Pneumatic Test after Hydro & Stamp - Code Implications? 3

[ConstantEffort]

I'm the owner of a vessel designed to 100psig per ASME Sec 8 Div 1. The 3rd party fabricator hydrotested at 130psig, stamped it, shipped it to me.

Now I have B31.3 piping attaching to the vessel with a design pressure of 100psig. For process and logistical reasons, hydrotesting is not practical. I desire to pneumatic test the piping along with the vessel at 110psig (i.e. 10psig beyond the stamped MAWP).

Questions:

1) Aside from the obvious danger of pneumatic testing, is there anything about this that just screams unsafe?

2) Is this considered by any ASME, NBIC, or API code/standard as a test of the vessel, permitting me to exceed the stamped MAWP?

By strict interpretation of ASME I don't think so... but I am not familiar with NBIC or any "field hydro" operations.​

3) Am I still subject to Section 8 Div 1 paragraphs concerning pneumatic testing, like UW-50's requirement for MT/PT of welds at openings?

It would seem that these are for the purpose of detecting cracks prior to a hazardous pneumatic test. Having already hydrotested the vessel at a higher pressure, shouldn't I be reasonably assured that such cracks do not exist?​

4) Does B31.3 prohibit such a test configuration by omitting a pneumatic equivalent of hydro's paragraph 345.4.3 concerning hydrotesting of vessels with piping?

After ensuring that my crews will follow a vetted and thorough pneumatic test procedure, I am inclined to allow this course of action even if I can't point to a paragraph in a code that says this exact situation is permissible.

 

[moltenmetal]

I don't see any code reason to not permit you to do what you suggest, though I defer to those who have these codes tatooed to the inside of their eyelids which I do not. More importantly, I do not see a valid safety reason not to do what you suggest. We frequently follow hydrotests of piping (intended to demonstrate integrity) with pneumatic tests intended for more sensitive leakage detection.

To my mind, the hydrotest of the vessel has removed most of the hazard of such a pneumatic combined test of vessel and attached piping, assuming the vessel has the larger volume of the two of course.

Any pneumatic test requires a pre-test safety evaluation, a properly developed procedure and trained staff. A calculation of the stored energy should be part of evaluating the test's safety, i.e. both the pressure and volume of the system under test matter to the safety of the test.

 

[zdas04]

As moltenmetal indicated, there is no reason not to do what you planned. The pneumatic test would not allow you re-rate the vessel (e.g., if it is stamped with an MAWP of 100 psig, a 110% test to 130 psig would not allow you to re-rate the vessel to 118 psig).

I published an Engineering.com paper that might help understand the risks better. What you are proposing sounds perfectly reasonable to me.

[bold]David Simpson, PE[/bold]
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

 

[SNORGY]

Can the vessel not be isolated from the piping during the test, or are the nozzle connections welded? People used to do that all the time with ported test blinds.

 

[zdas04]

Even if you can isolate the vessel, you have to ask "why should you?". There is little downside to taking a 100 psig MAWP vessel to 110 psig beyond needing to isolate or remove the vessel PSV which is likely set at 100 psig.

[bold]David Simpson, PE[/bold]
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

 

[ConstantEffort]

Thank you all for your helpful responses!

I used 100psig as a nice round number. The actual design pressures are higher, up to about 1500psig, with correspondingly higher hydro and pneumatic test pressures. I figured someone might balk at my proposal due to the pressures involved rather than the logic of pneumatic at a lower pressure after hydro at a higher pressure.

Pneumatic tests are how my industry operates, unfortunately. Even an ounce of water remaining in the system would cause hundreds of thousands if not millions of dollars in damage during startup. Our installation sites are remote and lacking in water treatment facilities so that sufficient quantities of water would require a fleet of trucks to bring the water in and also to dispose of the water in the end. It's a situation I only reluctantly accepted after discussions with the old timers who came before me, but I have been working constantly to mitigate the dangers since.

 

[ConstantEffort]

RE: Isolation, for Snorgy and Zdas...

Isolation of the vessel is possible in many, but not all, cases. Where it is possible, I do recommend that to my crews. They still prefer to test through vessels because it is expedient: more piping under a single test, and it is the only practical way to test some of the vessel trim connections with an inert gas before introducing the more hazardous working fluid.

 

[weldstan]

1500 psi is a far cry from 100 psi. You need to perform calculations to determine applied stress relative to specified minimum yield strength to adequately determine safety.

I would recommend isolating the vessel from the pneumatic test. Treat the tie in weld like a "golden weld" and perform RT on it after completion to comply with B31.3.

 

[moltenmetal]

The calculation of stored energy of a vessel at 1500 psig is likely to convince you that the pneumatic test is a bad idea, unless the vessel itself is very small indeed.

 

[EdStainless]

The last one of these (HP gas testing) that I saw done went smoothly, but they had a lot of experience.
The gas used was not air but high purity nitrogen and the pressure was about 1,000psi.
There was not a person within 1,000 yards of the unit under test, and any equipment closer than that had been shut off and inerted.
In the day of wireless video (Go-Pro) this is much easier. They had about 25 cameras around the unit watching valves, gauges, and other items. I asked about watching gauges when they have DCS reading them, they said that if a signal fails during test they can't go in and check without depressurizing, so they use 'remote visual' as a backup.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube