STRESS IN WELD ROOT OF CS & LS.

[BPVFEA]

A pressure vessel (ID 710mm,THK 8mm,tangent length 1290 mm) with torispherical ends is having (1) one CS (full penetrated, single V, 3mm root gap, without back strip) welded from inside, (2) one LS (full penetrated, single V, 3mm root gap, without back strip) welded from inside and (3) one CS (full penetrated, single V, 3mm root gap, with back strip) welded from outside.
(For all above welds included angle of butt V is 30 deg. & no root face)
Welds 1 and 2 are back chipped and re-welded from outside.
With regard to roots of above welds, I would like to know about following.
a) Under normal circumstances, joint is welded from outside. Is it right statement ? Please let me know the reason.(In case of single V joint, Standards & codes generally indicate required geometry of the joints, welded from outside)
b) Does any clause of any standard or code mention about required weld geometry of the single V welds which are welded from inside of the vessel?
c) What type of stress (tensile/compressive) is acting in roots of welds 1, 2 & 3?
d) Is it true that, weld root should always be under less stress comparing to stress
in other side of weld? Please let me know the reason.
Thanks in advance.

 

[prex]

a)Don't see any difference between V welds done from inside or outside, nor does the code. It's more usual to weld from outside just because the accessibility is generally better.
b)Don't remember any. By the way the code states very little about weld geometry, as this varies widely with the process (I guess you are not using electrodes, as your 30 degrees angle seems a bit low for that, but it's up to you to decide and to prove that your process is correct).
c)Don't see any reason to worry about stresses in the welds: as they are full penetration, the stresses are the same as in the parent metal.
d)Again don't understand this point.
prex
motori@xcalcsREMOVE.com

http://www.xcalcs.com

Online tools for structural design

 

[AEF]

I agree with prex.
a)Normally the root pass is done from the ID then ground back from the OD to sound metal and welded out. This tends to give you better control/quality of the finished product. You'd normally weld the root pass from OD, with or without backing, only when you don't have access to the ID. Your fabrication code should tell you this.
b)The geometry of the weld depends on the WPS for the process your using, thickness and type of base metal, etc.
c)Full penetration welds are in the same stress as the parent material.
d)I'll try this one. As long as the vessel operates in the elastic region of stress, the highest stress (tensile) should be on the OD membrane. Long seams are always in higher stress then circ seams (hoop stress calc).

 

[twjag]

Follow on to prex & AEF posts:

DMJadhav did not indicate which Code.
Following comments are only for ASME Section VIII, Div 1.

Weld joint details are not specified by Code. All welds must be performed per a Weld Procedure Specification (WPS) qualified (PQR) per ASME Section IX. Per ASME Section IX:
1. Joint details used for PQR must be recorded on the PQR.
2. For the WPS, joint geometry is a non-essential variable.
4. Welding from one side only, or welding from two sides are essential variables, with each requiring separate PQR's.
3. The WPS may list or discribe all joint geometries desired or to be used. The WPS can be revised to add or remove joint geometries without need for requalification, so long as all other essential weld varibles are satisfied.

ASME Section VIII, Div 1 classifies welds (see Table UW-12) by Type (butt-welds welded from 2 sides, welded from 1 side with backing, welded from 1 side without backing, lap welded, etc). The table lists allowable joint efficiencies based upon combination of Type & level of RT inspection.