Pneumatic Pressure Testing

Don't get the soap in your eyes.

Yes, don't stand near the pipe....

...and read this....

http://www.engineering.com/DesignerEdge/DesignerEdgeArticles/ArticleID/8180/Pneumatic-Testing-of-Pipelines-as-an-Alternative-to-Hydrostatic-Testing.aspx

Compressed air is dangerous. How much pressure are we talking about? What is the system being tested?

Leaks should be deducted during the pressure test by observing pressure from a save position.

Soap solution could be use at low pressure to determine where the leak is but it depends on what you are doing.

Pneumatic test of 1" line at a pressure of 13bar - reported observation , instead, without reducing pressure to the design pressure, prior to inspection, the teams would continue with the line inspection, supposedly all other joints were leak free.

Is there any safety concern : reduce pressure to the design pressure, prior to inspection

Yes there is a safety concern.

Nobody will be liable here if he tells you its not.

Observe pressure test. Look for pressure decline. Expect increase/decline with temperature changes. Same pressure + temperature drop might still be a leak.

Stay away from the lines when pressurised. 13 bar could be fatal.

What is the application? Whats the fluid in the lines? What does the regulation say?

Did you read the link from SE?

What is the design pressure? Quoting this without a number would have to make us mind readers to follow.

If in doubt hire a specialised engineer who knows what he is doing. Make a proper risk assessment.

The concern is kinetic energy of pressurised air in the system which could rapidly release and cause serious damage and death.

If you are saying that you want to perform pneumatic testing on a pressure vessel instead of the accepted hydrostatic test, don't. It's dangerous, against code, and won't give reliable results (hence being against code).

There is no safety procedure for doing something you shouldn't be doing anyway.

The old soap test ploy.....

conduct a risk assessment to determine the correct PPE before moving onto point 1.

1. Make sure you have the correct soap to water ratio

2. You have a suitable applicator for the mix you wish to apply to the pipework

3. ONLY put the mix on joints that are cold, and not just been welded or soldered in the case of gas pipes.

4. Set your air regulator to a low pressure.. you determine that bit (you've got to do some part of this)

5. once you are happy you've reached your stable max test pressure, shut the inlet valve

6. Wait for the bubbles to appear

7. No bubbles? Move on and repeat 1 to 7.

Pneumatic test of 1" line at a pressure of 13bar - reported observation , instead, without reducing pressure to the design pressure, prior to inspection, the teams would continue with the line inspection, supposedly all other joints were leak free.

Is there any safety concern : reduce pressure to the design pressure, prior to inspection

Do not carry out a pneumatic leak test without having done a structural hydraulic test to the same or a greater pressure beforehand.

To PWSlack.

I am out of touch with today's installation practice having been retired for 15 years. But during my previous 50 years at work, I have never carried out or come across a compressed air installation that had a hydraulic pressure test.

Individual hydraulic tests on pressure vessels for proof testing definitely yes, but not once installed and piped up.

Filling the system with water would have been disastrous for the equipment.

Air pressure was applied slowly, and checked for leaks. That's all. Albeit, fixing leaks was the real problem, finding them was easy.

So what happens today, are hydraulic tests required?

One example that is not a recent development would be that pressurized cylinders that falls under DPT purview are generally required to have a hydrostatic test every 5 years.

.

I'm a little fuzzy on why a pressurized air system would be damaged by the intentional filling and subsequent removal of water. The drier pretty much work all the time, indicating water is being removed, ergo sum it is in the system to be removed.

.

Certainly certain equipment could be damaged by moisture in the line, but that equipment should not be receiving anything from the line during the test.

.

Hydrostatic testing is simply so much safer than pneumatic testing, it is certainly worth the extra time required to drain and purge the lines.

I want to avoid the possibility of a misunderstanding arising here - I am not suggesting hydraulic testing is not required - it might well be today - but that is a problem for today's engineers. My comment was in the context of my past experience.

I would never put water in the system. It gets everywhere I can assure you. Isolating sections of the system to avoid water effectively meant the 'system' was not tested.

But then, my systems were all relatively small and mostly below the PV limits of mandatory testing anyway - and when tests were required it had to be done with compressed air - it was not for proof burst pressure but for leaks - and carried out in carefully controlled conditions.

Horace,

Without hydraulic test certificates you can't get insurance in the UK for receivers etc. It has been that way since I started in engineer 40+ years back.

TonyS

Yes, you are right. I cannot recall a time since starting work back in the 50's when receivers and pressure vessels did not require a test cert. As far as I know, it has been a requirement to prevent steam engines blowing up stemming from the time railways were built

But in the context of this post, and in line with my previous comments, and today's rules, all products used in an installation will have the appropriate pressure test approvals and all the certificates to go with them - and where appropriate, on-going certs as required by law (and product warranties ??)

Insurance companies can impose their own rules I suppose. Although in my experience they do not do this. They rely on compliance with legislation and take your money, only to duck out of paying up in the event of an accident if you can't prove compliance.

Where in the case of an accident causing personal injury in the UK - you have the HSE to convince first.

I have never come across any law that specifies an hydraulic test on a complete compressed air installation.

On what? Pressure tank? Piping system? Hydraulic system? Water? Air? Liquid hydrogen? Nitrogen?

Type of material? Steel? Iron? Plastic? Glass? ?

Where? India? USA? Greece? Canada? (It makes a difference.)

In the USA ASME Code Section 8 covers some of the test requirements for pressurized vessels.

However there are many other sections of the code which are application specific that must be adhered to for safety reasons.

You must be familiar with all test methods and procedures that are required by the governing body where the application is located.

Don't pneumatic test it. First reason is air is a compressible material and if you have a failure things don't stop flying until all the pressure is out of it. Do as they do on the installation and testing of natural gas pipe lines. Fill it with water, remove all the entrapped air, pressurize the water to the test pressure, valve off and plug the connection to the water and the pipe, pressurize the pipe and monitor the pressure for test time length. If you can't get it up to pressure you have a small or up to a large leak (look for the water leaking). If the pressure slowly drops you have a very small leak. If it maintains pressure there are no leaks and you can do a "happy dance"!

If it passed the test, dewater it, clean out any residual water and put the line in service.

With a compressed air leak, almost any leak will probably increase in size and blow apart. This is why pressurized water is the favored system, "fewer injuries and fatalities".

Good Luck, Old Salt

Pressure Equipment Directives apply to the UK and Europe. Minimum PV limits apply (explosive release of stored energy). Above which a CE Mark is required by law. You use CE marked products. You install all equipment in accordance with the appropriate standard. Logically all that is left is the integrity of the joints - which usually will be leaks.

An underlying principle of these Directives is the caveat 'where reasonably practicable' - and filling a small scale compressed air system with water, and consequential draining and drying, definitely isn't.

Risk assessments by competent persons, and suitable procedures, can be used to find safe ways to test a system with pressurised air - to find leaks.

Over-pressure to proof test burst pressure is definitely a hydraulic exercise.

I can't talk for natural gas pipelines, but I guess hydraulic tests are essential - even if only to find leaks. We don't want any gas leaks at all.

Interestingly we had to test a 'plastic' pressure vessel for burst pressure, where the object of the test was find out how high the pressure had to go, and when exploding, to see how well the surrounding mesh safety screen restrained any fractured parts. And that required a compressed air test. All conducted safely inside an enclosed steel container.

The words "usually", "reasonably practical", "definitely isn't' are very ambiguous. If the customer wants you to test something with the procedure they specify you do it that way. It may be over kill but that isn't for the contractor to determine. Just do it his way. If they want a hydro test on a compressed air line instead of the pneumatic test you feel is ok you do a assessment of it and do it his way if it can be done safely some other way hydraulically. You think things out fully, do risk assessments, haz ops to find a way. These are the things that prevent more pneumatic tests from being done

Having seen some pneumatic test failures I wouldn't do one unless it was the safest way possible and the hydro test wouldn't be anyway suitable. Pneumatic test failures generate shards, projectiles and other nasty things with almost continuous pressure behind them. If a hydro test fails you get a puddle of water from a very quick flow of a small amount of water.

Can you imagine it if a 42" natural gas pipeline was pneumatically tested and had a failure? Same thing that happens when one of those catastrophic natural gas leaks happens. Metal everywhere including in people. It is also the same as when a large container such as when a reactor is pneumatically tested. BIG BAD BOOM!

Good Luck, Old Salt

"Definitely isn't" definitely is unambiguously unambiguous.

Ambiguity of meaning allows judgement to be exercised and progress to be made. The wording might be vague and open to interpretation, but UK law here is absolute.

Regardless of paying attention to every detail, and every control measure, and every precaution to prevent risk to health and life, an accident resulting in injury to a person is clear evidence of failure to comply with the law.

Very likely the application of the appropriate standards and codes-of-practice will prevent accidents, but if not successful, you will end up liable, but their use can act in mitigation and consequent damages.

Which leads to cost of compliance, and whilst it is permissible to use 'cost' as a decision tool in the choice of the means of compliance, it is not permissible to avoid cost by omitting the essential safety requirements or the control measures.

You might say the ambiguity tests the resolve and character of the engineer, and since it evolves to an 'argument' between engineers and accountants, you get a situation where the engineer usually loses, and has to find the 'cheapest' solution.

horace40-

Not being intimately familiar with the UK laws and standards as well as someone from there. I do know that in this here United States the are legal and there are "common sense laws". There is no "legal system" law that says you can't try to fly off a cliff, go scuba diving down to 140 ft with only a 30 minute bottle and no NITROX, run in front of a charging elephant (except maybe harassing wild life), water your lawn every day (except in drought conditions), and so forth. There are no laws about these so how can they be broken? The "Law of common sense" is applicable for these and similar conditions. If I attempt to clear an obstruction from my lawn mower and amputate part of a finger there is no legal law but the common sense one is there.

DuPont years ago, as part of their S.T.O.P safety program, did a study of the frequency of accidents. They found that for every true accident (one recordable as per OSHA standards) there are 330 potential accidents. Of these there are a lot that have no law against doing them. OSHA has one part which is generally called something like the "General working conditions". Risk assessment and Haz Ops prior to the incident and behavioral modification are what does the major part of preventing accidents. Unless it is an act of God, there really isn't any accident that could be prevented. Take that 1:330 and make it a 1:660. Don't stop there, make it 1:1,320 or even higher to keep everyone safer.

The ambiguity that causes engineers and bean-counters to "argue" is not an insurmountable problem. If engineers stopped alking "engineering" (used to be called pocket protector or slide rule geeks) and the bean counters stopped putting $$$'s in front of everything and communicated with an amiable solution as their goal it is usually safer, less costly, and quicker.

29CFR1910.120 tells me I have to use a level "A" protection suit for certain haz mat conditions. It doesn't tell me I have to use chem tape to seal the arms to the gloves. Neither does it tell me whether the glove goes inside or outside the sleeve. In almost every situation these things are resolved for before hand. Have the risk assessment figure it out beforehand. Don't depend upon the laws and the "sharks" lawyers to sort it out. Don't let them happen in the first place!

Good Luck, Old Salt

To Old Salt #23

PS to my post. I should have made it clear that chatting in general about the laws here, we are largely bound by the Health and Safety at Work Act.

Domestic DIY jobs in general are outside the scope of the Act. But hopefully commonsense prevails.

And where I said cost, it is not just money for damages, breaking the law could result in a criminal charge. A fine, a disqualification, or In the extreme you could end up in prison for an injury or death resulting from a breach of a Prohibition Order.

But probably wandering a bit OT. The basic question stems from trying to do it right. The advice we give might differ, but I think we are in the same choir and we all join in the same health and safety chorus.

In the UK we have the term ALARP. A line must be drawn, or anyone could be accused of criminal liability for, e.g. serving a drink in a glass container (which could break, and so injure someone), or building an edifice with a story above ground level (someone could fall from a window, or fall down the stairs).

ALARP is most frequently used in connection with radiation protection, but the principle can be applied generally. It could be described as ambiguous, but ...

If everything was made 100% safe, nothing much would happen (except for all the people dying in bed).

That doesn't sound like '100% safe' to me. Layered risks from things like sedentary lifestyle, diminished in-person social interaction, and early onset boredom; surely would pose significant risks. What of the risk of missing out, or risk of regretting never having had a particular experience?

.

It all becomes very equitable, if we just take a wider view. Specifically if we widen, by an order of magnitude or two, our perspective on time in the sense of an arbitrary span by which we consider lifetines either short or long, then the ultimate risk for everyone becomes equally probable.

Perhaps not really all created equal, but it gets resolved by the end.

Depending on the location of the piping to be tested, and the duration of the test, the effect of temperature must also be considered.

A 24-hr test will show a decrease in pressure during cooler outside temperature (such as after sunset), so the ambient temperature as well as the the pipe temperature (thermocouples) should be part of the data recorded so that a slight decrease in pressure in not necessarily interpreted a a leak.

In Australia, this is what you should normally follow:

"Australian Standard™ AS 4037-1999 Pressure equipment-Examination and testing

S E C T I O N 1 8 P N E U M A T I C T E S T S

18.1 APPLICATION Pneumatic testing should be avoided, but may be used in place of the hydrostatic test in special circumstances …

18.6.2 The pressure shall then be reduced to a value equal to 80 percent of the test pressure and held at this pressure for sufficient time to permit detailed visual inspection of the vessel." i.e. don't inspect at test pressure! Buy the standard for the full story.

Treat pneumatic testing with respect and use a soapy water solution that you know works i.e. I've seen some learners/shonky operators use solutions akin to old washing up dishwater. Going on memory, we used to use 1 part glycerine/glycerol, 1 part detergent and 4.5 parts water (please confirm for yourself) for big jobs otherwise use proprietry solutions/sprays.

Regards

Malcolm

If you are doing pneumatic testing, or checking for leaks in a gaseous carrying pipe, in temperatures below freezing add some Ethylene Glycol, automotive antifreeze, to the soap solution. Unless it is extremely cold 1/4th to 1/8th of the solution is good. The glycol does not detract from the generation of the bubbles.

Good Luck, Old Salt

Follow the pressure testing code applicable to the local jurisdiction or imposed by the burst/collapse indemnity insurer's requirements. In the absence of an applicable code, above 215 bar-litres a wise Engineer would hydraulic test above the design pressure beforehand, except in exceptional cases such as vehicle tyres.

• One day recently, after dark, a team of technicians were pressure-testing a 600mm diameter line 1.5km long using compressed air. As the pressure rose towards the test limit, the end of the pipe blew off and demolished a scaffold. The escaping air flung a piece of scaffold pipe 500m towards a building, which building it entered end-on via the roof and impaled someone who was in bed sleeping. Needless to say the individual did not survive.

'...Needless to say the individual did not survive....'

.

A relative of mine was the surgeon on call when a construction worker fell several stories to land on and be impaled by thick rebar (in around his hip, where emergency crew sawed off the rebar) thwn up through his torso to exit the opposite shoulder.

.

He live to tell the tale. He of course spent a fair amount of time in recovery.... so a lot of people got to know him and he seemed to enjoy that everyone called him by the nickname he received in the ICU...'Spike'.

Anyway, the point is, it wasn't less necessary to describe the outcome of the injured person. Conclusions are rarely as foregone as people like to think.