Hydrostatic testing of fittings and valves 1

[JMcCartney]

Under B31.8
Are these statements facts or misunderstood application of the code
1. Hydrostatic testing is for pipe integrity (the physical component not the system) and fittings don’t require testing because they are tested by the anufacturer.
2. Valves do not need to be tested as they are tested by the manufacture.

I have read through the code but it doesn’t seem to be black and white and is subject to interpretation. I have worked around piping in conpressor stations and have always heard this but nobody (Engieers and working professionals) can actually site the code section.

Please advise

 

[LittleInch]

JM,

I'm not quite sure what you mean by "fittings".

Section 841.3.1 states that "All piping systems.." shall be tested except for pre-tested fabricated assemblies - see below.

Once you tie-in those non welded (i.e. flanged or screwed) connections you then need to test those connections to a leak test of the max pressure available.

"Fittings" tends to mean fabricated items such as flanges, elbows, tees etc. These are not pressure tested by the manufacturer as they are normally included within the pressure test of the piping system.

Valves and fabricated assemblies which will cover special fittings like say sampling equipment or dosing pumps will be classified under the "pre-fabricated assemblies" provision, IMHO.

General Provisions. All piping systems
shall be tested after construction to the requirements of
this Code except for pre-tested fabricated assemblies
and welded tie-in connections where post construction
tie-in testing is not practical.

Now whether you include valves (normally partly opened) in the pressure test varies between parties. I've seen systems made up of lots of pre-tested spools and valves which are then finally all bolted together or with a minimum of "golden" welds and then leak tested and I've seen entire systems built and then all pressure tested as one single item with everything installed and occasionally the same, but with spool pieces for the valves.

Personally I think the grief factor of having to install dummy spools and then un bolt them and then connect in the valves then leak test is higher than just testing the whole lot in one go. However you have a choice.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[JMcCartney]

Littleinch,

I appreciate the reply...

Yes, my definition of Fittings is as you stated. Flanges, elbows, Tees, etc.

I am working in the Northeast part of the United States. I guess we need to define hydrostatic testing and leak testing.
My "simple" definition of hydrostatic testing is to expose the pipe, piping assembly (pipe and fittings), or piping spool to a pressure of 1.5x MAOP (not to exceed the fittings) for 4-8 hours (depending). This can be a prefabricated piping spool or the full construction of the system. After that you make the minimum allowable tie in welds, perform a radiographic (X-ray) examination and then turn in service.

And leak testing would be to inspect the connections (weld and bolt) once it is in service?

But I have seen it to often that you are allowed to install a fitting between two pre-hydrostatic tested spools and not subject it to 1.5x MAOP. Or drop a Tee or flange assembly in a line that is already in service. It would just be exposed to normal operating pressure of the facility. (It would be x-rayed before service)

All this seems to be common among all companies both DOT regulated and Midstream operating in this area. Just trying to tie this practice back to B31.8 for justification.

JM

 

[LittleInch]

Ok,
That should then be covered by the part a of 843.3.1

a) The circumferential welds associated with connecting
pretested assemblies, pretested repair pipe
lengths or sections, and welded tie-in connections not
pressure tested after construction shall be inspected by
radiographic or other accepted nondestructive methods
in accordance with para. 826.2.

Now I would take that to be a single weld between two sections subjected to hydrotest, not two welds on something like a tee.

Therefore I don't believe there is the justification in the code for what you are looking for as this seems to be, on the face of it, poor practice which has been allowed to develop over time.

The code to me is quite clear - you need to test "All piping systems". Now this can be in one go or in parts but all sections exposed to the pressurised fluid should be tested. The code allows for (limited) welding or jointing to be tested without hydrostatic test as it is not always practical to test entire systems in one go, but these should be mimimised.

There is specific reference to pre-tested pipe and I would take that to also refer to pre-tested fittings if these are inserted into the a system which has already been tested.

Hope this helps.

LI

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[moltenmetal]

My comments assume B31.3 pressure piping is being discussed.

You are permitted to bolt two flanged pipe spools to one another, and if each has been independently hydrotested, you don't need to re-hydrotest the bolted assembly. You don't need to "prove" the flange (be it in the run or a blind flange), the bolting or the gasket at a pressure higher than design.

You can drop in a flanged valve between two pre-tested pipe spools, bolt it up and use it, without hydrotesting again. The valve is assumed tested by the manufacturer. The same goes for a pump or other piece of pressure-retaining equipment which is either exempt or separately tested.

Any weld between two things that retain pressure must either be hydrotested or subjected to a special process of NDE which may eliminate the need for the hydrotest. Only in category D (nonhazardous) services may the hydrotest be waived in favour of an in-service leakage test, and that's only with the owner's permission.

Leakage testing of flanged joints post assembly is something not addressed in the pressure piping code, but something which obviously needs to be done. Whether it requires a special test with another fluid or can be accomplished via an in-service test depends on what your fluid is and what the consequences will be if it leaks. It may be the subject of other codes and certainly should be part of the operating company's standards and procedures.

Does this apply to piping joints of types other than welding? In my opinion, yes it does. We hydro or pneumatically test every threaded and tubing compression joint in any fluid service other than category D, because that's how we read the code- all piping joints must be tested. And surprisingly, we have had to argue with some of our client/owners, who feel that this isn't necessary and don't do it themselves...

 

[JMcCartney]

I appreciate your feed back. Do you know if any good text books that explain the B31.8 code in greater detail and it’s application?

 

[GD2]

JMcCartney,
Look this way - hydrostatic testing is required for proofing both leak and integrity of the the entire piping system that includes pipes, fittings and valves including pipe supports, if there will be any. Leave out the equipment for simple understanding.
The reason behind it is that even if the pipes (used for pipelines), valves and some pre-fabricated sub-assemblies are pre-factory/shop tested, welding joints NDE tested, not every joints may be 100% NDEed (depending on class location), pipe supports not tested for loads and flange leak proof not guaranteed (for those joints interconnecting pre-tested components/sub-assemblies). The best way to avoid these uncertainties is to do a hydrostatic test and get a piece of mind. No matter how the code writes and how you interpret, the client will like a system that is safe and reliable.

GDD
Canada

 

[DekDee]

I agree with GD2's last statement.
"No matter how the code writes and how you interpret, the client will like a system that is safe and reliable."
The code gives minimum requirements - the clients specifications will often add additional requirements.
If you are the contractor it is in your best interests to give the client (if you want more work) a leak free piping system.
If they have failed to request a system hydrotest (and are happy with only sub-assembly testing) then the call is yours, obviously at more cost to you, the contractor.
Cheers,
Shane