hydrostatic test duration for b31. 3 prefabricated spools 3

[unclebensrice]

Dear all,

Our company made an order for thousands of prefabricated pipe spools. The purchase order stipulates that the spools shall meet b31.3 criteria. The fabricator says he can only pressure test the spools for 1 minute as opposed to the requisite time of 10 minutes as per b31.3. Can this reduced test time be accepted ? Thanks.

 

[XL83NL]

The owner has ultimate responsibility and may waive (or add) requirement in (to) B31.3.
Why cant he do a hydro for 10 minuten, but only for 60 seconde?

 

[unclebensrice]

hi XL83NL,

The fabricator says he has a machine which can pressure test only one individual spool at a time and so 10 minutes per spool multiplied by thousands of spools will undoubtedly delay delivery time. He says 1 minute per spool is enough and still have b31.3 compliance. I don't know really.

 

[unclebensrice]

So if a spool is pressure tested for 1 minute will it meet b31.3 requirements ??

 

[XL83NL]

No, refer to para 345.2.2(a). Ten minutes.
However, the owner may waive that requirement. You then need his permission.

 

[Duwe6]

Talk to the Owner's Inspector -- per B31.3 the Owner is required to have an Inspector. My usual take on "10 minutes" is that 10-minutes is up when I have laid hands on all welds and found them to be dry. So if the spools are fairly simple, and the Inspector is there for each hydro, 1 to 2 minutes seems to be a reasonable time; but it is up to the Inspector.

 

[unclebensrice]

Thanks duwe 6,

I'm the owner and I'll be having a third party inspector assigned for this project. Yeah the spools are just a meter in length. I'm kinda leaning towards 1 to 2 minutes myself but my inspection team and my boss insist on the 10 minutes saying that's what the code says ! I mean in the here in field we test our pipelines for hours not 10 minutes due to extensive length and so if we extrapolate that for spools then 1 minute is enough. A leak should be instantaneous right and plus the weld joints will be right there in front of my TPI so he won't need a full 10 minutes to do a walk around.

 

[BigInch]

I think BOSS is the owner.

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

 

[weldstan]

Will you not be leak testing the completed piping systems? If you will be, that is when the 10 minutes applies. Note that "golden welds" do not apply to thousands of welds. Your boss may well be spending the company's money unwisely.

 

[unclebensrice]

Yes the completed system will be leak tested in the field. The spools once they arrive will be part of the completed system. The purchase order for these spools mentions b31.3 our design engineer should have specified the holding time also but he didn't so now our division is left to enforce the purchase order my boss says wants to adhere to the purchase order religiously.

Even the design engineer says a spool on its own is not a piping component it is a piping system therefore 10 minutes per spool !!

And on a different note you can have more than one closure/golden weld.

 

[weldstan]

It is standard Engineering practice to specify B31.3 to a Pipe Fabrication shop for pipe systems designed, constructed and tested thereto but not to require hydrostatic testing of the spools. If it was not specifically stated in the P.O., if I were the Fab Shop Owner, I would back charge you big time based on customary practice and per 345.1, "each piping system shall be tested to ensure tighteness." A pipe spool is by no stretch of the imagination a piping system, especially when thousands of spools are involved.

And I do understand that there can be more than one closure weld.

 

[LittleInch]

How many spools can you bolt or clamp together in one go to hydro? I've done this many times where you bolt up multiple spools just to hydro to reduce overall testing time as you have to allow for all the filling, draining venting, emptying etc as well as your 10 minutes per test and then just unbolt them all again to transport / fit in position. If they're only a metre long you should be able to get 10 or 20 or more bolted together in one go or even just join bolted spools together with a pressure hose. If you've got to bolt on two blank flanges for each one then it's just as fast if not faster to bolt spools together as you only need one bolt up per spool apart from the end spools.

My motto: Learn something new every day

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

 

[unclebensrice]

The spools are not flanged at the ends. They are made up of api 5l pipe pup pieces welded to b16.9 elbows mostly. The fabricator would like to avoid joining multiple spools together by welding just to leak test then cut them separate again to him it's a waste of material not to mention extra welding and cutting.

 

[DekDee]

unclebensrice,
You (or your boss)are about to make a huge mistake which will cost your project huge amounts of unnecessary time and money - purely because someone (specifically your design engineer) does not understand the applicable code.

"Even the design engineer says a spool on its own is not a piping component it is a piping system therefore 10 minutes per spool !!"


Here are some excerpts from B31.3

erection: the complete installation of a piping system in
the locations and on the supports designated by the
engineering design including any field assembly, fabrication,
examination, inspection, and testing of the system
as required by this Code.

piping components: mechanical elements suitable for joining
or assembly into pressure-tight fluid-containing piping
systems. Components include pipe, tubing, fittings,
flanges, gaskets, bolting, valves, and devices such as
expansion joints, flexible joints, pressure hoses, traps,
strainers, inline portions of instruments, and separators.

piping subassembly: a portion of a piping system that
consists of one or more piping components.

345.2.3 Special Provisions for Testing
(a) Piping Components and Subassemblies. Piping components
and subassemblies may be tested either separately
or as assembled piping.

Notice I have placed the emphasis on may


If you have gazzillions of dollars and a couple of years up your sleeve you may test each spool individually but the code does not require it and never has done.

Regards,
DD

 

[LittleInch]

"The spools are not flanged at the ends.". So how exactly is the supplier pressure testing them? I agree with dekdee that these items are not a piping system and hence should not need individual test certs for each spool.

I am kind of intrigued to figure out what item needs "thousands" of small 1m long spool pieces pre-made.

My motto: Learn something new every day

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

 

[unclebensrice]

Hi guys,

First of all, a piping (sub)-assembly is not equivalent to a piping component. Please refer to the answer of question #1 of b31.3 interpretations volume 23.

Second, most of these spools consist of pipes with beveled ends welded to elbows with BW ends. They will be used for flowlines at road crossings.

Some spools have one or two flanged ends out of three ends and those spools are for tee branches to be used in the flow lines.

The order is for almost a thousand spools.

The hydrostatic test procedure is unclear. It says "the coupled system will be filled with water and then after the pressure is raised all connecting pipe excess those for pressure measurement shall be disconnected."

So what do you think now 1 min or 10 min

Thanks for all your valuable input fellows. Appreciate it.

 

[unclebensrice]

Sorry for my above post I forgot to mention the interpretation number
It is Interpretation # 20-24, question #1 of b31.3 interpretations volume 23.

 

[moltenmetal]

Any assembly which is "plain end" is pointless to hydrotest for any duration, since it will need to be re-tested after the completion weld is done. The alternative is very careful NDE (plus a sensitive leakage test?) which may not be less effort than a hydrotest.

If the service is Category D (nonhazardous) per B31.3, the owner may elect to test the piping in service with the service fluid rather than hydrotesting, if I recall correctly.

Any spool terminating in nothing but flanged ends may either be hydrotested or pneumatically tested in a code-compliant way (10 min) now and then NOT re-tested post assembly, or it may be tested with the assembled system.

It is not necessary to re-hydrotest flanged spools which have been individually hydrotested, after the flanges have been bolted together, merely to test the integrity of the flanged joint. However, it is obviously prudent that these flanged joints be leak tested prior to being put into service. That leakage test has a different test design, different conditions, and totally different hazard profile than a hydrotest.

 

[DekDee]

unclebensrice,
I will try again.
It is quite clear a piping component and a piping subassembly are not the same as noted in my earlier post.
The interpretation you quoted is based on the 2002 edition.
The 2002 edition does not list piping subassembly in 300.2 Definitions and that would be why the query was submitted and interpretation raised (subsequently added for future editions).

You have a pipe (one component) and an elbow (another component) so they form a piping subassembly - which is then a portion of a piping system.

This is the definition of "piping system"
piping system: interconnected piping subject to the
same set or sets of design conditions

Is the pipeline running either side of the road crossing subject to the same set or sets of design conditions as the spool being used for the road crossing ?

Cheers,
DD

 

[unclebensrice]

Hi gentlemen,

Just to answer the questions above :

- our service is normal fluid service not category D.
- the reason we request even flanged end spools to be hydrotested during their manufacturing even though we can just hydrotest the completed system once during field erection as per 345.2.3 is because we are purchasing these spools from abroad. We are not manufacturing them in house so it's kind of a precautionary measure. The purchase order asks for hydrotest.
- yes the the pipeline running either side of road crossing is same as the spools design code which is b31.3. 3.

Thanks again guys.