Does a 600# flange need to be tested? 2

[Bigfaz]

New 24" natural gas pipeline in WV, USA.
Inspection had me remove all 600# flanges and weld temporary test heads on prior to hydro of a section of pipeline. Design pressure is 1480psi, material is X65. Hydro pressure was 2400psi due to elevation difference. Inspection said the flanges are tested at manufacture site. I don't agree. I was under the assumption 100% of the pipeline had to be strength tested including the flanges. Please advise, Bigfaz

 

[LittleInch]

flanges are good for 1.5 x max rated pressure. Why is you test pressure greater than this? what code and wall thickness do you have? Where is this pressure being measured?

You are essentially correct, but there is something strange here. A section diagram / sketch would help showing elevations and heights, pressures etc.

What you are testing with flanges is the connecting weld, not the flange itself. The flange is not tested per se, but forged and built to standards.

My motto: Learn something new every day

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

 

[dcasto]

So, it appears you are measuring the pressure at the bottom of a 416 foot hill and loaded with houses in the area.

I always find it quit hysterical when engineers mandate that pipelines be rated to 1480, then all the devices that create the pressure are rated to 1200 psig. So all your compressors are ANSI 900? That's what you need to get to 1480 on the pipeline. If you test to the line at 2220 at the low point, then it's rated to 1300 psig. You can operate it at 10% over that or 1430. You lost 50 psig of room. You'll never miss it and save a bundle of money up front. Think about this just a couple of decades ago, the 600 class flange was rated to 1440.

I also love it when piping designers spec sch 80 6" in plant piping in order to get a full 1480 piping, then every vessel that pipe is connected to is rated at 1000 psig. Put in sch 40 rated to 1200 plus psig and save a bundle.

Technically, they are right. You could pneumatic test 1600 or so and get your full 1480

 

[Bigfaz]

LittleInch,
the elevation difference is 400', the wall thickness is .500, the test pressure is being monitored at the lowest elevation. The minimum test pressure is 2220psi. The hydro is now complete and the line dried. The flanges were welded back on and the welds x-rayed in accordance of API 1104. The flange welds were not hydro tested. Inspection removed the flanges prior to hydro because the said the flanges would not hold the test pressure. If that is so, I think engineering would have used 900# flanges. Just my thoughts. Thanks, Bigfaz

 

[LittleInch]

Thanks for the info.

I kind of agree with dcasto - pipelines should be designed and rated for what they have to do - that's why pipeline design codes exist - not the max cold working pressure of the flanges. That always sounds to me like a piping engineer got hold of the design without realising the implications such as the one you have just faced. Having a desing pressure of 1360 (I get 1360 compared to dcastos 1300) would have made all those issues go away, but probably very little impact on the operation of your pipeline.

FYI ANSI B 16.5 states that the max system pressure test is 1.5 x the cold rating of the flange (2220 in your case), but higher is "the responsibility of the user"

Whilst the inspector is erring on the cautious side, the flip side of his decision is that you now have an untested weld on what sounds like a piece of pipe subjected to welding, cutting off, welding a test end, cutting that off and re-welding the final flange. I think I know what I would have done....

My motto: Learn something new every day

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

 

[moltenmetal]

Utterly dumb-@ssed in my opinion. It would have been far safer to leave those flanges in place and hydro them as part of the assembled system, even if adjusting the test pressure downward by putting a more realistic limit on the design pressure was not possible. I'd far rather modestly over-hydrotest a flange pair and risk leakage during the test, than substitute NDE for a hydrotest on two welds and risk a brittle failure in service! I know no rules have been broken- you are sometimes in a position where one weld cannot be hydro'd for some reason and you need to substitute careful NDE, but I would not characterize this situation that way.

I think LittleInch and I would have done the same thing.

 

[Duwe6]

what MoltenMetal said. . .

Since RT has been acknowleged to be incapable of imaging fine, tight Lack-of-Fusion that did not trap slag, 100% RT is not particularly a good method of testing. A "Full Flange" hydro, to 1.5 times the cold MAWP of the flanges is a superb method of assuring that the line is well built. If it is indeed a hydro, vs. a pneumatic, those few more 10's of seconds of running the pump are damn cheap insurance. If everything was designed and built properly, your flanges are your limiting element. Hydroing to 1.5 x Limiting Element just makes sense.

 

[Bigfaz]

First, thanks to all for their time in answering my question. Reading between the lines of all answers, I think what everyone is saying is that a DOT audit won't have a problem with the flanges not being in the hydro. Thanks again to everyone, Bigfaz

 

[BigInch]

Don't read between the lines with DOT. Do at the very least EXACTLY what the text says. All components in the pipeline must be tested after assembly, flanges must be included.

Yes, exactly Inspection removed the flanges prior to hydro because the flanges would not hold that high test pressure. Engineering should have specified 900# flanges, or reduced the test pressure to +/- 2160 psig, OR LESS. Maybe less because it is not often on a pipeline where you must test to 1.5 times design pressure, so it really is a mystery why the test pressure was set so high. Normally 1.25 x design pressure is sufficient on most parts of a pipeline, but it depends on your Class Design Factor at the test section's location. Where on this pipeline is this, compressor station, road crossing, downtown, rural area?

Right now this seems to be a major mistake by your engineers, for which if true, you need to let them have both barrels ... of a 12ga.

Learn from the mistakes of others. You don't have time to make them all yourself.

 

[dcasto]

By DOT code and b31.8, you can build a 500 mile line and never do a hydro test on the entire pipeline and components. The operator is free to pretested a joint of pipe, then weld that pipe on each end and do a 100% X-ray and then operate the line. That's how repairs are made. So nothing is wrong by just testing each piece of pipe on a 500 mile line.

However, there isn't a sane person that would accept that. If I was a DOT inspector and saw that you didn't test that flange assembly, I'd make your next few months not a very fun. It's not the fitting that's at issue, it's the weld and the bolts that are at issue.

If you can't live with less than 1480, then as stated above, put in 900 class fittings. Just like I said you'd have to do at your compressors because of the temperature derate.

 

[BigInch]

No. Not possible to build 500 miles, or any length more than a repair assembly without hydrotesting or pneumatic testing.

841.3.1

All piping systems shall be tested after construction to the requirements of this Code

Yes there are a few exceptions,

except for pre-tested fabricated assemblies and welded tie-in connections

But not always are those excepted. You can only make those exceptions where

testing is not practical.

that pretty much limits it to golden welds.

Furthermore, that work must be done in the context of,

for the purposes of repair or replacement.

While it is possible to not do hydro/pnuematic tests on limited repair pieces, it is overwhelmingly obvious it is not allowed, or permissible, or practical to construct 500 miles of pipeline without hydrotesting, or pneumatic testing under those particular circumstances which do allow pneumatic testing.



Learn from the mistakes of others. You don't have time to make them all yourself.

 

[dcasto]

Big inch. As operator you make the call on all those clauses. If I declare it impractical to test, then if you don't believe me , I'll see you in court. Read further in my post I said it would not happen. I've replace a few hundred feet at one time under the exception. We consulted with the trrc before and they agreed.

 

[BigInch]

I read the whole post and agree with the general conclusion, but stating that you do not have to hydro and that you can build 500 miles of pipeline without one was just too much for me keep silent about. A couple hundred feet is a long, long, long way from 500 miles. Perhaps you are guilty of simple exxaggurraation.

If I can do it, it's practical. If I can't, it's not. If you can't do it, I'll show you how. That's practical. Lawyers get that part. See you in court.

Learn from the mistakes of others. You don't have time to make them all yourself.

 

[dcasto]

In our discussions with the TRRC, their chief field guy actually threw out the possibility of a "hundreds of mile long" pipeline built that way. Remember, interstate 10 ends with mile marker 810.

Also, the pipe would all be hydro tested, just not the welds. Or manufactured components

 

[BigInch]

Not all that surprising. I would hope that his boss give him a swift kick.
That typifies the usual problem with agency field inspectors; they know it all, and it's often the only regulatory interface with the owner/reps, other than submitting a permit itself.

Learn from the mistakes of others. You don't have time to make them all yourself.

 

[Duwe6]

". . .would all be hydro tested, just not the welds."

The pipe was tested at the mill. The entire purpose of a final hydro is to test the welds.

 

[LittleInch]

The pipe might have been tested at the mill (for about 10 seconds...). After that it has been transported many miles, lifted onto stacks, off stacks, welded, maybe cold formed into bends, lifted into a trench, surrounded by "soil", stressed a bit more to get it to fit at a few crossings / in the trench, settled in the ground, especially when water has been added - oh yes it's only the welds which are being / need testing....

The entire purpose of a final hydro is to test the system in its "as installed" position, ready to accept its contents. That means everything possible connected with the system (pipe, welds, fittings, flanges, supports) should be tested.

My motto: Learn something new every day

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

 

[BigInch]

Whether it's a good idea or not (it's not), 31.8 does allow no hydro on pipe, if it can be established that it was mill tested and transported in accordance with API rules for marine & rail transport, however given the above, that would only apply to pipes that will be used for repair purposes, with a radiograph of the golden welds sufficing.

Learn from the mistakes of others. You don't have time to make them all yourself.

 

[LittleInch]

I take it you mean 841.3.1 second para. However this could be taken as meaning only single pipes or multiple pipes which have been tested as a single string (e.g. a diversion or uprating). I would argue the case that this lack of hydro only applies to single pipes or strings which have been hydrotested.

My motto: Learn something new every day

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

 

[BigInch]

Yes. I think. It has always been my experience that it can mean a single pipe, or a string of pipes that were prefabricated & pre(hydro)tested in place, or before, then carefully dropped or pulled, or laid gently into place, with the final tie-in welds radiographed.

The same has been applied, in my experience, to block valve and tie-in prefabrications, beach pulls, platform risers and river crossings that had higher class location and test pressures than the mainline. After they were pretested at higher pressure (for the fabricated items DF=0.6, or riser = 0.5), were tested through again at the lower pressure consistent with the Class 1 test pressure. That, even though they were not specifically intended for "repairs".

Learn from the mistakes of others. You don't have time to make them all yourself.