Pneumatic test pressure less than 1.1 times design pressure

[bck2ftr]

Dear Friends,

Here I am having a lines for which Design Pressure is 39 Kg/cm2g.
The line needs to be complete dry, hence we are going for Pneumatic test.
As per Code ASME31.3, the Pneumatic test pressure comes out to be 1.1 x 39 = 42.9 Kg/cm2g

As I understand, such high pressure is not suitable for Pneumatic test..
I am trying to look for any reference where Code will allow me to conduct the test at lower pressure than 1.1 x Design pressure.

Is anyone aware of such Code reference?

 

[zdas04]

I've done tests at far higher pressure than that. I don't think you are going to find a code reference that sets a magnitude for the upper limit of a pneumatic test.

I really hope that kg/cm^2 is not the "future" you are going "bk" to. What a ludicrous unit.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. —Galileo Galilei, Italian Physicist

 

[weldstan]

There is no lesser pressure other than as permitted for a sensitive leak test. In terms of safety, you need to look at the diameter, wall thickness, stress in hoop versus Specified Minimum Yield Strength of the pipe, welded or seamless pipe, required NDE of welded pipe, toughness characteristics of the pipe, and test temperature. Additionally the test apparatus must be confirmed safe.

 

[jte]

Hey zdas-

What, would you prefer slugs per square fathom? More seriously, I really doubt that bck2ftr is choosing the units which his/her facility uses. Just as pressures in facilities in the US are often referred to in "pounds" (and I get the roll-eye when I ask them if they mean per square inch), as you know there are several versions of "metric".


bck2ftr-

What product is the piping expected to transport when it is put into service? Many times the products are inherrently hazardous. B31.3 explicitly allows occasional operating pressures as high as 133% of the design pressure. In the corroded condition. Would you argue that air at 110% or 120% in the uncorroded condition is safer? I'm not saying that a pneumatic test doesn't have real concerns which must be addressed (see my response a few minutes ago in thread794-372956). But why would you be happy with say propane at a higher pressure if you are concerned with air at a lower pressure?

 

[zdas04]

jte,
A kg/cm^2 is 1.02 bar. Bar is based on energy terms like Pa is. kg is a mass unit like lbm is. Because pound is a mass unit used as a force, you need "g_c" to do the unit conversion in a lot of equations. That silliness has been an embarrassment to people who used Imperial units since WWII when the Slug fell out of favor. The SI guys have sneered at me over that factor in all of the classes I teach. People who what to use kg/cm^2 deserve to have to develop their own version of g_c

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. —Galileo Galilei, Italian Physicist

 

[BigInch]

No, since a kg_mass/cm2 makes no sense whatsoever, "Kg/cm2" implies it is actually Kgf/cm^2 using kilograms-force, hence 1 Kgf/cm2 = 0.980665 bars

g = 9.80665 m/s2
1 Kgm x 9.80665 m/s2 = 1 Kgf
1 Kgf/cm2 = 9.80665 Newtons/cm2

9.80665 N/cm2 / (10 N/cm²/Bar)
=0.980665 Bar

I think you will find that this is the same result as given by Katmar's Uconner conversion program.

you must get smarter than the software you're using.

 

[BigInch]

BTW per B31.3
Paragraph 345.5.4
The test pressure must not be less than 1.1 x design pressure, nor exceed 1.33 X design pressure, nor cause stress in excess of 90% yield strength of any component at the test temperature.

UNLESS, per 345.2 C subparagraphs 1 & 2, the following conditions exist.

If both the process, or piping linings would be subject to contamination and where a pneumatic test would present an undue hazard of possible release of energy stored in the system, the Alternative Leak Test may be used, as stated in 345.9

345.9 requires a sensitive leak test in accordance with para 345.8

345.8 (a) only requires testing to the lesser of 15 psig, or 25% of design pressure.

you must get smarter than the software you're using.

 

[zdas04]

BigInch,
You are right about the number, I went from bar to kg/cm^2 in Uconeer instead to the direction I should have gone.



David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. —Galileo Galilei, Italian Physicist

 

[LittleInch]

bck,

Make sure if you follow the 345.1 C) and 345.9 route that you read the relevant section and understand that it needs ALL welds - including any in the pipe manufacture - radiographed, UT inspected or liquid penetrant / MPI. Also the sensitive leak test is a bubble test with a stated sensitivity, not just a pneumatic test to 15 psig...

Yes pneumatic testing should not be undertaken lightly and you need to work out a safe and practical way of doing it, but 40bar is not particularly high, but size of pipework has some bearing on it.

Also be aware that in 345.1 it is the owner who decides. There may also be some regulatory and inspection hurdles to overcome before so you story about both a hydrostatic test and pneumatic test being not possible or presenting an "undue hazard" had better be written down and make a compelling argument.


My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[zdas04]

For an article on pneumatic testing please see Pneumatic testing of pipelines as an Alternative to Hydrostatic Testing at ENGINEERING.com.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. —Galileo Galilei, Italian Physicist

 

[BigInch]

Not that I'm a fan of pneumatic testing, but with rigorous safety procedures in a well controlled environment, it is allowed.

you must get smarter than the software you're using.

 

[bck2ftr]

Thank you LittleInch and BigInch for your helpful information.
I wonder how I missed out on this clause.

 

[BigInch]

99 out of 100 times it is because you are searching, rather than reading the code in its entirety. right?

you must get smarter than the software you're using.