Weld surface finish / ASME B31.3 item 341.3.2 / Severe Cyclic Service

[Deccocwb]

According to ASME B31.3 table 341.3.1, severe cyclic service piping welds have surface finish criteria (roughness) of 12,5um Ra.

Somebody experienced measuring this roughness over weld bead, once there´s a numerical criteria?

Somebody knows the influence of roughness in fatigue life?

What if i don´t mind this criteria?

In my opinion the weld shape (weld toe) is much more critical than roughness to fatigue life. I measured weld roughness over weld beads made by FCAW, SMAW and GTAW, and found roughness little bit over the specified 12,5um Ra. (up to 18,0um Ra)

I´m talking about process piping in oil&gas industrie.

I´m digging the web looking for answers, I would appreciate a little help here!

thanks

 

[mk3223]

The control of the surface roughness is to prevent the material from the fatigue failure in the severe cyclic service. In the cyclic condition, the surface defect is main stress concentration of the material which could initiate a crack. One can check many incidents happened in the past for details.

 

[Deccocwb]

Mk3223, I agree with you if I flush the weld reinforcement. I agree the roughness could be a stress concentrator in any material aplied in cyclic condition, but when you put a weld in the middle, the roughness became less critical than the weld profile (in as welded condition). Considering a circunferencial weld joint, in as welded state, in my opinion the weld toe will prevail the weld roughness in stress concentration to initiate a crack, that´s why I don´t understand why put this criteria to the weld.

thanks.