Weld acceptance under B31.3 - defect or not 2

[XL83NL]

In view of this discussion I have an example of which Im not certain if it contains a B31.3 designated weld defect. Could anyone indicate if there's a weld defect in attached picture, and when applicable which type?
I cant find any real good example of non-acceptance acc to the criteria of the referenced topic, nor acc fig 341.3.2, keeping in mind the required weld metal thickness has to be 10S (see below).

The weld end of the 2 fittings welded in the picture were counterbored on the inside from 40S to 10S. The fittings have been welded without any root gap. There was no external or internal misalignment during welding.
The welds are already RT examined & accepted under B31.3 critieria. Code of construction = B31.3.

 

[metengr]

You would need to review the RT for this butt weld for proper examination of weld imperfections per 341.3.2, of B31.3. Visually from the OD surface the weld crown does not look pretty but it does not appear to contain undercut.

 

[XL83NL]

Ill review the RT report and film and see what I can find.
VT per B31.3 will check more defect types than RT, so given VT could there be any indication to report any defect?
PS: the picture shows the ID of the weld (thus root), not the OD side.

 

[dhengr]

I’m having some problem imagining that your picture is of the ID (inside, root side) of a pipe or fitting or some such, and I don’t have a copy of B31.3. It sure looks like the OD of a pipe or fitting to me. What is the light blue and the white material in the photo? Nor do I understand some of your abbreviations ‘acc fig 341.3.2,’ or 40S, 10S and VT. In any case, that weld looks kinda like pretty poor chicken sh-- to me. I’ll bet the spec. intention wasn’t to fill the groove to 80 - 90% of the finished surface, but that’s what all the concavity we see shows. While there may not be any spec./code defined undercuts, I can’t imagine all the irregularities and unfilled cavities/puddles in the weld surface are much less critical than an undercut. And, the two sharp tits we see, at left on the top and at right on the bottom look like trouble to me. If the picture is of the OD, those tits could be ground down, but I don’t understand how they got there; and if the picture is of the ID, those tits (sharp cavities) are serious stress raisers and must be filled.

 

[gtaw]

Unfortunately, ASME has no criteria for ugly.

Best regards - Al

 

[XL83NL]

Attached picture is better, and no, the photo is not from the OD, it's the ID. The picture confirms this. The light blue and white is just unclearness/resoluation of the camera.

‘acc fig 341.3.2,’ means according fogure 341.3.2 of ASME B31.3. 40S is a schedule size or wall thickness designation. See wikipedia or google what it means (not being sarcastic here but hey, if ya dont know...). 10S => same as 40S, except another will thickness designation. VT is visual testing, or visual examination according ASME B31.3 para 341.4.1 and 344.2.

What I was curious to find out is if this picture shows undercut or lack of fusion, as defined by the Code. I believe it didnt, which seems to be indirectly answered here as well. I do agree it's not the best/bmost beautiful weld, but thats not the question here, as neither the question is if we have to accept this. Is it 'Code accepted'? So far the answer seems to be what I anticipated.

 

[gtaw]

You didn't tell us what the fluid service classification is, so I assume it is Normal. You provided no dimensions, so that we cannot assess the depth of root concavity.

Generally, if the weld reinforcement is within acceptable limits, i.e., not too high, and if the root concavity is such that the thickness through the weld is at least equal to the thickness of the adjacent base metal, the weld is accepted.

Best regards - Al

 

[XL83NL]

Yeah I shouldve mentioned those, although it's not important for the question I asked;
- Normal fluid service,
- 1" 40S fittings,
- with butt weld ends machined internally to 10S.

Im only curious if anything in the root is considered to be a probable weld defect, not whether if its within acceptable limits.
Reason Im asking is because the ID has been machined to 10S. (assuming) The weld metal thickness is what it should be.
The weld end prep may not have had the best configuration to B16.25, the taper should've been located outside the 'weld zone', so to get a flat profile adjacent to the root.

 

[Kiwi2671]

XL83NL,
Is this an experiment ?
Can't understand why else you would machine Sch 40 on both fittings down to Sch 10 ?
If you were welding a Sch 40 fitting to a Sch 10 fitting then I would understand.
You have an unmelted tack weld at 3 o'clock but that is not rejectable (as far as I am aware oxidation of s/s is not allowed by most specs but it is not addressed by the code).
What I find "may" be rejectable is the unfilled prep in the bottom right hand corner of the photograph, however, cannot comment without viewing the graph.
If the density in that area is darker than the parent metal (and it certainly looks like it would be)then IMHO it would be rejectable based on the root concavity clause.
Regards,
Kiwi

 

[stanweld]

The counterboring has created a stress riser as depicted at 6:00, yet the root would meet Table 341.3.2, Visual Examination requirements. Whether the stress riser is detrimental, cannot be answered without design specifics. It is also noted that the heat tint shown may also decrease corrosion resistance depending on the fluid/service contained.

 

[moltenmetal]

There's a defect in the wall of this assembly, formed not by the welding process but by poor preparation.

If all, or most of the sch40 wall thickness is actually required to meet the minimum code-required wall thickness, I would reject this regardless of what the code said, since you've got what amounts to a defect which is more than 10% of the wall thickness to my eye. Even if the preparation defect were acceptable from a code perspective, you've got a lovely point on the ID for corrosion or stress cracking to initiate in future, if not a stress riser which might just fatigue crack over time without the help of corrosion/embrittlement.

If you used sch40 fittings despite sch10 being adequate for the duty, then I defer to the code inspection experts here, as you have adequate wall thickness even at the "defect"- I am not such an expert, so can't comment on whether what is plainly a defect in PREPARATION counts as a defect in the finished weld if you're left with more than adequate wall thickness under the defect. Then again, I've never fully understood why excess reinforcement was a defect either, aside from perhaps the stress concentration it can cause at the edges of the weld- and I'd be more worried about a deep ID notch such as what I see in your photo than that.

I too see no reason for back-boring BW fittings to sch10S unless they're welded to sch10S pipe or other sch10S fittings. And when we back-bore, we try our best to back-bore far enough that the RT examiner can see the edge of the back-bore in the radiograph, distinct from the weld, so he/she knows what they're looking at. What really happened here: the fitter screwed up, or the wrong fittings were purchased?

 

[XL83NL]

Yes, its an experiment, started prior to production welding. All welds/components will be pickled&passivated in&outside after welding. Let's say the line class requires 40S for small bore piping, and (I believe starting) at 1" and above it requires 10S; since pressure is low, 10S suffices, 40S is only required for extra mechanical strength (e.g. when someone steps up using the piping system, or when relative heavy inline equipment has to be supported). Hence tee's have to be 10S on the run, 40S on the brach. Thats how they were ordered, delivered, and thus welded as a test. Although it turns out to be a crappy solution, lets just leave that for what it is (knowing next time this way of counterboring isnt the solution).

The RT's where thus made such to determine code acceptance. Radiographic films were OK. This doesnt relieve us from accepting the welds; as moltenmetal pointed, the stress raiser and notches may be a concern for fatique and corrosion.
To answer the last question, what happened is that the supplier counterbored the fittings to meet the purchase order requirements (the PO e.g. stated "TEE EQUAL, 1'x3/4", 10S/40S"). The shop then welded 2 tees together (worst case scenario), and didnt care to look beyond their glasses, let alone looked at some useful pictures in B31.3 or B16.25 for weld end prep (or where too lazy too make a good counterbore with porper taper detail). Again, its just an experiment, I want to be one step ahead of any insepctor.

Thanks all, my question seems to be answered.