Pneumatic Testing - Show me the failures 3

Take a look at thread481-348164 for a great discussion and some of the pictures you are looking for.

mjcronin posted the actual incident report from the picture you attached (and I'm eternally grateful to him for posting it). It is interesting reading. The tank was not even part of the test, but the plant was testing some piping into the tank against closed valves. The valves leaked faster than the tank could exhaust the gas (I'm thinking that they may have taped over the vents for painting, happens all the time, but it wasn't spelled out in the report) and the tank launched. Is that really a pneumatic test failure since the tank wasn't part of the test? In my class I use that incident as an example of an Engineering failure. Some of the other examples he posted fit into the "bone headed Engineer" category or the "brain dead operator" category instead of the "pneumatic testing is reckless and irresponsible" category.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

I am not a fan of pneumatic testing.

There was one report of at least one death occurring during a hydrotest involving connecting hoses being disconnected with trapped air inside. I'd not forget to mention that hydrotesting can be a dangerous enough operation on its own.

I'm looking for that report now. If I find it I will post the link.

Not the one I was looking for, but serves to illustrate my position,



Independent events are seldomly independent.

Just ran into this "Hydrotesting Student Manual" by California Fire Marshall office.
Perhaps useful to you, even though it is directed towards hydros.

Pressure Testing Requirements for Hazardous Liquid Pipelines in California:

Student Manual; click here.

Another note you might want to incorporate:
Pneumatic testing for regulated liquid pipelines is not allowed.
B31.4 requires hydros. No provision for pneumatic testing.

Independent events are seldomly independent.

Here we go!


Independent events are seldomly independent.

Thanks, it's all good stuff. Though, I did actually have a good bit of it. That one about the dewatering incident was new to me.

Edward L. Klein
Pipe Stress Engineer
Houston, Texas

"All the world is a Spring"

All opinions expressed here are my own and not my company's.

I really hate those ones that just kind of jump up and get you, even though you think you've taken every possible precaution already, even to the point of NOT actually doing a PNEUMATIC TEST. Especially ironic.

The "fat tail" theory. Even a miniscule probability of a very big event can still deliver significant consequences. Especially when the problem is thought, even by supposed experts, to be simple, but in hindsight still turns out to be not sufficiently well understood. My(?) definition of complex systems. I think it has something to do with the probability of independent events in complex systems actually summing, contrary to the accepted math, to be higher than 1.00 Should a safety factor be applied according the number of events that are being analyzed. One, two independent events = 1, three = 1.25, 4 events = 1.5, 5 events = 2, 6 events = 2.5, or something on that order. The probability and statistics that applies to dice and cards I'm thinking cannot be extended to "real" events. "Luck" for lack of a better term, or the lack of it, starts to crowd out the math. Every once in a while, I beat the casinos, but only when I stop doing the math. Like looking at a photon, does measurement change the problem? I think that's an interesting question. Sorry for the rant.

Independent events are seldomly independent.

Sounds like an interesting project. fyi if you would like to make a sub-section of your presentation, the most serious incidents are of course those that result in fatalities. There have been many. Here are a few now with easily accessible iinks (I am aware of at least a couple more, that for whatever reasons the links are no longer active. You can rest assured there are many more, before and since that I am not aware of, or was not able to find reference of on short notice):

,
, ,
page 2 of the newsletter at , (I have heard that in the latter one three workers were killed, with one being virtually disintegrated into hard to find pieces, while working on a line under pressure with very high pressure nitrogen from test), ,and finally (while I realize it doesn't quite meet the fatality criteria) see another case of pneumatic testing gone quite awry recounted by Mr. Joe Facer AFM #237 near the bottom ofthe page at including the quote, "..An apprentice in my Local was paralysed for life when someone tried to test a plastic piping system with #100 air. The 6" plastic pipe ruptured/exploded and threw him across a room and crushed his spine..."

I suspect there were many photographs taken of at least some of these incidents, and who knows you might be able to obtain some with info/contacts provided in these sites (i.e. unless records have been somehow sealed as a result of settlements or other proceedings etc.)

This is not the type of coupler that was demonstrated in the OSHA accident report that I would use for such test.

The pneumatic test failure mentioned by rconner in Shanghai is probably the most spectacular and most costly in recent memory.

It could have easily happened in the US......

On almost every job I have been on, the mechanical contractor who was awarded the job wants carte-blanche permission to use a pneumatic test in lieu of a hydrotest.

Of course he does..... this saves him money and speeds up the schedule.... he, of course demands written permission from the engineering firm on his "innovation"...

Typically, the dumb-a*s MBA that I end up working for is usually on his side and demands from me detailed reasons why this is not acceptable.

I keep selected websites with pictures of pneumatic test accidents on a thumb-drive.....

A good paper on hazard/risk assessment for pneumatic testing: