Piping disasters and Incidents

Hi
I'm working with an 'engineer' who is of the opinion that piping design is all about routing piping and the structural aspects are not real and that I should 'get out more' [grrrrrrrr!!!!!!].
Anyway I thought I'd do a bit of digging and come up with some references or articles on disasters or incidents where poor piping design lead to large scale loss of life or assets to remind him of what happens when we ignore basic engineering principles. Paradoxically it does seem sometimes that the field is now so safe that people have become remote from what can go wrong and the consequences of that.

So far I have come up with
Flixbourgh - http://www.hse.gov.uk/comah/sragtech/caseflixboroug74.htm
BP Grangemouth - http://www.hse.gov.uk/comah/sragtech/casebpgrange87.htm
Tay Bridge disaster[not piping related but a facinating bit of history regarding forensic engineering] - http://www.open2.net/forensic_engineering/riddle/riddle_01.htm & http://www.open2.net/forensic_engineering/riddle/riddle_07.htm

Reading up on these has been quite illuminating and is great background information for everyday use.

Can anybody else provide links to piping design related incidents so we can all learn more from the mistakes of previous generations?

Kenny,

Excellent idea.

Quote - "You must learn from the mistakes of others. You can't possibly live long enough to make them all yourself." ~ Samuel Levenson.

http://www.hse.gov.uk/research/rrpdf/rr253.pdf (Recommendations for avoiding failures)

http://www.hse.gov.uk/comah/sragtech/systems10.htm (Compendium of failure modes)

Here's one down to poor piping design, poor communication and penny pinching. The company standard was bellows on every nozzle, and were included in the design before a stress engineer was taken on. He modified the function of several to improve the design, but one required change was not communicated to the manufacturer by the company, and the stress engineer was laid off so unable to check the final design.

A bellows able to take 3mm lateral movement was thus installed where about 50mm was imposed. It survived about 3 months then ruptured, shutting down the local town.

http://www.hpa.org.uk/chemicals/reports/cir17_jul2000.pdf

MoverZ

You seem to have significant information that is not in the document available to you concerning this incident, first hand knowledge.... hmmmm


Frankly the common blame for failures are usually bad materials, bad fabrication, bad installation. Seldom does the design get faulted... although IMHO it may be at fault just as readily as any other root cause.

A partial list of famous piping disasters:

Piper Alpha (1988, North Sea) - Offshore platform explosion, 164 deaths
Bhopal, India (1984) - Methyl Isocyanate release, thouands of deaths
Mina Al-Ahmedhi Refinery (2000, Kuwait) - flammable vapor release and explosion - 7 deaths
Petrobras (2001, Brazil) - Offshore oil platform explosion, 10 deaths, platform sank
Ammonium Nitrate Explosion (2001, Toulouse, France) - 31 deaths
Baytown, TX (2004) - Naphtha unit explosion and fire
Mojave Station - (1986) Hot Reheat seam-welded piping failure
Lodgepole Well - (1982, West of Edmonton) - not strictly a piping failure, but a failure of the drilling operation led to a release of billions of cubic feet of methane
Three Mile Island (1979, Harrisburg, PA) - blockage of a feedwater line led to a release of radioactive steam
Chernobyl (1986, near Kiev, Ukraine [then USSR]) - likely water hammer incident triggered failure of reactor cooling water flow, leading to a major release of radioactive material. Portions of the site are still off limits.

hi, kenny

this topic is really intresting. this will improve our knowledge in this field. this is very usefull .


regards
B.Sureshkumar

Two related references that may be of interest:

"Defects and Failures in Pressure Vessels and Piping", Helmut Thielsch, VP Research, ITT Grinnell Corporation, Robert I. Krieger Publishing Co., Florida, 1977 (Mr. Thielsch has several other related publications.)

"Pressure Systems Casebook: Causes and Avoidance of Failures and Defects", Ed. by J.B. Wintle, Professional Engineering Pulishing, 2004

I also recommend, "To Engineer Is Human." It discusses failures of all kinds, throughout history. Can't remember the author, but I think he's at the University of North Carolina. Interestingly, his thesis is that we have learned way more from the (relatively rare) engineering disasters throughout history than from all the successes combined.

One can see, from the changes to the seismic design codes as a result of the Northridge and World Series earthquakes in the 1990's, that he may indeed have been onto something when he wrote this book. Not that I would recommend any of us intentionally trying to cause a disaster!!!

Wouldn't intentionally causing the disaster nullify the learning experience? Learning from your mistakes only really comes if you've made the mistake thinking your answer was sufficient.

I assume everyone saw the Steam pipe burst in New York 7/19/2007? Its on cnn.com today.

To Engineer Is Human: The Role of Failure in Successful Design by Henry Petroski

# Paperback: 272 pages
# ISBN-10: 0679734163
# ISBN-13: 978-0679734161

Actually I think the issue with Lodgepole was sour gas (H2S) not specifically methane.

The following report on the transmission pipleines may not be directly related to the accidents. However, it includes some risk related information may be very useful to some readers.

http://onlinepubs.trb.org/onlinepubs/sr/sr281.pdf


Ibrahim Demir
Clyde Babcock Hitachi Australia

I have found another one that you might find very interesting:

"Property Risk Consulting Practice

The 100 Largest Losses 1972-2001.
Large Property Damage Losses in the Hydrocarbon-Chemical Industries.

20th Edition: February 2003.
A Publication of Marsh's Risk Consulting Practice.

MMC Marsh & McLennan Companies"

www.marsh.com


Ibrahim Demir
Clyde Babcock Hitachi Australia

Another reference which may or may not prove to be the cause of the NYC steam line failure.... http://www.kirsner.org/pages/jackSteam/jackSteamTextAlt.html

The latest one happened in Insaleh Facility"somehere in may this year during commisioning. Flare line was undersized and veclocities were very high. I will try to forward the mail which has intresting finding about the whole stuff but fortunately no fire or explosion occured.
we have to keep thier "eyes open
Regards

Alok,

The linked paper below appears to be from an earlier incident (at In Salah - same place?) although flow velocities are topical. Turbulence induced fatigue Vs. acoustic fatigue.

http://committees.api.org/standards/cre/scprs/mtginfo/docs/s06/7-fatigue.pdf

Stan,

Thanks for posting this paper on the In Salah incident. The client on my current project is taking a very aggressive approach in addressing fatigue from acoustic vibration and sites this very incident as a reason why. While I'm familiar with MTD 99/100 and CONCAWE 85/52, this is the first time I've seen the report that you posted.

Thanks again,

Stan,
No, picture is same, so I believe it`s may 2005 not 2007, however the report is bit comprehensive.
Could anyone let me know how to post/attach the document over here
regards,

Click the [Reply] option (to a post). Just below the "text input box" is a link for "File Manager". This will let you browse for files to upload, and it will automatically put a link to them in your post.

pls find the report
best regards,

Another good, easy and cheap read is:

"Why buildings fall down: how structures fail", Matthys Levy and Mario Salvadori, W.W. Norton and Compay, 1992, ISBN 0-393-31152-X.

Redundancy, shortcuts, etc...

If you can find a copy, I also recommend

"What Went Wrong?" by Trevor Kletz
http://www.amazon.com/What-Went-Wrong-Fo...85998760&sr=8-3

I haven't read this one, but I see he's also written a follow-up called "Still Going Wrong"

http://www.amazon.com/Still-Going-Wrong-...85998760&sr=8-4

can anybody of you have any documents related to LOF criteria for piping component evaluation from stress point of view

"Mina Al-Ahmedhi Refinery (2000, Kuwait) - flammable vapor release and explosion - 7 deaths"

This incident was due to un-noticed corrosion on the pipe line. There was no design issue. the damage line was in service for morethan 25 years.

Here's another incident.

en.wikipedia.org/wiki/1998_Esso_Longford_gas_explosion

A hot oil system shut down, the natural gas brought the temperature of the heat exchanger down below its embrittlement temperature, and the thermal gradients induced when the hot oil restarted caused a brittle fracture of the exchanger. It released about 10 tons of natural gas, which ignited.

Moral to the story - consider upset conditions!

Just to emphasie why we need to get it right first time, every time!

http://edition.cnn.com/2008/US/02/18/refinery.blast/index.html


It certainly brings it home what can happen when engineers get it wrong.

Very lucky there were no casualties, but luck will eventually run out.


Regards
Roy

In previous post meant to say "Lucky no Fatalities"

I've just discovered how to add an attachment, Hope it works!

Roy

kenny

relly it is very good idea and u must makerealize people like ur engineer



thanks
rakesh

while this is not a disaster as yet surely it is a disaster waiting to happen?

Dear all,

These also could give contribution in piping disasters and Incidents

Dear Richard Ay

I coppy the input files of my model but when coppying a message was shown like "datas cannot coppied". I continued coppying by clicking "ok" but my client sent mail that they can't open the files and want from me to copy all files. How can I solve this problem, I have to copy all of input files but I can't. Is there ny way to solve this problem or did I something wrong? If you help me I would be very apreciate.

Best Regards...

Hasan Burak Basar

P.S. : I open my models and can analyse them, but I can't copy only the data files. If you want more information please ask me for solving this problem. Thanks

Another incident...

The line slipped off the structure which was designed without taking care of large thermal movement, support, and reaction.

Hi, does anyone know if there are also accident caused by failure of rubber expansion joints??

I read on case were the joint was used in the wrong way:

"A bellows able to take 3mm lateral movement was thus installed where about 50mm was imposed. It survived about 3 months then ruptured, shutting down the local town."

Can anyone tell me about problems or faillure of rubber expansion joints??

Best regards,
Sören Blomaard

Dhirubhai 1 FPSO shut in Dhirubhai 1 the first floating production storage and offloading vessel production oil from Block S6 in India`s Krishna Godavari basin , has been shut in . Operator Reliance Industries said a rupture in a short pipe connected to the floater`s flare header forced it to shut in operations on 9 december. There was no fire , no oil spill or any casualty.

http://news.xinhuanet.com/english/2009-02/07/content_10779866.htm

The above was caused by a very high pressure PNEUMATIC PRESSURE TEST. In the USA we sometimes use pnuematic (Air) tests for very large diameter lines but the design pressures are low ~50. psig, not 2200. psig.

Quote:

The explosion was caused by a sudden flange failure while the workers
were conducting pneumatic testing of a 36 inch diameter line, 600m
long segment, laid out in an S shape.
The testing pressure was 15.6MPa (2,262psi), and the explosion
occurred when the system pressure reached 12.3 MPa (1,784psi).
The explosion was caused by the rupture of a flange at the end of the
test section.

Hi to all,

Im doing stress analysis for the 8" PTFE (Poly Tetra Fluoro Ethylene) lined Carbon steel Pipe using CAESAR 4.5. As a part of Client requirement the Pipe is to be insulated by thermal cellular glass. The insulation thickness for the pipe is 1” and the same is being entered in the CAESAR input data. And now im unable to provide the lining thickness (3mm) for the pipe as I observed from Caesar Documentation “If a negative value is entered for the insulation thickness, the program will model refractory lined pipe”. Kindly advise me whether to consider lining thickness for doing stress analysis. Does it have an impact on thermal expansion of piping?
Incase lining to be taken into account clarify me on the consideration of both lining and insulation thickness on the input.
Kindly suggest on the above clarification which will be greatly helpful for me.

Best Regards
Deepak.S

http://home.versatel.nl/the_sims/rig/i-fatal.htm

This is a test post

I'd like to think that it would just blow the gasket under hydrotest, how did you discover this, surely it would've been hidden?

hi
I want to know how to calculate the weight on anchor bloc ; with caesar I put restraint ANC and durin the anlyze I found the value 9000 N loads , then I change the configuration and install a bend LR away from the ANC and run caesra again the I foun the load on the ANC 5956 N ; better than the first case, but hox to chosse the dimensions of the concrete block?
please can some one can help me
tanks in advance

Algerie,

Please start a new thread, this on is for "Piping Disasters and Inceidents".

I'm going to delete this post on Monday.

Hello Sam,

What was the design temp of the line that fall off the support.Just being curious, so that when I encounter same type of problem I know what to do...This is with regards to the expansion loop incident.

Thanks and Cheers,

(off topic)

The incident report for the DWH in the gulf is going to be an intresting read.

I bet everybody on the rig knows the problem(s) yet the official report won't mention any of them.

It is not a real disaster but it is a disaster what is happening with the new generation of engineers. I think that in their run to power they do not think to potential disasters. As safety was quite good in the latest time they simply can not realize what they are doing. No offence, this is the world we are living. What will be?

Comments, remarks?

Regards,

Hi Dan,
it seems that you have problems with the new generation engineers. I also saw such remarks in other threads. As I am one of these new generation engineers (age 31), I want to reply to your post.

Nowadays engineers are mainly trained to use computer programs. This is a problem. I'm of the opinion that one needs to know what happens in the "black box" (computer) and what the basis for the used calculation methods are (like 3d beam elements). Also knowledge of material behaviour and code requirements and their background is important.

Unfortunately this is almost not tutored in school nowadays. Although one can learn from internet, one must learn by doing. And one can only learn from seniors that are willing to share their knowledge.

Partly I can agree with your opinion, but I think you're too much generalizing. Also don't underestimate the role of seniors in the learning process.

Yes, you are right also. My arrow was not against the stress engineers, but piping. They are not to be blamed because of ignorance regarding the stress problems but for the overconfidence that these aspects are only secondary as importance.

I'm not so old (42) and I'm also trained to use computer programs but I 'm not thinking that all is rubbish.

And yes, you're right again then notice that almost nobody is willing to share. But part of this is also because of "we only need some PDS skill"

Best regards,

Hi Memebers,

This is very good debate topic relating to the new generation with computer calculation and old generation with manual calculation. I am a new generation (23 yrs) but I’ll support for old generation because of the following reasons,

(Please note that the following reasons are not for all new generations).

Drawbacks of new generation:

1. Piping engineering study is not available in school or collages, we can learn only based on our experience, available case studies etc.
2. Fast moving world, no time to think and work.
3. Rarely find experienced person in a concern to guide us.
4. Life is money, money get work and time, no time to think and work.
5. In spite of intelligent work, we work cleverly.
6. The fact suits to our topic "one time writing is equal to ten times learning by voice." Now days we don’t find time to solve the problem manually because of the readily available computer program, so not using our brain effectively.
7. Invention of computer programs made the new generation to become sluggish. I am not criticizing the developer of computer programs (In fact, it is a good invention, which saves lot of time to carry out the work, but the only problem is that the user scope is not defined properly (i.e.) from my point of view, the computer program shall be used only by the experienced persons who are expert in the stress analysis. For all other users they shall continue doing the stress analysis manually. But the owner of the computer program may not accept this because of the money which depends on the number of users. We cannot blame the owner since they are human who love money.
8. Atleast the final hope should be the client but even there we can find some experienced people only in their “Age” but not in their work (not all – some people). They simply throw their s-h-i-t on their contractors, stating that “all the responsibility shall be taken care by the contractor; we cannot review your documents, give comment on it etc.”


Thankyou,

Let the debate continue...........

can you tell me please how to attach file or any document in forum with post.

"Lessons learnt from industrial accidents, IMPEL Seminar, 2011" can be found from the following link:

http://impel.eu/wp-content/uploads/2012/03/brochure_gb_impel2011.pdf

Hope it might be useful for some of us.

Regards,

Ibrahim Demir

Hi,
I sent the pictures that disaster happened in plant,in order to warn the engineer it is real dangerous when u treat the design carelessly.

Hello Everyone,
Can anyone performed pipe whipping analysis or Pipe break analysis?
Please provide the references, if any.
Thank you.