak.

Total is "putting in place the plans to allow crews -- outside specialists from Wild Well Control accompanied by Total personnel -- to safely go on the platform in the next couple of days," said company spokesman Andrew Hogg.

If they find a connection on the wellhead to latch onto, lugging around 20K Chiksans in full BA will not be an easy job.

I wonder who of the Total personnel will be so lucky, sounds a little risky for your their own people, that is what they employ contractors for!

I wish them well

Edit:

Here is a current photo of the wellhead platform and the leaking gas. Maybe it is the wind, but the flow of the gas looks very directional, therefore leaking from a single point.

http://au.news.yahoo.com/thewest/a/-/world/13314602/fears-remain-as-flar...

Can some pull the picture on this article and post it please. My HTML skills don't stretch that far. Thanks.

to post most photos you find on the the web replace my square brackets with the arrow brackets on the 'comma' and 'period' keys
[ing src=url]

this photo's url is http://l.yimg.com/fv/xp/afp/20120401/05/992736064.jpg?x=400&sig=t5D2QPyG...

That is found by right clicking the photo. Then just highlight and copy the address(url). That is how you do it on a PC a Mac has no right click so you'd be on your own there. Always hit the 'preview' button before you hit 'save' on your comment. If the photo is too big you don't post it unless you size it down-that will be another lesson.

Thanks Luke,

For the post and the lesson.

Give a man a fish, you feed him for a day
Teach a man how to fish, and feed him for a life time

or words to that effect.

Thanks again

unfortunately I had a typo in my lesson-sorry

the brackets should read
[img src=url]

didn't mean to send you off to a dead zone

as for sizing down an image

[img src=url width#,height#]

the spaces must be there-a single space between img and src and single space after the photo's url-the numbers shouldn't exceed width600,height400 or so on this site, smaller is fine if it works. My simple way of keeping the image proportional is to use a stick ruler to measure the image on my screen and then divide height by width to get the ratio then just mulitply that result times 600 or whatever you decide to use for width and insert it after height no space between that number and the final arrow bracket. Hope I've no typos in this try and that I haven't totally muddled things for you with this last part. Try it out with the preview it will work or it won't

If the original image is large you should create a smaller image (50-100 kB) before posting it. (You can use MS Paint or any graphics package.) You put the small image on flickr or somewhere like that. Give a link to the full-sized image for those who want it.

Resizing with width=whatever in the html downloads the full picture but displays it resized, i.e. you have not saved any bandwidth.

good point, generally the stuff I grab that is already on a web is 100 kB, give or take, in size. The very non techy way I reduce a file size is by emailing myself the larger image. Generally the medium size option that comes up for the email attachment is small enough (under 130 kB) and sufficiently detailed. Then I just drag the image file attachment from the email (still unsent) to the upload panel (I use ImageShack) and grab the 'direct link' they send back and paste that both in my email (so I have handy access if I want to post it again) and as the url in the html code in my post reply box.

I use that step on pdf files as well after I've captured the screen displaying the pdf image (on my keyboard with the 'print screen' key on the top row next to F12) and paste the image into and editor (I use Paint.NET, a free download, simple and plenty powerful enough for my purposes). Once I've cropped the image getting rid of superfluous border material captured by the 'print screen' command I save it to a dedicated file in my desktop 'briefcase' from which it is easily emailed and I carry out the step in the paragraph above.

Messing with pdf images does require several minutes and additional settup time right off if you don't have a decent image editor or an online image storage account but often the very best charts and diagrams are only available in that format until you snip them out and load them into your own online image storage account--I created a one dedicated to a username similar to the one I use here. Just passing on what others at TOD have taught me with a couple tricks added in that have made the process very simple for a not so very computer literate guy like me. There are certainly other and probably better ways to do what I have described.

It is also a good idea to increase the compression of the posted image. You can often reduce the file size by 1/2 or more with little image degradation. The full size, linked image will still have all the detail.

NAOM

Here's an image taken the following day, 27th March. Click image for High resolution version

Hi res direct link http://img213.imageshack.us/img213/5154/1340n.jpg

"Energy giant Total has located the source of a gas leak on an offshore oil platform in the North Sea, the company told CNN on Thursday. The leak is not underwater, but is on the deck level of the well head platform, Total said."

http://edition.cnn.com/2012/03/29/business/north-sea-gas-leak/

So if the leak is on the deck as reported now, what is the explanation for the accounts of the large volume of gas bubling up from the sea under the platform?

Smells of misinformation to me. I am sure that once again, TOD will have the straight story soon enough.

The link from joules below gives the clearest picture I’ve seen yet:

“An immediate concern for the company is whether it should, or can, extinguish the flare burning off excess gas... Experts say that if the wind changes, the gas...might be ignited by the flare...”. Should or can? Not consistent with the following statement.

“... the flare was on a separate platform from the leak...The flare is still burning but is not posing a risk. The leak is on the wellhead platform and the flare is...300 feet (away)... and it was not possible to shut it down remotely.”

Not a risk? A tad inconsistent with their other statements. Shell appears to have a differing opinion of the risk: “Shell has moved 120 non-essential staff from the...drilling rig, about four miles from the Elgin.”

http://au.news.yahoo.com/thewest/a/-/world/13314602/fears-remain-as-flar...

From this link above, it apears the flare has gone out. I guess that the speculations that this was condensates evaporating and burning off was right.

Total still evaluating possible action plan for North Sea gas leak

He said that some weeks ago Total engineers had decided to pump mud into redundant piping on a gas reservoir which had been plugged about a year ago. This might have caused the escape of gas from the outer casing of the well.

Can someone explain "redundant piping"?

joules - In 36 years I've never seen the term redundant piping used to decribe any aspect of a well. It maybe be the translation from French. I'll take a guess and say the are talking about other wells in the field that penetrated to depleted resevoir but had never produced from that zone. But not sure why the would pump heavy mud into those wells. The risk wouldn't be the reservoir flowing but the mud being lost ino it if they had a bad cement job. Maybe they are talking about pumping heavy kill pills into the annuli of those wells to compensate for potential poor cement isolation. Maybe they discovered some annuli had pressures that should not be there. In the federal offshore and in most US states you're not allowed to "have pressure on the backside": annular pressure. Not sure what the N Sea regs are but I would suspect the same if not tougher.

Remember that this was the first production well there (though a dozen were put in at once, before production). Compaction, sand infiltration, liner deformation. So, they plug up the middle with mud (a year ago?). Then something happens a few weeks ago, and they decide to put more mud in somewhere. But where? And this caused the problem? Or, perhaps it just didn't fix it.

this WSJ article mentions some of the problems pressure drops due to depletion has been causing in the high pressure/high temperature fields. It's a couple days old don't know if you saw a similar one. Apparently rock shifts associated with reservoir pressure changes are playing hell with the casings.

The steel casing that lines a well can be buckled or deformed by movements in the rock, Mr. Bergerot said. "Eventually, the liner may become sheared off completely," he said.

Most well liners on the Elgin and Franklin fields have been affected in some way by this process, Mr. Bergerot said.

Total has said that in the current incident, gas is probably entering the well through a leak in its casing.

and a bit farther down

It was a surge of gas and condensate from the well onto the Elgin platform that forced the evacuation Sunday.

David Hainsworth, health, safety and environment manager at Total U.K., said the company has been actively managing these issues for a long time. "What we've been doing over the past two or three years is plugging and abandoning those old wells," which have had their casings compromised, he said.

The well that is now leaking was shut down 12 months ago on the suspicion that it had failed, "but we hadn't got round to the final plugging and abandonment exercise; we'd just been monitoring the well," Mr. Hainsworth said. There is nothing to suggest this job was done inadequately prior to the leak, he said.

Luke - This problem doesn’t occur with just HT/HP wells. Casing collapse can be caused by “point loading” in a normal pressure water drive reservoir. And hydrocarbons can leak into an annulus for a variety of other reasons. I’ll make a very opinionated statement and claim that all their focus on the HP/HT aspect and the casing collapse (as almost an act of God) is to distract folks from asking why they didn’t have adequate cement in the leaking annulus and exactly how the NG came to escape. The HP/HT and the casing collapse is why there was high pressure NG in the annulus. But those weren’t the cause of the blow out. The blow out was caused by the uncontrolled flow of NG out of the wellhead. That’s the critical question IMHO and not how the NG got into the annulus. I’ve probably had annular pressure (“pressure on the back side”) at least 30 to 40 times in the last 36 years and none of them blew out. Again maybe the translation from French isn't reflecting what they really mean. Consider:

"but we hadn't got round to the final plugging and abandonment exercise; we'd just been monitoring the well…There is nothing to suggest this job was done inadequately prior to the leak..” OK…they said they hadn’t done the P&A job yet but the job (that they haven’t done yet) as done adequately.

Again, pure speculation on my part. But what isn’t speculation of how you kill pressure on the back side. You hook up flow lines to the proper fittings on the wellhead that allow you to pump a heavy kill fluid into the annulus. Once you’ve killed the flow you pump cement and permanently seal the annulus. My WAG: the valves allowing access to the pressured up annulus failed. Maybe a seal failed. Maybe they were trying to pump in the kill fluid and the valve broke. And maybe someone accidently turned the wrong wheel and opened the annulus to the atmosphere and they ran away without closing the valve. Yes…I’ve seen that done. Millions of cu ft of explosive NG blowing in your face can easily cause such a reaction. Total may not know exactly how the pressured NG got into the annulus. But they do know exactly what their hands were doing in the minutes before the well blew out. And they haven’t felt a need to share that info yet.

Thanks for the comeback ROCK, just came in from shovelling about an inch of that stuff you called 'magical' off a thousand foot or so of my driveway-probably never have seen it in Ghana ?-) Not bad duty, beats paying to go to a smelly health club.

Anyway, Hainsworth was blowing what smoke he could from what I could tell, but my interpretation to

There is nothing to suggest this job was done inadequately prior to the leak

was that the initial work on the producing well was fine until the the earth moved and cracked the casing.

But his statement that they hadn't got round to the final plugging and abandonment exercise yet certainly contradicts This NY Times article of a couple days back

The leaking well on the Elgin platform was no longer producing gas, according to a statement by the Department of Energy and Climate Change, and workers carried out an operation on Sunday to plug and abandon it. Such work usually involves injecting cement into a well at several locations, including the top, to prevent any gas remaining in the reservoir from leaking.

After the work was done, remote monitoring of the well then revealed that gas continued to be released, according to the statement.

Which paraphrases Britain's DECC ministry statement

The incident occurred during work to plug and abandon the well, which was no longer producing gas. Total has now shut in all wells but remote monitoring has revealed that gas continues to be released.

Well maybe technically they hadn't gotten round to the final plugging as the damned thing blew out before they could finish the job--but that is a heck a way to say that. I'd almost have to say Hainsworth was flat out lying in the the next phrase we'd just been monitoring the well but then there is just enough squirm room in his usage of 'just' (when taken in the temporal sense) for him to wiggle loose-he seems oily enough to slide out of a pretty tight spot. I'm guessing no translation from French was needed as Hainsworth is safety and environment manager at Total U.K. Those statements were very, very, very likely made in English originally.

Also very, very, very likely that all the hands actually involved in the incident have been told in no uncertain terms to keep their mouths shut. Thanks for putting you WAGs out here, I figured you'd been in similar spots more than a few times. Real world eyes are always appreciated. No doubt, getting resources from mother always has a hairy side.

Not just HP/HT, but Geo pressured. Gas in the pore space supported the overburden, release the pressure and the pore space collapses, resulting in 'movement'......duh !

The blowout was probably caused by failure to consider the consequences of this movement. How in the world is a couple inch sheath of cement going to prevent casing collapse, if indeed that is what happened.

The solution, in my opinion, would have been to maintain pressure in the reservoir, with CO2 or N2 for example.

Apparently, Total failed to properly address the geo-pressured problem, end of story. Then there is the geo-pressured Hod formation above. Haphazard development in the hubs of hades ?

Solvent injection had/has potential to recover condensate from this rich gas condensate reservoir. Inject enough solvent and the condensate will come out. Evidentally, solvent injection was rejected for 'economic' or technical reasons.

I'm interested in the marine environmental implications of the leak (assuming it is bubbling up from the sea bed): anyone out there got any idea of the chemical compounds in both gas and condensate?

Am I the only one who is getting flashbacks to the Gulf experience?
This time we have a very complex and difficult well, awaiting final plugging, developing unexpected leaks.
Because there are multiple wells and multiple productive regions, and because there may have been leaks at various levels because the well liners became very stressed as the lower level reservoirs were depleted and lost pressure, we do not know which well was the source or from what level the leak originates. Reports of gas leaking around the base of the platform are unconfirmed, but might indicate more leakages at higher levels.
Meanwhile, TOTAL issues reassuring noises, but very little hard information, in a replay of the "mushroom treatment" practiced by BP in the Gulf.
Thankfully, no one has gotten killed thus far, but the idea of a large platform sitting atop a dozen or so possibly leaking wells with no one within a couple of miles is worrysome.
It would help if some of the veterans of Macondo would give us their best guess of the situation and their assessment of possible outcomes.

So somewhere in a pub in Dundee right now is a bunch of unhappy oil men figuring out who is going to draw the short straw.

The luck winner will get kitted out with full scuba breathing gear then plonked into a rubber dingy upwind of the platform. From there he will paddle up to the rig with a gas monitor around his neck and all going well he will climb up the ladder onto the platform and get the work underway.

Since a radio or torch or anything electrical is a no-no, I can only assume he will have a mirror or a flag to use to signal the all clear to his mates huddling behind the biggest boat they can find 2 miles away!!

Good luck Jock, rather you than me!!!

nigwil – Very good. LOL. I actually experienced a similar bar room situation about 20 years ago. A flow line was leaking a relatively small amount of NG and the valve on the wellhead was stuck open. One of the production engineers would have to shut down the production equipment in a proper sequence to prevent an explosion.

I recall two funny things. The hand that was going to do the deed ordered lunch. His wife had been on his butt about losing weight. But he said “Screw it….give me the chicken fried steak with a double order of mashed potatoes and extra gravy on the side”. After all these years I remember his exactly words as though I heard them yesterday. And when we went out to the parking lot the other engineer asked him for the key to the motel room they shared “just in case”. LOL.

I’ve had similar thoughts to yours. There is very little you can function with the equipment without significant mechanical force. How is that force going to be generated without a potential ignition source? Even a battery assisted tool could pop a spark. Start up the generators on the rig? That sounds like a death wish. Maybe some clever device powered by compressed air? But that would have to be very heavy so how would it be positioned by hand? Similar to Macondo: if the BOP fails there’s little to do other than hope it kills itself or drill relief wells. But unlike Macondo there are multiple wells in the area they would have to intersect the out of control well bore. A complication for sure.

Nigwil

Radios, torches, even cameras are used all the time in potentially explosive atmospheres - they are "ex" type, that is to say sealed so that the potential ignition sources do not come in contact with the gas. Standard on all rigs in North Sea. Problem comes when the batteries are flat as you must have a safe area and power source to recharge. In this situation the intervention team would simply carry sufficient spares so that recharging on site is not necessary. Machinery located in hazardous zones will also be to this standard.
The bigger problem (as Rockman noted above) will be to get some form of power so that the machinery (Pumps, tuggers, etc, and not least all the instrumentation, can be used again. The rig is almost certainly at thier highest level of ESD (Emergency ShutDown) so electrical power to everything will be shut off, and all generators stopped. They will be able to check which areas of the rig are affected by gas and which are clear, but it would be a brave man that restarts a generator under those circumstances - the concern will be that you not only power up the stuff you want, but send power somewhere unexpected. Will need personnel very familiar with the rig systems.

Yeah sure. We've had a coal mind disaster here in Pike River. Gas explosion - and a mine now full of gas.

Problem is getting any gear or personnel through the zone of gas density that is explosive. OK(ish) to work on full recycle respirators in high gas density where risk of explosion is low(er), but how do you get across the boundary of gas density to approach the work site in a way that is not going to briefly illuminate eastern Scotland?

If this piece of information is correct, it may not be simply bull heading a bit of heavy mud down the annulus.

http://www.bellona.org/articles/articles_2012/elgin_flareout

Is Total withholding information from workers?

Molloy has slammed Total over the last 24 hours for withholding information from workers who were sent to Elgin to deal with the instabilities that resulted in the leak.

He said they repeatedly raised safety concerns about rising gas pressure but were told a leak "could not happen" only hours before one took place.

"The workforce had raised their concerns to the Offshore Installation Manager who in turn consulted with onshore technical authorities at Total...they were repeatedly told that a failure in Annulus C could not happen...and even if it did, a design feature would prevent a gas leak," Molloy said.

"Several discussions between workers and Total technical authorities happened throughout the preceding weeks, up to and including a few hours before the event," he said, referring to the leak that led to the evacuation.

A spokeswoman for Total in Aberdeen reached Friday said she could not immediately comment on the matter.

http://total.com/MEDIAS/MEDIAS_INFOS/2155/EN/elgin-franklin-VA.pdf
Page 11, has a graphic of the typical well
A Annulus tubing to 10 3/4 - 9 7/8 casing
B Annulus 10 3/4 -9 5/8 casing to 13 3/8 casing
C Annulus 13 3/8 to 20in casing

Now my previous assumption was a leak in the B annulus, 10 3/4-9 7/8 to 13 3/8 casing set at 3,600m, giving a reasonably deep set shoe. This new information if correct means the leak is in the 13 3/8 to 20in annulus, with a shoe at 900m. This shoe depth will not be able to withstand 12,000psi (800 bar) pressure of the Hod formation, potentially leading to a sub sea blow out,ie gas bubbling up through the sea bed.

So what are the options,
1/ perforate the 9 7/8 casing at 4000m or deeper, and circulate mud down the casing and squeeze the mud up the annulus in a dynamic kill. Problem, what machinery do you use over a actively leaking well. Quick and risky
2/ drill a relief well and dynamic kill. Slow and safer, as long as the well doesn't wash out too much.

A bit of a Hobson's Choice. But this situation is dependent the "C annulus" information being correct. Plenty of miss information around.

Gas is seeping into the annulus, or pipe casing, of the G-4 well, which was plugged in February 2011

pusher - Good catch. Caught my eye: "Gas is seeping into the annulus, or pipe casing, of the G-4 well, which was plugged in February 2011". Plugged to what degree? You know there are several steps in a P&A. How many cmt plugs did they set and at what depths? Etc, etc, etc. So let’s make some more assumptions. They had completely plugged the tubing/production casing. Now nothing left to do but pull the wellhead off. Can't do that with pressure in any annulus. Perhaps what they're indicating the well had pressure on the backside for at least a year. Big question: why wait so long to deal with the situation? Perhaps not an important question compared to why they just didn't kill the annulus when they finally got to this point?

Now more guessing. You know the equipment better than me. Given that I've heard N. Sea safety regs are tough I would think the wellhead was designed to deal with such potential situations. So what went wrong? The wellhead didn't receive proper maintenance all those years? Was it not installed properly? Yes...that can happen: I was saw a drill rig with part of the BOP installed upside down. How many safety valves have you seen installed backwards?

They apparently sat there for a while trying to deal with the situation. Killing the backside shouldn’t have taken more than a couple of days. Something was working. You know better than me what happens when you’re burning rig time and not making progress. The message comes in from the bank: “What are you *ssholes doing out there? Make it work NOW!” And what do hands tend to do? Correcto: jerry rig it. And we both know how that turns out sometimes.

Rockman,

On the same doc as above which I didn't notice before. Page 12,

http://total.com/MEDIAS/MEDIAS_INFOS/2155/EN/elgin-franklin-VA.pdf

"C" annulus Synthetic based mud + cement Max. allowable pressure = 20 bar.
"B" annulus Synthetic based mud + cement Max. allowable pressure = 250 bar.
"A" annulus Inhibited fresh water + N2 gas cap. Max. allowable pressure = 860 bar.

20 bar or 300psi will not give them much to play with. I suspect they were attempting a lubricate and bleed on the C annulus, to try and lower the pressure, but if they were up against 800 bar, they were never going to win the battle, as 17.5ppg mud at 900m = 2600psi, and Total state 20bar or 300psi as the max allowable.

It is going to be an interesting kill, however they pump the mud!

See

http://www.elgin.total.com/elgin/PressRelease.aspx?contentid=653

Downloads available are a text explanation of what Total think has happened and a nice well schematic.

In summary, failure of casing in Hod formation (non-producing Chalk reservoir) giving access to A annulus. Progressive leakage to outer annuli and following pressure observations 25th February, decided to kill annuli and P&A well. Started on 4th March but on 25th March, during kill operation with 2.05sg mud, C annulus failed and gas is leaking from 30" conductor. Reported wellhead path is via D annulus.

Released in a press conference 30th March but the weekend media seem blissfully unaware. Not hard to find via Google though.

Quaking,

Thanks for the link.

http://www.elgin.total.com/elgin/content/documents/elgin_gas_leak_press_...

On 25th Mar we believe the C annulus failed.

In other words the 20in casing burst. Fun and games.

Edit, They seam to be saying by reading the diagram, that the 9 7/8 casing failed, the 13 3/8 casing failed, and now the 20in casing has failed.

The 30in casing may be 1 in wall but it is for structure, not pressure.

The 30in casing may be 1 in wall but it is for structure, not pressure.

Does it mean that if they close the valve on the rig the gas will blow out on the bottom of the sea?

Are failures this high up in the well not unusual?
Separately, if the well is compromised to that extent, is not the least dangerous solution to simply flare the gas for as long as necessary?

e - Annular failure should be relatively rare up shallow. Just to be sure everyone understands why we pump cement into the wells: it's not to keep the casing inplace. The casing isn't going to move. In fact it's typically damn hard to impossible to pull it out. The primary purpose of cement is to seal off the annular spaces. Every annulus is a potential conduit for any hydrocarbons that are present. And we don't just pump any cement. It's formulated to withstand a specific pressure. And then we pressure up on it after it sets up. And test to a level of the max pressure that could be encountered.

Here's the problem with the shallow annular sections: often high pressures are not anticipated. The deeper annular spaces are cemented and tested to hold those high deeper pressures. Thus the shallow annular spaces either have no cement or relatively low strength cmt. As pusher points out above the shallow casing strings typically aren't strong enough to handle high pressures. The cement in the deeper annular spaces keeps those high pressures from reaching the shallow intervals.

Unless something breaks in the production tubing that allows the high pressures to bypass the isolation cements. That appears to be what went wrong for Total. But there are relatively simple measures that can be taken WHEN THE WELL IS ORIGINALLY DRILLED. Just my WAG but I suspect Total had not known about this potential problem initially. Once the well was completed and producing the problem might have become apparent. The problem could still be fixed but now it would cost a great deal more. Not surprising that management didn’t jump on the problem earlier if that was the case. Easy to wait to be promoted to another position or retire and let it be someone else’s problem.

We’ll just have to wait to see how much details Total eventually releases.

Thanks for the link, maybe you can elaborate a little on this picture for us duffers

To my inexperienced eye it looks as if the annulus just inside the 30" conductor is at the edge of the well head control hardware. What is the nature of the seal at that point? You said the 30" casing was for structure not pressure does that mean once the 20" casing was compromised there was essentially no designed pressure containment at the wellhead for the annulus inside the 30" casing?

Luke,

Thanks again for your handy work, I did try to grab the picture, but I don't have a snipe tool, so I gave up due to the PDF.

Now to answer your question, the 30" conductor is normally a pure structural joint ie, it is a structure to provide support to the wellhead and BOP. In many cases there is no gas seal in the "D" annulus (30" to 20" annulus), several platforms I have worked on have actually had holes cut in the 30" casing for various reasons. The only time it would be absolutely required to be fully sealed is if shallow gas is suspected, though in most cases it does have a seal but it is not high pressure.

30" conductor can be installed 3 ways to my knowledge,
1/ Drill 36" hole run casing cement
2/ Pile drive with hammer, most likely
3/ Jet in place by drilling an undersized hole followed directly by the casing, and let it get stuck, This method is mainly used in deep water.

My point being, most likely this 30" conductor has just been pushed into the ground to a depth of 223m, but never pressure tested, so any back pressure in the D annulus will result in the gas/pressure leaking off at the shoe of the 30" and potentially bubbling out of the sea bed. In a lot of ways it is better off blowing away in the well bay, until they can come up with a fix.

I hope this helps

Thank you, toolpush and rockman, for adding expert understanding to this evolving problem.

Two questions:
1 Is it practical to flare 200,000 cubic meters/day of gas and if so, can that be set up reasonably safely under the current conditions? Would that allow the platform to return to some operation?
2 How can the well be killed with a bottom injection of mud if the well is compromised all the way to the top? There would be very little mud weight above the leak initially, so would not the mud be simply blown from the hole as it gets pumped in?

If the situation is as suggested, what are the precedents and what examples should be studied to get more insight into the potential options for TOTAL?

E – That much NG and more could be safely flared. In the field I drilled in off the coast of Equatorial Guinea they flared many times that volume. But here’s the trick part to that answer: the NG has to be piped to the flare stack. If I understand correctly the NG is flowing out of the wellhead perhaps through a broken seal/valve. Don’t know enough details about the wellhead but I doubt the piping could be set up. The source of the NG originally being flared is still unclear to me.

You’re correct. I doubt a static column of mud with sufficient weight and height could be put in place to kill the well. Especially true if the well is flowing. That was the problem they had with the top kill on Macondo: can’t build head when the NG is blowing the mud back into your face. But if you can stop the flow you can pump cement at high pressure and force it into the annulus. But the NG would cut the cmt and make it nearly possible to harden properly. OTOH remember the premise: you have to stop the well from flowing before you pump anything down. If they can shut the wellhead in then they just solved the biggest problem: keeping the well from blowing up. Then they would have time to drill the relief well.

The relief well??? Given the different thoughts on which annulus (or how many) is flowing NG I don’t want to start contemplating how difficult the RW will be. IMHO we’ll have plenty of time to beat that topic to death if the well doesn’t kill itself anytime soon.

Options? Just my WAG: just the relief well.

http://books.google.com.au/books?id=OZnGJROkPG8C&lpg=PA94&ots=rTYVzCtj6h...

Offshore blowouts: causes and control
By Per Holand

search for "Casing Leakage" Page 94

Ruptured casing is not a very common event though casing casing can be worn through by the rotating drill pipe in a high angle dogleg, but even less common when it is combined with a blow out, because these are the things we try very hard to avoid.

It is hard to believe what Total have described on their release, of multiply casing failures, and the fact there has been so little other comment. I beginning to wonder if I am reading too much into the situation.

But to answer you question to flaring. It is hard to see where it would be possible to be able to capture the gas and direct it to a flare via a direct connection. Maybe some sort of hot tap could be employed on the 30" conductor to connect up to a flare line, but that sort of equipment will come with the well control experts.

The platform will not return to any normal function until this well is under control.

"There would be very little mud weight above the leak initially, so would not the mud be simply blown from the hole as it gets pumped in?"

We do what is called a dynamic kill, where you pump the mud as fast as you can/formation can handle and this allows the friction of the mud rising to exert extra back pressure on the formation.
Now when this was done in Australia a couple of years ago, the extra velocity of the of the gas induced by the pumped mud, created static and you guessed it, the rig burnt down. So at no stage are you out of danger until the well is dead.

Now how they get 17.5 ppg mud back to surface with no sound casing in the hole, I will leave up to Total and their merry men. I am sure they will be fully entertained for quite a while. A lot will depend what depth the ruptures are at, and if the mud flow is in communication with the formation or not.

Thanks, that was what I expected, but that leads me to the next question. How does gas leaking from the 20" casing communicate with the 'D' annulus?

If I understand correctly the 30" casing is essentially a piling driven at most a couple hundred meters into the seafloor, which, according to the diagram, is thousands of meters above the chalk formation that is producing the leaking gas. Is that gas following the drilled hole in the rock the 20" casing was punched through and just jetting a larger path through those thousands of meters of rock in the process? If that is the case it would seem every conductor out there in this type formation best be getting a 'D' type annulus cement job pronto.

But then how do you cement inside a piling driven into the seafloor? The cement would likely have to be mostly in the section where pipes are passing through the bottom of the water column. That would allow pressure build up in the section of the 30" conductor just below seafloor, which is often very soft mud if understand correctly. I could be way out in left field here--but I think I have a basic grasp of the principles behind plumbing and mechanical systems.

I guess another option would be to require a real seal at the top of the conductor but also to have a sufficiently sized outlet to a flare stack just below that point--all right from the get go. More money and time but if these high pressure/high temperature formations are the bear they seem to be a bit more precautionary engineering would seem fitting. What's it cost to have this platform and all nearby operations down for at the very least weeks on end? That's not mentioning the possible human life costs, which is immeasurable to those (including family and friends) at risk but is quite quantifiable as increased insurance premiums to the corporations involved. (I've been at the write off end of that stick.)

note: I added another post up the page which went into posting pdf images. The material contained in those files can make it worth the effort for those who have the time--which in this case I had, glad to be of some service. You and ROCK bring plenty to the table here that can't be learned messing on a laptop for half an hour, thank you both.

http://www.elgin.total.com/elgin/content/documents/elgin_gas_leak_press_...

This link PDF doesn't seem to exists any more, the first article from Total with some actual details of what happened, now it vanishes. Maybe too much detail? people may actually be able to work out what is going and realize what a cluster, these gents have got themselves into.

It is possible it contained wrong information, but then surely they would post a rebuttal, rather than just vanish the PDF.

Hi toolpush I think you have the wrong URL for me is working:

http://www.elgin.total.com/elgin/content/documents/elgin_gas_leak_press_...

Edit to say: I have it saved as well

Thanks Gel,

My link is working now, I think Total must have been having trouble with their site, as I was getting 404's.

So all is good, but I also saved the PDF as well. A lot of handy info in that file.

Total to get go-ahead to kill leaking North Sea gas well

Along with the good flare/bad flare tango, we have

But the discovery that the source is just 1.5 kilometres away from the main gas reservoir, one of the world's deepest, suggested that the amount of gas feeding the leak may be larger than first thought.

I thought they said the gas had very low H2S, indicating that it was not coming from the producing reservoir.

A senior union official said on Friday that Total had repeatedly assured workers a leak was impossible until just hours before evacuating them.

The second boxed quote sounds eerily familar to what the BP man said prior to Macondo's blow-out.

Karl - Sadly in my personal experience such statements are not uncommon. To be fair it's typically a judgment call. Of course that's a function of one's experience as well as pressure from management to move forward as fast as possible. But it's also not uncommon for such judgments to be challenged on the spot. Recall the response of one of the Macondo hands after his concerns were dismissed: "Well...guess that's why we have a BOP". I can promise you that wasn't a spur of the moment adlib. I've heard that expression more times that I like to remember. I've used it and was almost run off the job for it. And it's always done in front of witnesses. Basically saying: "If it goes bad I'm not going to cover for your stupid *ss". It sounds like the union rep made such arguments. If so and matters turn very bad out there, such as an explosion and/or death I suspect we'll see more comments from the hands out there.

Thanks for your honest reply, Rockman. In your opinion, how long will a shut-in well, one whose production is deemed insufficient/over, on the ocean floor remain shut?

k - If a well is properly designed and tested it can stay shut in safely for decades. Much longer than you would likely ever leave a well shut in. But I wonder if you meant to ask how long a plugged and abandoned well would not eventually leak back into the evironment. If a well were P&A as per the specs of the Texas Rail Road Commission it will likely not ever fail IMHO.

Did that give you what you were looking for?

Is that still usually the case in the event of subsidence along the well?

It seems TOTAL has noted that subsidence is likely in its reports on the Elgin field development.
Could that also help explain the multiple seal failures this well appears to have experienced?

Consider the challenges involved in plugging a large leak from a damaged well in a set of about a dozen, drilled deep in subsiding ground, with high pressure producing zones at 4000 and 5500 meters. Are there comparable situations that have been dealt with successfully?
Separately, if subsidence continues, is the relief well not likewise at risk?

This seems an unusually difficult problem, even disregarding the gas leak.
So is the first order of the day to capture and flare off the leak? Is that why nigwil mentions the idea of a top hat?

E – The compaction effect could cause casing failure. The pressure at those depths are great. Shifting the stress vectors can produce forces that can easily break casing strings. Might only require a movement of a few inches.

The greatest concern with the offset wells is near the surface…within a couple of thousand feet. The wells have been directionally drilled so at greater depths there’s significant separation. But within the first few 100’ the wells are only tens of feet apart.

Yes…as long as the NG is flowing out the wellhead it’s difficult to imagine conducting any ops on the platform. The problem with the top hat approach is that the NG isn’t shooting out 5,000’ below sea level. I can’t imagine any system that could be lowered over the wellhead that wouldn’t allow even a very small amount of NG to leak in the immediate vicinity of the wellhead. So the risk of explosion remains. Just the process of moving any piece of equipment on the platform carries a high risk IMHO. As someone else mentioned just a discharge of static electricity would destroy the platform and kill ever hand onboard. It might sound odd but if the NG had ignited and was burning now the process of capping the well would be safer. Dang difficult but safer.

Thanks, rockman, for another enlightening input.

Logic would indicate that if the surface platform is subject to explosion risk, the work must be done underwater.
Is there much experience doing a 'top hat' like procedure (swapping out the well header at the ocean bottom) using free swimming vehicles?
Short of suspending the equipment underwater from floatation devices, which has not been done anywhere afaik, is there any way to handle heavy gear without some brave operator on the surface directly above it? Maybe some sort of long cable between two points on the surface, with the device hanging from the middle of the cable?
It seems TOTAL will be pioneering new techniques again at Elgin.

e - And therein falls the problem" there is no wellhead at the bottom of the sea. This isn't Macondo. The wellhead and al the control systems are sitting up on top of the platform. There is no access to the well below water. The NG is flowing directly into the atmosphere on the rig floor. And mixing with the air. As I described before this condition generates the potential for the most powerful non-nuclear explosion ever known. If this were to happen it could complete;y shatter the platform and expose all the other wells to blow out potential. I know the situation may seem benign: no oil cover birds, no towers of flames, no charred bodies. But the considtions out there are some of the most dangerous I've seen in my career.

Just to be clear,

Are there no shutoff mechanisms for the wells below the sea surface? That is, if the well platform were to go boom, all the wells could conceivably be opened to the atmosphere?

Beyond just the current situation, I could imagine some other maritime incident, such as a large tanker or whathaveyou, ramming the thing.

JB

Joules,

All the producing wells have a SSSV, Sub surface Safety Valve, installed in the tubing string. They are fail safe closed, and require 6000psi to stay open. These have been in North Sea wells since the beginning from my knowledge. Note the Piper A, once the fire was out none of the wells leaked. Compared to Kuwait, where the well heads were blown up and they all leaked. I understand that the new Kuwait wells had SSSV installed.

The D annulus is never suppose to see pressure, and will have a couple of ball valves at best to isolate the annulus.

The D annulus is never suppose to see pressure, and will have a couple of ball valves at best to isolate the annulus.

Total is a bit circumspect about how high pressure gas is getting into the 'D' annulus.

How deep is the 30" connector likely pushed into the seafloor? The total diagram shows gas leaking from the 'D' annulus at the well head but only shows it getting out as far as the 'C' annulus at the seabed level (I missed those little arrows first look). That would means the 20" casing was cracked above the seafloor somewhere. Their diagram requires lots of cracks and for each successive larger diameter casing to have cracked closer to the surface, I guess anything is possible with straws stuck in shifting rocks--more and better info from total would be helpful, but I do have to wonder just how good of info they have on the failure mode/s.

In many cases there is no gas seal in the "D" annulus (30" to 20" annulus), several platforms I have worked on have actually had holes cut in the 30" casing for various reasons. The only time it would be absolutely required to be fully sealed is if shallow gas is suspected, though in most cases it does have a seal but it is not high pressure.

It certainly does look as if having a sufficiently sized outlet going to a flare stack from the top of the 30" conductor (which would have to be fully sealed at the wellhead except for the outlet) would certainly have reduced the hazard potential. That would be a big and controlled hole which would not allow pressure build up inside the connector. Would that just be extremely difficult to work around? Is this the first time you have ever heard of high pressure gas making it out to the 'D' annulus? I'm pretty sure I'm not the only one asking these questions.

I'm, not so sure we are talking about a "high pressure". I ran a sizing for a relief valve using a set pressure of 5 bar and a flow of 12,000,000 SCFD of methane. It works out to be an "N" orifice valve, about 5 sq inches. So if they broke off a 2 -1/16 inside diameter valve, that would be about right with only a 5 bar differential pressure.

1) Why were they in no hurry to address the problem? Did they assume a 5 bar annulus pressure was no big deal?

2) They say that the well spewed some mud, but there is little visual evidence of it.

3) They want to try a top kill first, rather than drill a relief well.

4) If they broke off a valve, they may be able to drive in a tapered wooden plug into the hole. 5 bar (73 psi) x 5 sq inches = 365 lbf on the plug.

Of course, IF (admittedly a big IF) this makes sense, then we go back to Plan A and drill a tiny pilot hole through the wall of the 30" conductor. With 9 bar outside and 5 bar inside they'll get flow inward. Let the water displace the contents of the conductor from the mud line to the surface and they don't have to send anyone aboard the rig. Just send down a diver with an itty bitty drill, drill a pilot hole, go home and wait for success. Gravity will do the rest.

Very interesting calculation. The 200,000 cubic meter/day numbers certainly would support your estimate of a modest sized hole and modest gas leak pressure.
Still, given the TOTAL indication that the entire well is compromised, with leaks through every annulus, does that not also indicate that the true pressure in the well, if sealed, would rise to equal that of the initial high pressure gas leak? So I am not sure the 5 bar pressure that is currently derived from the leak rate is really representative. Presumably that initial leak is very small, maybe just a fissure in the cement seal, so the larger leak topside keeps the pressure from building up. That would suggest that the leak could grow very considerably quite abruptly, if the plug(s) involved break(s) a bit more.
The really minimal information flow provided raises the concern that this situation is perhaps much more tenuous than the reports to date indicate, which would certainly be consistent with the expert opinions posted by rockman and other knowledgeable drillers on this site.

This reminds me of Macondo. They treat us like mushrooms, keep us in the dark and feed us BS. So we have to try and read the tea leaves.

If they have no one on site and no instrumentation, how do they come up with an estimate of the flow rate? My guess is they know the size of the hole because whoever broke off the valve told them, and they know the shut in pressure (5 bar) before the accident occured. So they solve the same equations I did but using diffent knowns and variables. They know area and pressure to solve for flow. I know pressure and flow to solve for area.

The shut in pressure at the well head (5 bar) would be the maximum before depletion. At one point in time they expressed a hope the formation would become so depleted the flow would stop on its own.

One thing I've noticed is how simplifying assumptions become Gospel Truth in this business. So people tend to think of the gelled mud in the well bore as being impermeable. But the solid rock in the formation is permeable! Imagine that someone did an Opticem simulation of the mud and it showed a HIGH GAS FLOW POTENTIAL!!!!!!! due to channeling of the ceeeee-ment. Now imagine the gelled mud in the well has channeling resulting in a high gas flow potential (12mmscfd). That is how they got the 5 bar shut in pressure in the first place. So you put in some more mud and overbalance the formation and the flow stops. (Just do NOT bullhead it, nice and slow and gentle. Say Ommmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm while you gaze at your LavaLite!)

Drilling a little hole at a depth of almost 300 feet is like drilling a 300 foot deep relief well without all the cumbersome equipment! There is no oxygen, so we don't have to worry about Rockman's fuel-air explosions. We get more than a 5 bar increase in hydrodynamic head. Gravity does all the hard work.

yeah I lost the bus somewhere

Hainsworth said that gas is being released on the Elgin wellhead platform at low pressure – about five bar - and that it is coming from a non-producing reservoir above the Elgin formation.

from the first article linked in the key post.

I'm guessing the Total diagram I posted is just cartoon they spit out to give themselves breathing room in the PR department.

I think that is pretty much what Toolpush said a little up the page

It is hard to believe what Total have described on their release, of multiply casing failures, and the fact there has been so little other comment. I beginning to wonder if I am reading too much into the situation.

Rockman,

Just to add to your reply, The wellhead is actually on the Elgin Wellhead platform (the yellow bit)on top of the 20" casing and the leak must be at the seal between the 20" and the 30" casing.

We know there is a Rowan jack rig (the blue bit) over the platform, but I have not seen anything to say that this drilling rig is over the G4 well, most likely it is, but I have not seen it confirmed. There are 3 other wells on this platform so anything is possible.

The other unknown is whether the Christmas tree is still on the wellhead or whether it has been removed, and the Gorilla V has installed its high pressure riser and BOP. But whether the BOP is installed or not, as the leak is external to the well head then the BOP will have no effect on the leak.

To me if people can work in the area, maybe on the lower deck, a hot tap through the casing to give the gas an alternate flow path, and to lower the pressure in the D annulus, may give the opportunity to seal the current leak path without putting back pressure on the formation. Then depending what/if they have any pipe in the hole, it maybe possible to pump mud down the 10 3/4" casing.

From what Total's diagram showed I looked like all strings of casing bar the 30" has failed/leaked/burst, that your pick. Now this sounds very extreme, and I would like to see more information from Total before getting carried away with that as a conclusion. Facts seem very thin on the ground one week after the event.

I am surprised that none of the North Sea drilling hands have not offered any of their wisdom, or inside information on the matter!

Found the following study on the Buncefield explosion. I am wondering if the rig infrastructure may cause this sort of effect, maybe it was involved with the DWH too. Any thoughts?

Impossible Explosion

NAOM

NAOM,

Great read, now I know why we don't have hedgerows on the rigs, and I though it was because there was no soil, and plants don't like salt water. lol

Though a few trees around would make the place look nice.

Claim it's for carbon offset, just make sure it's firs;) They seem to be pinpointing this as part of the reason for the intensity of the blast at Flixborough and some oil/gas installations with the maze of pipework and superstructure acting as the hedgerow. Struck a chord as I used to pass Buncefield quite a lot, on the road affected and early in the morning. It must have rattled the windows in my old house in the UK. Also I was doing a stint in a chemical works control lab a little while after Flixborough and the report (or maybe summary, long time back) was required reading. Chilling when we were sitting in the middle of a similar potential for a bang.

NAOM

Yes, thanks. The only term I could come up with at the time was shut-in, whereas plugged and abandoned was closer. I took your suggestion and found this great Well Plugging Primer pdf published by TRRC. TRRC says the average well plugging cost is $4,500, which is much less than I assumed, although it is much higher for bay and offshore wells, which is detrimental to the program's fiscal integrety. The history of the program shows the Texas Legislature to be somewhat responsible when it comes to dealing with oil, for yet another negated assumption, although much of that legislation was enacted long ago. And there is more good stuff, particularly the appendix's well diagrams.

So, it would seem that barring a major earth movement or erosional event(s), properly plugged wells will remain plugged, as you stated.

So are we back to images from the GOM where we have a top hat hung from a string from the surface poised nearby. A new flange welded to the outer casing (split flange) and then a robot with a wire saw cutting the existing casings and drill pipes off a metre above the BOP? Swing the top hat over, bolt to the flange and seal. Can men work at that depth?

Of course all surface work (like dangling any strings with top hats) will have to be done offset from the gas leak by quite a way.

A raging gale and a calm sea is an unusual condition in the North Sea, but it may be what to pray for!

nigwel - Yes...potentially similar to the Macondo blow out in the GOM. Except for one little difference: Total will have to conduct ops in an environment containing one of the greatest potential non-nuclear explosions possible. I suspect welding top side is off the able. LOL.

I see they are now talking about a top kill solution before the relief well. If, as is mentioned above, they're now reliant on the 30" casing string is there any liklihood this will work? Will that casing string take the pressure created by the presumably heavy mud solution used in the kill process?

The Elgin/Franklin Project: Developing the Largest High Pressure/High Temperature Fields in the World

Reservoir fluids are characterised as gas condensate with one compartment of the Elgin field particularly being quite rich, containing about 1.7sm3 condensate/ 103sm3 gas (300 bbl/1000 mcf), twice the liquid content of Franklin. The 3-4% CO2 and about 40ppm H2S impose additional metallurgical and processing demands.
The fluid, rock and pressure/temperature conditions yield field parameters for which there is little analogous experience.

High Pressure/High Temperature Production: Completing the Process Efficiently
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=OTC-12118-MS&so...

As discussed in associated papers the combined reserves, gas condensates, are large - about 750MM BOE. Sales gas reserves are about 50 BCM and liquid reserves (including oil and NGL products) are about 430MM bbl. Both fields are rich in liquids, but are sour. The Elgin field is richest, containing about 11 bbl/MCM. Franklin contains about 6.6 bbl/MCM.

New Approach to Production: the Elgin & Franklin Partnership Based Operation

Ultimate recovery from the 2 high pressure and high temperature fields is estimated at 48.5 billion cubic metres of gas with 57.5 million cubic metres of condensate. Reservoirs have an initial pressure topping 1100bar with temperature close to 200°C and will be produced through 12 wells. First production should take place by mid-2000 with peak production capability set respectively at 14.6 million cubic metres of gas and 27,800 cubic metres of condensate per day.

Estimation of Static Bottom Hole Pressure from Well-Head Shut-in Pressure for a Supercritical Fluid in a Depleted HP/HT Reservoir
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=SPE-124578-MS&s...

Fluid segregation mechanisms involve complex thermo-convective phenomenon, in association to gravity, which are strongly related to the temperature gradient in the well bore.
Segregated phase distribution has been observed on some available static pressure gradient surveys, where three gradients were identified. They correspond to three different phases, i.e. gas at the top, condensate in the middle and supercritical gas at the bottom of the well.

Depletion-induced reservoir compaction: Two geomechanical models and their application in the planning of subsidence monitoring

Production-induced depletion in water and hydrocarbon reservoirs leads to deformation, compaction, displacement, and stress change both inside and around the reservoir [1-3]. Potentially serious consequences for production include well damage, permeability reduction, and vertical displacement at the Earth surface (subsidence). Well damage can occur inside the reservoir by e.g. buckling of the casing due to compactioninduced along-well shortening. Wells can also be damaged by compaction-induced fault slip, which is reported to occur mainly above the reservoir [4, 5].

Subsidence at the seafloor reduces the safety-gap between the average sea level and the base of the platform structure. Large compaction in shallow-buried reservoirs is likely to cause inhomogeneous subsidence, controlled by lateral and vertical distribution of reservoir compaction and by overburden geology. For instance, stepped-subsidence patterns (terraces) could form in case of compaction-induced near-seabed normal faulting. These could destabilize the near-surface sediments and damage production installations at the seafloor.

The Hod Geohazard : a unique overpressured interval. Cooperation in it's recognition, evaluation and risk mitigation.

Identification of this Geohazard as a tight, overpressured gas bearing horizon has been important. It was realised that cemented casing was unlikely to hold this gas back during the production lifecycle of the wells. Consequently, casing annulus pressure management systems were installed on the facilities. In addition, increased awareness of Hod gas influx during drilling was included in the planning of subsequent wells in the region.
The Hod Formation consists of chalk, both clean and slightly argillaceous which was deposited in a warm, shallow sea. The unique signature of the Hod Geohazard consists of three clean, thin limestone strata over an interval of c.50m. It is the middle of this limestone which is overpressured and contains hydrocarbons. Despite studies, an adequate theory for this isolated zone of overpressures has not yet been proposed.

Edit:Cut down version - sorry about the length.

Team Prepares to Board Stricken Platform

LONDON—Total SA said Tuesday a specialized team could board the abandoned Elgin North Sea platform as soon as Wednesday evening to begin securing the area in advance of efforts to stem an onboard gas leak.

Total engineers and specialists from Wild Well Control—the company that helped tackle the 2010 Gulf of Mexico oil spill and Kuwait's raging oil fires—will likely board the platform Wednesday evening or Thursday morning. That will mark the first stage of an operation to "kill" the problem well responsible for the gas leak off the Scottish coast.

Once aboard, they will conduct an initial reconnaissance of the vessel before attempting to secure the area so workers can safely begin a "top kill" operation. In a top kill, heavy mud is dropped into the well in the hope that sufficient downward pressure is exerted to stop the flow of gas.

A sudden drop in temperatures in northeast Scotland means that the weather has become a consideration. Heavy winds and rough seas add an extra layer of complexity to what is already a tricky operation. Total said it would delay any return to the platform if conditions worsen to the point where higher waves and more forceful gales jeopardized the safety of the teams.

Total is also proceeding with a separate plan to drill two relief wells to divert the flowing gas, with drilling rigs being moved into position, although the relief wells could take as long as six months to complete. Initial relief-drilling work is expected to begin around April 8, Total has said.

also Total to airlift team to oil rig