safe distance from a steam pipe that ruptures 7

[ss123]

In light of the recent fatalities at the japenese nuclear plant caused by the rupture of high pressure steam pipe which caused fatal burns, could someone tell me how I could estimate what safe distance a person needs to be away from the pipe in case of rupture.

My steam pipes carry 10barg saturated steam. I've been told that a temperature of 70C is sufficient to cause fatal burns.

 

[davefitz]

You could predict the jet penetration distance using the same correlations used to predict steam jet penetration length from steam sparger pipes installed in condensers. One would also need to know the critical heat flux needed to burn the skin.

On the other hand , you would never know when or where the failure would occur, so what is the point of knowing how far away to stand from an unknown location?

If 70 C at 10 barg is a concern, I guess we should be real worried about the leaks from a 250 bar,g 565 C HP main steam line . Knowing what I now know regarding the way the P91 piping has been fabricated , maybe I should buy some life insurance.

In 1988, there was a steam line failure at the Mohave station in Nevada, USA. The hot reheater P11 pipe failed , at 31 bar,g and 593 C temp ( design temp was only 540 C, and it was overheated for 1 yr prior to failure). It failed outside of the plant cafeteria, and killed 11 operators who were eating lunch in the cafeteria.

In 1998, there was a new coal fired plant in China which expereinced a slag fall , where a 50 ton piece of coal slag fell 60 meters to the furnace hopper. It casued the furnace feeders to fail, and discharged 200 bar g saturated liquid ( which flased into steam) at 343 C,, and killed 22 workers who were rodding out the hopper slag trap.

 

[rmw]

If you operate on the roads, you live with the specter that a head on collision could take you at any moment.

If you work around steam pipes, you live with the specter that the grim reaper could call your number at any time.

Is any distance safe?? In some areas, the steam velocity and the temperature might not get you, but you might suffocate when the atmosphere was replaced with steam.

I am not pessimistic, only a realist. Unsafe practices are accidents looking for a place to happen. A lot of what our profession is all about is preventing others from getting killed.

One of the first lessons taught me when I was introduced to steam shipboard was to always have an escape route. Since the boilers were deep in the bowels of the ship, the old hands route was to pull the deck plates up, and jump into the bilge water below, rather than chancing a 10 meter climb up a ladder in a live steam atmosphere.

I have never forgotten that.

Now, where did VPL get that soapbox emoticon. I need it now.

rmw

 

[davefitz]

to add a correction to a prior post:
The failure in China was on 10 Mar 93 at the BeiLunGang plant, Zhejiang province, and 23 workers were killed

 

[epoisses]

You don't design for a pipe to fail. Operators don't stand next to pipes (only engineers do when it's sunny). Besides, I guess the safe minimum distance depends a lot on the kind of rupture. I would spend the time and money on better ways to improve plant safety!

 

[pmover]

while all postings are of value, there is a book titled "what went wrong" that describes "case histories of process plant diasters".

i believe the 5th edition is now available.

i consider books of this type to be of high value to designers, engrs, etc. whom are involved with plant design (process, power, etc.).

perhaps other forum readers may offer additional reading recommendations.

-pmover

 

[LuizSouza]

Pmover, Who is the author of this book and where can I find it?

 

[saxon]

ss123, In some facilities there is the hard and fast rule that "NO ONE is allowed in operational areas except by permit that is controlled by the operations superintendant" These are issued on a daily as needed basis only. Reason being, processes go critical, equipment fails, piping fails, instrumentation fails, etc. By minimizing individual exposure, you minimize potential injuries and deaths.

Hope this helps.
saxon

 

[ss123]

I would like to thank everyone for your responses to date. I would still like to hear from anyone who has done some design calculations to estimate the distance the jet of steam needs to travel from the ruptured pipe before its temperature reaches 70C.

I've estimated the steam velocity for 10.5 barg and 320C to be about 325m/s assuming choked flow . However, what happens beyond this point involves some complicated heat and mass transfer between the steam jet and the entrained ambient air.

Can anyone help?

 

[CHD01]

My first thought is that you might simplify things (ALOT) and try to apply API gas dispersion formulas based on steam properties to estimate dispersion distance horizontal and vertical for the steam jet in air. Then separately approximate jet temperature at various dispersion distances based on a simple ratio of steam to air concentration. I envision obtaining two answers, one baserd on dispersion and the second based on temperaure drop; with best estimate based on a plot of the two functions.

I will think about this somemore, as there are surely more precise methods to model this, but the above is as close as I can think of using already developed formulas.



The more you learn, the less you are certain of.

 

[ss123]

Thanks CHD01 for your comments. Could you direct me where I can find the API gas dispersion formulas? Also, to put an extra spanner in the works, the temperature drop of the calculation will also involve taking into account the latent heat of condensation, and subsequently the flow of the jet will quickly become 2-phase flow.

I've thought of another idea as well. That is to think of the problem as being like the a steam ejector, where the choked flow of the steam entrains the ambient air. One can then assume that this mixes instantly with the steam and so cooling it rapidly. Do you or anyone have any idea as to what the entrainment rate of atmospheric pressure air into a steam jet produced by a supply of 10.5bar steam is? Alternatively, do you think this approach is ok?

 

[athomas236]

davefitz,

The accident in China was caused by the operating company continuing to operate the boiler against all good sense and advice even though it was aware of the slag build up.

As you can imagine the American company that suppied the boiler is very sensitive on this matter.

athomas236

 

[MJCronin]

ss123,

Your question should be, "What measures are commonly taken to inspect, evaluate, maintain and test high energy piping systems ?" It is, in my opinion, impractical to develop "exclusion zones" around all complex industrial, commercial steam systems. Part of the answer to that question is incorporated in the recommended mainteneace programs found in the ASME B31.1 piping code appendices.

You should also note that in the late 1980s, the USNRC has developed and enforced a nuclear piping inspection and asssesment program under the diretion of the Electric Power Research Institute (EPRI). This program involves identifying and ultasonically testing of suspect locations where erosion/corrosion may exist. The program was implemented because of a feedwater piping failure in a nuclear plant in Virgina (as I recall)

Pmmover,

The book that you are refering to is titled "What Went Wrong ?" by Trevor Kletz (now in the Fourth edition and available on AMAZON). It is, essentially a valuable casebook on process plant disasters, but has little mention of steam piping systems. Kletz followed this book up with a sequel called "Still Going Wrong " which is more of the same. Kletz hates process plant management who seem to care for little besides the bottom line. He also scorns a system where there is, essentially, no "lessons learned mechanism".....this, plus his black and bitter sarcasm, make him one of my favorites.

davefitz,

I too am curious about the nature of this steam piping failure...however Japan runs mostly PWRs and the operating temperature range of the steam systems means that mostly plain carbon steel systems will be in use. The failures that you mentioned happened mostly with Hot Reheat steam piping fabricated from A335-P11 (1.25 Gr - 1 Mo) with an axial welded seam. After some time at elevated temperatures, the seams gave out due to high temperature creep failure. Many utilities replace the P11 with higher alloy P22 (2.25 Cr - 1.25 Mo) or P91 materials. EPRI again led this effeort for the utilities.

Gimmie a star....

my opinions only

 

[TBP]

In over 25 years of working around steam systems, the two guys (who I both knew personally) who suffered the worst burns, suffered them from LOW PRESSURE CONDENSATE. Most people are pretty careful around steam - especially high pressure steam. Condensate usually gets treated much more casually. One of the men was in hospital for several days before they were even sure he would live. His accident was caused by someone else doing something careless & stupid. He was off work for a year. The second man was hospitalized for weeks, and off work for months. I can't prove it, but I firmly believe that his accident was due to downsizing/understaffing. Having anyone, let alone someone in their late '50s, doing a physically demanding job 6 or 7 days a week, with the days often extending 10 - 12 hours, for months on end, is just an accident waiting to happen. It's a good thing he stayed on his feet, because if he'd fallen, he'd likely be dead You can be very badly injured by having your work boots filled with 200*F condensate.

 

[grandnobi]

ss123,

I remember having read an interesting article about ruptures in steam systems. It can be downloaded here:

http://www.kirsner.org/kce/media/pdfs/KirsnerSurvive.pdf

Regards, grandnobi

 

[davefitz]

MJCronin:
No star yet.

The Mohave hot reheater pipe failure was due to several errors, all of which are being repeated today ( with P91) and which are inviting similar failures.

The Mohave unit was designed according to an early 1960's ASME section I code, which overestimated the high temperature allowable stress for SA335-P11 piping, and did not provide for a weld creep strength reduction factor when the weld is utilized in longitudinally welded piping ( as typically used in reheater piping).

The plant was a slurry-coal fired plant , which experienced excessive slag buildup in the furnace waterwall , which led to high furnace exit gas temperatures and resulted in hi-hi hot reheater steam temperatures,beyond the ability of the reheater attemporator spray to control to the 1005 F design temperature. The unit was operated for about 1 year at 100F over design temperature, which leads to a creep life that is about 20 times shorter than if operated at the correct design temp of 1005. F. To date , ASME section I does not proibit such operation above design temperature. The pipe failed near an elbow , at a weld that was already weld repaired twice for prior evidence of creep cracking and seepage.

Similar errors are being repeated today with the widespread use of SA335-p91 , which has apparently been implemented in an incorrect manner. The engineers and designers who have specified this alloy are generally not familiar with the extra precautions needed for fabrication of components with advanced creep resistant steels. In particular, hot bending is not being accompanied by normalization and tempering, and when normalization is done it has not been accompanied by the correct cooling rate ( not less than 9F./min min permitted cool rate to avoid ferrite formation)and requisite hardness testing and photomicrographs are usually lacking. Similar errors are being promulgated at large dissimilar metal welds ( as at turbine stop valves) and welding procedures in general have not included the appropriate low hydrogen procdures and correct post weld heat treatments. And, of course, and explicit weld creep strength reduction factor is not yet defined by code .

 

[philbob]

A likely way to get a burn is not so much a pipe rupture, but a pressure relief device activating properly and flooding an area with steam. It's tempting to write "Vent to safe location" on a drawing and forget about it.

Apologies if this has been raised previously!

Philbob

 

[Latexman]

ss123,

All good posts! The best way is to have a good piping design, the right materials of construction, a qualified installation, a proof test, and a never ending test and inspection program with timely maintenance.

Some of the previous posts mentioned the resulting steam flow jetting into the area and how to get a feel for that. What I didn't see was - what is the effect of the expanding steam at the instant of rupture. A good way to get a feel for that is calculate the work the steam would do on the environment IF the entire piping instantly disappeared. This work would be the integral of P dV. Convert that into equivalent pounds of TNT and look up separations based on that. There's lots of references out there for this.

This may require some abstract thinking to apply, like how long of a run of pipe do I consider, but it'll give you a feel for this component of a steam pipe rupture.


Good luck,
Latexman

 

[sailoday28]

Latexman (chemical) On the work term, how would one look at the change in volume, ie in a fixed rigid compartment, volume would be constant?
However, if rigid, one could compute the pressure build up from internal energy change. U(kj of compartment and what was a finite source) =U (initial of compartment) + U(initial of finite source.

 

[Latexman]

Sailoday,

I was envisioning the area where the pipe ruptured to be open, or large, or small with sufficiently large openings that the resisting pressure to the expansion would essentially be atmospheric pressure. W = ? P_res dV, where P_res is the resisting pressure. At time 0-, the steam is at some high pressure and confined by the piping. At time 0, the pipe disappears but the steam is still at this high pressure. At time 0+, the steam has expanded to atmospheric pressure. Initial V would be the compressed steam and final V would be steam at atmospheric pressure. This is basically a closed system that becomes open.

If the steam pipe ruptured inside an “air tight”, closed compartment, your method would be correct.


Good luck,
Latexman