Tuesday, June 01, 2010

Drilling relief wells -- how do they know they're there?

As has been noted, pretty much the only proven way of stopping a flowing oil well that you can't close off from the top is by drilling a relief well and killing it from the bottom. So you have a few miles of drill casing going deep into the Earth, and you want to intercept it 4,000 feet below where it enters the surface so you can inject some heavy drilling mud into it, then, once the mud column gets heavy enough to stop the flow of oil, inject concrete into it to plug the oil well once and for all. It sounds like you're trying to hit a needle in a haystack, right? Well, yes... except it's not quite as hard as it looks.

One of the more revolutionary things to happen in the oil industry during the past thirty years is directional drilling. Old oil fields have been re-drilled with directional drilling rigs to send out radial "feeder" wells from the main bore hole to collect oil that was otherwise uncollectable, and oil wells can be steered to the exact location of a reservoir as imaged via geological "thumping" (the use of sonic waves to map the underlying structure of the Earth). This has allowed exploiting reservoirs that previously were considered uneconomical to develop, and extended the lives of other reservoirs that were considered near their ends.

Needless to say this requires knowing where, exactly, the head of the oil well is at given points in time. MWD tools are used to do this. The following measurements are generally taken by the logger at the top of the well and integrated by a computer into a map of where the MWD tool is at a given time: Temperature, bore length (as measured by how much pipe has gone down the hole) X/Y/Z gravitic measurements (pendulums) to determine inclination of the well which, with the bore length, tells you how far down you actually are, and X/Y/Z accelerometer or magnetic field measurements to tell you where on that plane of X feet deep that you're located. The reason for the temperature measurements, by the way, is simple -- these devices are mechanical devices (at least the gravitic and accelerometer devices are) and are temperature sensitive, measurements must take temperature into account. Each tool is unique, and is calibrated by putting it into a temperature controlled chamber, zeroing it out, then moving it to a predetermined angle and direction at a variety of temperatures, then plotting a temperature response curve.

In case you're wondering why I know so much about this, I wrote the software that did the calibration and testing for one of the early directional drilling tools back in the 1980's. But anyhow, while they don't run MWD tools all the time because the tools are expensive and the vibration and temperature tends to kill them over time, they do run them often enough (generally when they have to pull pipe to change the bit, which is a good time to do a survey) so that they have a pretty good idea of where the hole is. But remember, we're talking about needle in the haystack, "pretty good idea" means "within 100 yards" and we're trying to hit a 32" casing in that 100 yards? Sounds like a fool's chore, right? Well, not exactly, because the casing is metal, and thus interferes with magnetic fields. Once you're within that 100 yards, all you need to do then is look for the pipe using the same magnetometer technology as the "metal detectors" that you see occasionally at the beach.

So anyhow, the core problem now is simply getting the relief wells down there. The formations appear to be rather fractured and the original well was a PITA that required regular interventions to get it through the formations in terms of adjustments of mud weight and drilling tools. The wrong mud at the wrong time can lead to the hole collapsing and the drilling tool being lost, requiring significant work to re-route the hole or fish out the tool. But with two relief wells being drilled, the chances of one making it to the gusher and doing a bottom kill on it are as close to 100% as you can get. And, alas, that's the only thing that's going to kill the gusher now, because according to the oil people at the Oil Drum, it appears that the integrity of the casing and blowout preventer at the top have been compromised to the point that closing in the well from the top simply won't work, it'd just blow the casing and BOP right out of the seabed, resulting in an even worse situation since then you have no (zero) resistance stopping the oil from turning into basically an oil volcano.

So that's the inside scoop. Bottom kill is the only proven technique for dealing with the current situation, there is no (zero) technology that will successfully execute a top kill on a flowing well, and given that we can't close in the well to stop it from flowing because it'd blow out the casing and turn this into an oil volcano, bottom kill is "it". This situation is why many nations require deep sea drillers to have a relief well drilled to bottom kill vicinity before penetrating the reservoir, so that a bottom kill can be executed within days, not months, if necessary. But that doesn't happen here in the USA because that would cost money to have two drill rigs out there, even if one of them could be idled after the relief well was drilled, and we can't have that, right? I mean, spend money just to keep from poisoning the whole upper Gulf of Mexico? Perish the thought!

-- Badtux the Oil Penguin

5 comments:

  1. And here I was thinking you're an economics nerd!

    ReplyDelete
  2. DA, I've done a lot of things over the years before settling into my current role as a professional penguin. The economics nerd thing is fairly recent, actually, and occurred because of my devotion to, err, actual *facts* rather than just repeating opinions that my "betters" spew. The Internet makes it easy to become a reality-based penguin rather than a talking-point-based penguin by looking at actual data rather than depending on other people to "educate" you according to whatever their agenda is. Alas, most people just don't care about facts, they have their opinions and they're stickin' to them...

    ReplyDelete
  3. I am going to love to hear the argument against requiring a relief well to be drilled at the same time as a new well after this experience.

    ReplyDelete
  4. Badtux already gave us the argument. Because it would cost too much money, silly person. And a blowout like this is just such an unlikely thing to happen, why the odds of it happening again are just inconceivable.
    But seriously, what keeps the relief wells from then collapsing or leaking? And how much faith can we put into the concrete plug, given what we're now hearing of Haliburton's track record on them? Is there any other company that does these cement jobs?

    ReplyDelete
  5. The relief well would be kept full of mud, and would not be drilled all the way down to the reservoir, so it would neither collapse nor have the possibility of a blowout. Regarding Halliburton's cement job, the problem is that cement takes time to cure, and BP didn't allow the cement time to cure, BP started pulling the mud out ten hours after the cement was poured. The cement was cured to about 2,000 psi strength, and was subjected to 15,000 psi when it blew out. If the cement had been allowed to cure for at least 48 hours before the mud was pulled, the cement would have held. But there was a bonus resting on not waiting that 2 days to let the cement cure, so...

    - Badtux the Oil Penguin

    ReplyDelete

Ground rules: Comments that consist solely of insults, fact-free talking points, are off-topic, or simply spam the same argument over and over will be deleted. The penguin is the only one allowed to be an ass here. All viewpoints, however, are welcomed, even if I disagree vehemently with you.

WARNING: You are entitled to create your own arguments, but you are NOT entitled to create your own facts. If you spew scientific denialism, or insist that the sky is purple, or otherwise insist that your made-up universe of pink unicorns and cotton candy trees is "real", well -- expect the banhammer.

Note: Only a member of this blog may post a comment.