Pressure Test of Flanges in CSA Z662 1

[302Hugo]

Good day,

I have ready CSA Z662 and I find it is not clear and I'm hoping someone can clarify if they have the same experience/knowledge. In the following situation, is it required to pressure test the flange by itself:

We are cutting into an existing piece of pipe and will be welding on a flange complete with blind. We have determined that it is impracticable to complete a pressure test on the whole system and because of the size of the piping (20") we are unsure if we can have a carber plug hold for a hydro. Therefore we plan to complete an alternative test (closure weld) as per 7.10.3.1. However, our construction group is saying they need to weld a cap onto the flange and pressure test it, then cut it off and then weld it onto our pipe. They state this is necessary because the flange is "part of the piping" and any piping needs to be pressure tested. Now, I can not find anything in the code that says flanges are part of the piping (we do not complete pressure tests with valves installed). My response is that the flange is an ASME fitting to a pressure class so it should hold pressure and does not need tested, and the only thing we need to test is the closure weld. B16.5 states that the flanges do not need pressure tested and Z245.12 doesn't say anything about pressure testing...although those are the fitting codes and not the piping codes. As well, welding on the fitting to a test head and then cutting it off cannot be good for the integrity of the flange as we may encounter heat affected areas and the fitting has now been altered...and really we could consider that flange now used and we now need to meet Clause 5.6 Reuse of Materials.

Am I wrong in thinking we do not need to pressure test this flange? Help with this would be greatly appreciated.

 

[LittleInch]

Your construction group is plainly mad. Ask them if the "pressure test" every other flange they have ever installed. When they answer "of course not", simply ask them why this flange is now somehow extra special. Expect silence.

I believe your response is near perfect.

I don't know the CSA codes and don't have a copy, but in essence they are all the same when it comes to this sort of thing.

If they go mad and "insist" then I would add a pup piece at least 300mm long to the flange, weld your cap on, test, cut the pup piece back by at least 150mm and then either cut back your 20" a bit more than you would have or extend the flange and blind by say 150mm inline to get out of the double HAZ and do a pipe to pipe weld and do your alternative test on that. It would be easier to do an RT on that weld than onto the flange, but it's a lot of bother for no advance in safety or testing or anything.

My motto: Learn something new every day

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

 

[302Hugo]

Thanks for the reply and re-assurance!

 

[ch44all]

Another thought. Make your weld on the existing pipe, test against the existing pipe. Cut out portion of pipe to have flange communicate with existing pipe. This is a great technique on smaller fittings. I've never had to do it on a 20" line.[ul]
[li][/li]
[/ul]

 

[William662]

I like the motto but the "also" is forgetting that, "That's the way everyone does it" and "that's the way we've always done it" do not constitute a defense, when you failed to meet the minimum legal code requirements, and someone was injured or worse. Both Canadian and US standards are continually being improved. I think the ASME B31 series is on a 2 year cycle. What we did 5 years ago is no longer acceptable because it didn't work effectively.

I think if you read the Z662 you will discover it is not the same as all the others. Section 8 deals with testing and is very prescriptive.

The closure welds do not require testing, mostly because this term is not used. In the Z662, "tie-in" is a defined term and deals specifically with this scenario. Sight unseen, I think the weld to attach the flange will require 100% NDE, and the flange and blanking plate will have to be visually monitored during pressurization to meet the Z662 requirements. You may wish to verify this with your AHJ, as in Canada some jurisdictions will interpret the rules slightly different and may require the flange itself be tested. This will be rare, but can be arranged, and without welding. Hope this helps.

p.s. Almost all flanges, excepting only those that constitute "tie-in" joints, are to be tested for strength and leak tightness as part of a Canadian pipeline systems.