Kill Pill Vol.1

Tomorrow BP will attempt to “top kill” the well that is spewing-who knows how much-oil into the Gulf of Mexico(GOM). BP estimates somewhere around 5000 bbs/day are escaping from the collapsed riser and blowout preventer(BOP), but researchers at Purdue, using estimates from live feeds, estimate the flow somewhere between 40000-100000 bbs/day.

The Purdue results appear to be high; top producing wells in the GOM generate about 50000 bbs/day. This well’s location is in an area(Miss. Canyon Block 252) of the gulf that isn’t as productive as some other areas. So, I tend to accept BP’s estimates… within a margin of error, of course.

The “top kill” procedure involves pumping drilling mud into the BOP in pressures greater than the escaping oil and gas. The ideal is that the pressure from the pumps combined with the weight of the overlying column of mud will displace the hydrocarbons in the production string of casing from the well head to the base of the well, and then fill the casements-effectively blocking the leak. Cement will then be pumped into the string to seal the well… if all goes according to plan.

And, therein is the problem-nothing has gone according to plan. BP, and others, have speculated that the reason for the initial blow out was that a poor seal in the contact between the production casing and the protective casing at the base of the well. Apparently, the blow by was noted by engineers when mud pit volume increased out of proportion-well logs indicate that twice they stopped pumping mud in the two hours before the blow out!

In yesterday’s press briefing, BP noted that in finding cause of the blow out, the “investigation is focused on the following seven mechanisms.”

1. The cement that seals the reservoir from the well;
2. The casing system, which seals the well bore;
3. The pressure tests to confirm the well is sealed;
4. The execution of procedures to detect and control hydrocarbons in the well, including the use of the BOP;
5. The BOP Emergency Disconnect System, which can be activated by pushing a button at multiple locations on the rig;
6. The automatic closure of the BOP after its connection is lost with the rig; and
7. Features in the BOP to allow Remotely Operated Vehicles (ROV) to close the BOP and thereby seal the well at the seabed after a blow out.

Regarding 3 above: rumor has it that a well known, and very expensive geotech firm was on the platform to do cement-bond logs, and that in a hurry to get off site(so the completion crew could move in???), BP’s top guy on-site put a stop order on the c-b logs. One thing is known-the geotech firm confirmed that they pulled their people 6 hours before the blow out- which was 14 hours after the final cement job for the seal started pumping.

There are other rumors floating around that the well was “kicking”… hard, and that for that reason, a “dump kill” was recommended hours before the blow out.

I guess that we will find out in Congessional hearings.

There are a lot problems with the “top kill” plan. Some have speculated that the mud, under those pressures, will fracture the formation, and that this will lead to greater flow at the base. Fracking wells is a common procedure to increase production in reservoirs, but as the sediments in the gulf are very permeable(around 30%), and barely consolidated, it doesn’t seem that if the formation does frak, that it would a problem.

Though, a real possibility is that blockages in the string will force the mud out the BOP, eroding what constraints are inhibiting the flow. Another danger could be a blow out in the containment string near the surface… if the seals are bad, and there are blockages, the mud could force its way into the containment nearer the surface where that part of the string doesn’t generally encounter those types of pressures.

BP estimates for success are 60-70%. If unsuccessful, a “junk” shot will be attempted in which various sized rubber balls will be injected into the system; If that fails, an attempt will be made to cut off the top of the damaged BOP and affix the Lower Marine Riser Package(Top Hat) to the clean surface.

I wish the crews luck tomorrow. Hopefully, by tomorrow night the well will be sealed… but I doubt it. From the opinions of engineers and geologists, that I have read, this thing might be flowing until the relief wells are finished in the late summer.

Note: I am not a petroleum geologist, and I do not work in the industry. If I have misconstrued anything, or is there is more to add, let me know in the comments.


Tags: , , , ,

19 Responses to “Kill Pill Vol.1”

  1. markgelbart Says:

    I don’t believe BP’s estimates. Before they put the relief pipe in, they claimed 5,000 barrels of oil were leaking per day. After they put the pipe in, they claimed they were preventing 5,000 barrels of oil a day from leaking, yet the video shows that about 14 times more than that was still leaking. Leak is a misnomer–it’s gushing.

    I believe Purdue’s scientists over BP’s.

  2. Solius Symbiosus Says:

    There is no way the well is pushing 50kbbs/d-that is what unrestricted top producing wells in the area produce… and the flow on this thing is constricted. Too, it is located in a region(MC 252) that doesn’t produce like the top producing zones. The professor at Purdue was using estimates from a wide angle lens view; he then didn’t account for the distortion. This guy is joke among engineers and geologists knowledgeable in petroleum.

    The USGS did a comprehensive study and came up with 19-39 mil gal leaked. That averages out to 12-24 thousand bbs/day… which puts the profs figs off by 75% for the high number and 87% for the lower one. The prof estimated errors in his calcs at 20%.

  3. Jarrett Ledford Says:

    Very great article! Honestly..

  4. Solius Symbiosus Says:

    Thanks, Jared. While I was educated, mostly, by profs. that spent their early careers in the oil business, most of what I have learned is by following petroleum and oil message boards and blogs.

  5. markgelbart Says:

    BP admitted 5,000 barrels a day was leaking before they put the relief pipe in.

    Then they claimed the relief pipe was preventing that much from leaking, yet visual evidence proves this can’t be right because it looked like 14 times that volume was still gushing. I mean this is just common sense.

    When it comes to science, industry spokesman are never reliable sources.

    • Solius Symbiosus Says:

      What evidence is there that 14x 5000bbs/day is leaking? The only evidence that I have seen, that use those kinds of numbers, are from a flawed and discredited analysis that relied upon distorted imagery to obtain the results.

      USGS, using multiple sources, came up with 12-19kbbs/day. Most engineers and geologist put the number at 15kbbs.

  6. Kenny Oneil Says:

    If only more people would hear this.

  7. Charley Holland Says:’s done it once again. Superb read.

  8. Carmelina Sarao Says:

    I wanted to say that it’s nice to know that someone else also mentioned this as I had trouble finding the same info elsewhere. This was the first place that told me the answer. Thanks.

  9. Pam Nunez Says:

    Super interesting article! Really..

  10. Sandra Cameron Says:

    Heh am I actually the first reply to this amazing writing!?

  11. Mark Gelbart Says:

    According to news reports, BP now admits that more than 5,000 barrels a day are leaking.

    And as you noted, the U.S. government estimates 12,000-19,000 barrels a day are leaking, far more than what you originally wrote.

    The truth is no one really knows.

    • Solius Symbiosus Says:

      I qualified what I wrote, originally, but your are right, there is no way on knowing the exact flow without a gauge measuring the flow, and that can’t be done. But there are other methods of estimating the amount.

      The researchers at Purdue and Florida qualified their study by noting that they were not experts in petroleum production, and used only particle flow rates for calculations.

      Whatever the numbers may be, there is a consensus among industry professionals that the flow is no where near the alarmist numbers that were being used by main stream media,

      The Purdue study had some very fundamental flaws in its methodology. It has been picked apart all over the internet by knowledgeable professionals.

      The point is really mute as no reputable organization uses those numbers anymore,, since they have been shown to be patently absurd.

  12. markgelbart Says:

    The Purdue scientist has been vindicated. According to AP reports, BP knew that when they were claiming only 1000 barrels a day was leaking, that 60,000 barrels a day was leaking. The scientist was right.

    • Solius Symbiosus Says:

      No he hasn’t, his methodology was flawed. Too, BP has never provided a flow amount, all of those numbers have come from the US Gov. It is the media that has distorted the facts, and used hyperbole to to sell print.

      Read this:

      From an earlier thread on TOD outlining the flawed methodology and sloppy science involved in the Purdue analysis :

      1 – He calculated the cross section of a 20 inch ID pipe, the riser is 19.5 inch ID – this error overstates the flow by 5.2%
      2 – A video with wide internet exposure filmed May 1 from the other side of the riser clearly shows the drill pipe coming out of the riser and bent to what appears to be almost a 90 deg angle. Assuming the drill pipe has an OD of 6-5/8″ (it may be larger) then just the area of the drill pipe entering the riser would result in an error overstating the flow by 12.3%. The error could easily be twice that due to the restriction by the bent piece of the drill string.
      3 – He assumed the entire cross section of the riser was flowing oil when it clearly shows that a substantial portion of the flow is gas, not oil. At the time he did his calculation the information was that the GOR was 3,000. With complete oil/gas separation and the compression from the external pressure at 5,000 feet that would mean there would be 20 cubic feet of gas per barrel of oil. As a barrel of oil is 5.6 cubic feet that would make the ratio of gas to oil about 4 to 1. But some of the gas must still be in solution as from the video it appears that the ratio is more like 1 to 1. This probably means he overstated the amount of oil by at least 50% and maybe as much as 75%.
      I will assume that as his expertise is flow measurement that his particle velocity of 25.8 inch/sec is correct (see below for some reasons why that may be off as much as 10% or more). Using that as a basis and recalculating the flow using his numbers but correcting only for the above mistakes we get the following:
      Cross sectional area is 264.1 sqin based on 19.5 inch ID riser less the 6.625 in drill pipe.
      Volume of the total flow at 25.8 in/sec is 6,814 cuin/sec which is 60,679 bbls/day. Just to this point he has overstated the flow by 19%.
      Assuming that 50% of the flow is gas reduces the oil flow to 30,340 bpd. This is 58% less than his published estimate that caused all the media hype.
      The errors above are not some special knowledge limited to the oilfield. They are errors that any freshman engineering student should be able to catch. I have a hard time believing a professional would make such mistakes, announce his finding on NPR and then go on to talk shows and testify before Congress. I wonder if he will apologize to the American people for deceiving them.
      If you are think about getting a good engineering education Purdue might not be the best choice.
      The following errors are more understandable as the professor has stated that he has no involvement with the petroleum business and presumably no experience in underwater engineering. At the same time it seems incredible that a degreed and presumably registered engineer would blithely assume that he could do an analysis like this without doing even minimal research into the fundamentals of the subject he is expounding on.
      4 – The videos from the opposite side of the riser leak show a piece of drill pipe impeding the flow. Not only does this reduce the flow area as mentioned above but it is also positioned so it appears that it diverts the flow up and towards the side that he used to measure particle velocity. That diversion could accelerate the velocity which is the basis of his calculation.
      5 – ROV video lenses are very wide angle. As objects move out from the center of the picture they elongate, or if in motion from the center out seem to accelerate. If you are doing a pixel based measurement you would need to know the lens distortion and adjust for it. In some cases the visual distortion can be as much as 25% to 30% by the time you reach the outer edge of the field of vision. As shown on his slide the area where he calculated his particle velocity is about halfway between the center of the screen and the right hand edge. This would indicate that it is possible that the particle velocity is actually slower than shown on his computer by as much as 10% or more.
      6 – The riser section was broken and deformed at the end. This will have resulted in some degree of out of round. It is difficult to tell from the video because of the distortion of the lens but from the other side it does appear the riser is flattened to some degree. Without knowing the exact ovality it is difficult to say what the reduction is but it is probably at least 15% to 20%.
      I don’t know if the following items will make any appreciable difference but they are certainly questions that should be addressed before making definitive statements about the flow rate.
      7 – This is mixed phase flow which any process engineer will tell you will add considerable confusion to flow measurements.
      8 – The flow is at an ambient pressure of about 2,250 psi. I know that gases can have a substantial change in properties when under pressure and some liquids also.
      9 – A portion of the gas is probably going into phase change as it combines with seawater which might change its real or apparent volume and velocity.
      10 – The optical qualities of water are different than air and may compound the visual distortion discussed above.
      11 – It is difficult to tell from the video but the flow appears to be laminar changing to turbulent on exiting the riser.
      If you take all these facts, and assuming the particle velocity he measured is representative of the entire flow, above I think it is possible to argue a valid case for oil flow rates anywhere from a low of 8,000 bpd to a high of 25,000 bpd emanating from the end of the riser.
      But, there is no way to take Professor Wereley’s velocity observations and get even 30,000 bpd, much less in excess of 70,000 barrels per day.
      The professor also states that the hole in the kink of the riser is 1.2 inches diameter. I’m not disputing that measurement but I would greatly like to know where he got it. I have spent almost 4 decades working underwater or viewing underwater videos and I certainly could not get that dimension from any of the videos or photos I have seen. Even ignoring the distortion of the wide angle lens the riser is so severely deformed it cannot act as a baseline and the piping off to the side appear to be some distance behind the leak so even if you know the pipe OD you cannot use it as an accurate scale.

  13. free credit card numbers Says:

    Good day! Could possibly I move your article in Pdf file file format? I really want it in my training.

  14. Winford Nargi Says:

    You made several great tips in this article.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: