hazards of pneumatic testing? 1

[TomBarsh]

Last week I came across some calculations for a vessel that will be pneumatically tested. Test pressure is about 140 psi. The vessel itself is about 18 feet dia x 260 feet tan-tan.

Maybe it's just me but this test seem to be potentially hazardous to health if there's a failure. ??

 

[jte]

Sounds to me like a C3 splitter tower. I worked on a vessel of similar proportions which was field fabricated and field hydrotested. With a vessel this size, a pneumatic test would contain a whole lot of engery.

See thread794-26767 and let us know how many tons of TNT equivalent you have!

jt

 

[DSB123]

TomBarsh,
Strongly suggest you are nowhere near during the test. Preferable to be in the next state.

 

[deanc]

Go to the NASA link in the thread. I carry binoc's just for things like this.

 

[deanc]

http://smad-ext.grc.nasa.gov/gso/manual/chapter_07.pdf

Try this one. See index.

 

[davefitz]

For all the concern expressed against pnuematic tests, those exact same pressure vessels are operated continuously for 30+ years at close to the same gaseous operating conditions ( and the exact same degree of hazard) as occur during the pre-commission pnuematic tests . In fact , for the case where the working fluid during normal operation is hazardous ( ammonia, hydrogen, flourine, etc), the workers are exposed to the toxic hazards plus the increased risk associated with the vessel undergoiing thermal stresses, ovepressure during relief valve lift events, etc.

I am not underplaying the danger associated with the pnuematic test, but suggesting that people tend to get used to , and ignore, the exact same hazard during their day to day work life.

 

[vpl]

So should they hydrotest this tank first?

Verify the relief valve works?

davefitz, it sounds like you have some knowledge in this area. What should TomBarsh be ensuring is in the test procedures?

Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[codeeng]

I know of field erected vessels in a Polyethylene plant which require an air/gas test since the smallest rust bloom would poison the process. I recall a safe distance of over a km based on the following:

SD = 15 x (TNT)^0.5

SD = m
TNT = lb

This type of test can be accepted by your AI when properly justified.

 

[jte]

davefitz-

You're argument is valid to a certain extent. The consequence of failure during operation is typically higher than that of a failure during pneumatic or hydrotesting. However, the probability of failure is greater at the initial test: This is when fabrication foulups will most likely exhibit themselves. By limiting access to the area during the test, the consequences of a failure can be mitigated.

To throw an analogy out here, why couldn't I take the same argument to say that I shouldn't have had to take the PE test: After all, I studied for the test, and I'm more likely to forget something important later, so since I studied, just give me the PE without the test and let me get to work designing equipment. If I mess up on some design 10 years later, I'll surrender the PE.

jt

 

[TomBarsh]

All very interesting comments. I am not personally involved in design or test of this vessel but I became aware of it and thought "hmmmm, is that really wise?". Of course, I have few other details so it's not really for me to make any professional comments other than "jee willickers, Mr. Science".