Check of welds while the Plant is running 2

[Robys]

Good morning all,

In the Plant where I'm working we are facing several leakages from branches welds, like on sockolets or weldolets, on SS lines in corrosive service. Do you know if there are methods to check the integrity of these welds while the Plant is running?

Many thanks in advance!

 

[racookpe1978]

Visual inspection obviously. If the weld is leaking = it is a bad weld.

If the pipe is cold, you could use dye penetrant. Mag particle may work if no heat (no insulation) and no danger to the tester.

 

[Robys]

Dear Racookpe1978
I'm sorry my question was not clear. I'm looking for a method to check the other welds, the other apparently good, that are not leaking, while the plant is running.
Thanks and sorry again

 

[weldstan]

MT can provide limited subsurface detection of linear discontinuities depending on the operating temperatures of the pipe. UT and RT could provide highly limited detection as well.
None of these techniques will prove effective in finding small, shallow, linear discontinuities. Also you don't indicate the cause of leakage - SCC, pitting corrosion, crevice corrosion, general corrosion, fatigue cracking, etc.?

 

[gtaw]

Magnetic particle testing will be of no value on an austenitic stainless steel base metal.

Pentrant will be useless if the materials being tested has discontinuities that are "filled" with liquid such as water, oil, etc. The penetrant has to get into the discontinuity so that it can bleed back out to form a stain. If the discontinuity is "full" of something, the penetrant cannot get in: no stain.

Radiography will be complicated because hydrocarbon and water base fluids absorb ionizing radiation making it difficult to get a good image on the film.

UT on fillets, unlikely to produce acceptable results.

Best regards - Al

 

[weldstan]

My bad! forgot about the fact that it was austenitic stainless steel.

 

[sjrfc2]

We had problems with integrally reinforced fitting welds (MSS SP-97). We found that a lot had partial penetration. I THINK the problem was the smaller sizes but I'd have to check that. Bottom line is that there were some sizes that seemed as though they were more problematic for the welder's and we were able to categorize where our problem was likely to be.

I would start with similar size branches and go from there.

The fitting spec is non-dimensional (except for envelope). We found that we had a large variation regarding how different mfgs would supply these fittings (leading to inconsistencies regarding the joints the welders would have.

 

[DeleteriousPhases]

I'm in agreement with gtaw on this one.

schedule a shutdown, or plan RT/Dye for the next shutdown.

otherwise you're limited to dye/visual to try to spot the problems.

Any pictures of the leaks available? just interesting to figure out a cause to see what you're facing.

 

[Inspectengr]

Branch connections normally leak may be indicating the high vibration on the lines. First of all check for high vibration of lines. May be some machine or some control valve up stream is the reason. Second check the supports on that line such that they should be as per design. If you dont find discrepancies in support. Just check the small bore support connections. You can improve with installing gusset supports near small bore connections.

Visual inspection can tell you about poor finish welds that may cause fatigue cracks as well. Check for that as well.
I hope this helps.