Joint efficiency of weld with nearby laminations in plate

[fedefer85]

Hello. My case is a pressure vessel with no design data (0,70 by default). An ultrasonic inspection of welding were performed in order to increase this value. However, there were found laminations in shell, reaching the shell-head weld, so increasing vessel joint efficiency is not that easy.

Does anyone know if there is a table of joint factors for welds with defects, or something similar to that? If it not, it seems like it won't be able to has a joint factor greater than 0,70.

 

[pdiculous963]

This situation is concerning. If the laminations approach a head to shell weld, they are in a stress field where local bending occurs due to the discontinuity, which is a concern for plastic collapse. Also, API 579 would require a cracking assessment as well due to the proximity to the weld. I would recommend you hire an expert who is experienced in performing this kind of assessment.

 

[racookpe1978]

So, you have suspect joints - checked them by NDE and have found actual, real serious defects in those joints.
And you want to find some excuse to INCREASE the tank's original pressure rating? That's NOT a responsible action.

Why don't you repair the defects?

 

[fedefer85]

Thank you both pdiculous963 and racookpe1978 for your answear.
To make it clear, the weld has no cracks or other defects, this known by the Angular UT performed (probably a MT will be perform to ensure this). Laminations reachs the ZAC but are not inside the weld and nor cracks propagate from it.
And racookpe1978 I've said that if I don't have any Code/Fitness Analysis or something like that to increase the Joint Factor, it will be stay as 0,70, so don't judge.

 

[r6155]

With no design data this pressure vessel need to be out of service until an inspection and design be completed.

Regards
r6155

 

[metengr]

I know of no tables that assume a type of defect for joint efficiency. Attempting to increase the joint efficiency through additional NDT is a re-rate of the pressure vessel. However, laminations are non-injurious plate defects and not being in the weld are considered harmless based on their orientation relative to membrane stress. I would use a standard, like the National Board Inspection Code or API 510 for re-rating guidance.

 

[david339933]

What is the purpose of increasing JE? NB does not allow post construction NDE to increase JE for the purpose of increasing MAWP.

 

[fedefer85]

Thank you all for your comments. This vessel is a case of Equipment with Minimal Documentation (no Nameplate, no design data and not operating in USA), so a 7.7 of API 510, was conducted. Now, the following procedure will be done:

1) to ensure that the weld has no defects (already tested with UT) a MT will be performed, and when finding it satisfactory JE will be increased
2) to evaluate laminations a fitness for service analysis per API 579 will be conducted, in order to determine that this defects of fabrication are acceptable for safely service

Thank you all
Regards

 

[weldstan]

I assume that you do know the materials of construction. If not why are you contemplating even using this item?

 

[mk3223]

From the safety point of view, IMO, after the fitness for service assessment, a plan of repair or replace may be prepared and ready to fix the defect as the opportunity arises.

 

[racookpe1978]

Perhaps you misunderstand our intentions. Forget the joint for a moment!

The laminations in the plate itself are my concern. Those defects should be ground out and re-welded to safe metal. Regardless of the joint. The metal AWAY from the joint needs to be as strong as intended by the original designer for the second user to pressurize the tank safely.

If you do not repair the tank wall, then the "actual" tank wall is only as "thick" as the distance from the outside of the wall to the nearest lamination. Basically, half the thickness you "think" you have measured.

 

[BJI]

You are not going to grind out laminations and weld repair.. From an internal pressure perspective the effect of laminations will have negligible effect on the strength of the vessel, you would be better to leave it as it is and replace at end of life. However, if you had a set-on nozzle or an external attachment, then your effective thickness would be the ligament between the lamination and the attachment (unless only acting in compression). This is why you typically only check for laminations under such circumstances. You don't really see it much now with modern manufacturing processes but many older vessels and tanks have been operating with laminations for years without issue.

The better questions would be why do you want to increase the joint efficiency, you can't re-rate the vessel with the lack of design and documentation. This is not limited to the US, pretty standard worldwide.

 

[fedefer85]

This is not a re-rating, is a 7.7 of API 510, establishing a MAWP with conservative material SA-283 Gr C and JE 1.0 instead of 0.70 but backed up with satisfactory UT and MT performed on weld (and also RT if it necessary). In the hypothetical case that the weld had defects it would have been analysed by ASME IX/FFS-API579 and if necessary been repaired, JUST AS IT IS SAID IN THAT POINT OF API 510.

Thank you, regards