Acceptable Weld Straining Due To Forming (UG-79)

[Paulettaa]

Dear all

Please help me understand this issue regarding forming of heads fabricated from more than one piece.
ASME BPVC VIII-1 UG-79, regarding weld thickness reduction due to forming, reads:

(d) A reduction in weld thickness due to a forming operation
is acceptable, provided all of the following conditions are met:
(1) Prior to the forming operation, the weld(s) are
verified to comply with UW-35(a) through UW-35(d) by
the Manufacturer and the Inspector.
(2) The reduced weld thickness, at any point, shall
not be less than the design thickness of the component.
(3) The reduction in thickness shall not exceed
1/32 in. (1 mm) or 10% of the nominal thickness of the adjoining
surface, whichever is less.

Does this mean that for head fabrication, if the head is fabricated by forming a blank which has a weld seam, the weld thickness shall be checked before and after forming? And if the reduction in thickness is more than 1 mm then is the head rejected as a whole?

Here the code does not mention whether the head is cold formed or hot formed nor does it talk about carbon steel or any other material of construction. So is it a general requirement independent of the material or hot/cold forming?

Warm Regards

 

[r6155]

Please, read UG-79 in full. It is very clear.

Regards

 

[Paulettaa]

r6155 Thank you for reminding me to read UG-79 in full which is my question in the first place.
It was not clear to my eyes and that is why I have asked this question. It is silent on how to deal with the component if the reduction in the weld thickness exceeds the limit. Does the component have to be rejected? Can removing the weld and depositing new weld help accepting the component? If removing the weld is a method to save that component then does the weld have to be removed in its entirety or just locations where there is excessive reduction in the thickness?

Anyway, thanks for nothing.

 

[r6155]

You and a welding engineer must decide what to do when a limit is exceeded.
Use your engineering judgement.
Again, UG-79 is very clear

Regards

 

[david339933]

Paulettaa, what exactly is the issue? The weld is going to thin when forming a head no matter what, same as if it was a seamless head. As long as the minimum required is achieved and is not thinner than the adjacent material, there should be no problem.

 

[Paulettaa]

David the problem is that what happens if the thinning exceeds the limit of 1 mm (or 10 percent of the plate thickness)?
It is not about thickness being reduced to required thickness by design. It is a question of reduction in the thickness with respect to initial before forming thickness.
Suppose that the required thickness by design is 29 mm and I choose a 35 mm plate with a weld seam as the initial before forming plate. After forming the thickness in the knuckle region has reduced to 31 mm. The thickness is obviously above the required thickness but the reduced thickness is 4 mm which is above the limit as stated by UG-79 (d) (3).

 

[david339933]

That is what I thought you were saying....It will almost always exceed this limit. As long as you are not below the adjacent material there is no issue. I would be more worried if the weld didn't thin along with the rest of the material.

 

[Paulettaa]

David
The wording of the code does not allow a reduction in weld thickness above 1 mm. It says it is acceptable if this limit is not exceeded. How do you say this is not an issue? It violates this paragraph of the code.

 

[david339933]

This is not the intent of this paragraph. This paragraph is much like UW-35 which is talking about localized undercut type thinning. How do you propose to not thin the weld? So a flat plate can thin all it likes, but if you put a weld in it first it can't thin? Just plain silly.

 

[SwinnyGG]

This is not the intent of this paragraph....... Just plain silly.

I agree that it's silly (and I suspect Paulettaa does as well) but it's still pretty clear as to what the requirement is.

 

[Paulettaa]

OK David
Well to my eyes the requirement sounds like a limit on the weld strain not the plate strain. I have seen some engineering specifications recommend that if the heads cannot be fabricated from a single plate it is better to fabricate them in the crown and petal pattern and first form the individual pieces then weld them. This way no matter how much straining the plate goes through, the weld does not experience straining due to forming.

Besides, even if it might sound silly it is what can be understood from the wording of the code.

 

[r6155]

In this example reduced weld thickness is 31mm > 29 mm is OK. Then, reduced thickness of the adjoining surface is 31mm - 1mm (max.) = 30 mm is OK.

UG-79 is very clear

Regards

 

[Paulettaa]

R6155
The paragraph clearly states the reduction in weld thickness which means 35-31=4 mm.
Which is not good since it is more than 1 mm limit on the weld thickness reduction by forming.

 

[r6155]

Paulettaa.
Sorry, you are confused.
d) (3) The reduction in thickness shall not exceed 1/32 in. (1 mm) or 10% of the nominal thickness of the ADJOINING SURFACE, whichever is less.

Pay attention to the "ADJOINING SURFACE" (of the reduced weld)

Regards

 

[Paulettaa]

R6155
The adjoining thicknesses are 35 mm and 10 percent of 35 is 3.5. therefore, for this special case the reduction in weld thickness due to forming shall not exceed min( 1 mm, 3.5 mm) which means 1 mm. In fact if the thickness of plates to be formed after welding is more than 10 mm then the reduction in thickness cannot be more than 1 mm.

 

[r6155]

You reduce the weld and mantain the original plate thickness: IMPOSIBLE
Reduction thickness is not an abrupt change between plate and weld.

Regards

 

[jtseng123]

Then the head shall be rejected per code. It looks silly, not making sense of the code.
says, nominal 35 mm
after forming 29 mm of plate and weld.
per UG-79(d)(3), smaller of (1mm or 10%x29mm) ----> 1 mm governing
The actual reduction is 35-29 = 6 mm , exceeding 1 mm, hence head shall be rejected. Wow, first time learning this. Looks like most vessels fabricated in the past shall all be rejected.

 

[Paulettaa]

Jtseng
This has been added to the code since edition 2013. The restriction is only on the reduction in weld thickness not the plate and therefore one remedy could be using crown and petal by forming head pieces first and then welding the pieces together so that the weld does not strain.
I want to know if the welds are deformed and exceed the limit by the code is it possible to grind the weld out and weld again so that the head could be accepted?

 

[jtseng123]

Pauletaa, I have purchased hundreds of vessel for the past 30 years, and never ever realized there is such restriction in the code. No vendors or inspectors have ever mentioned this. To weld plate gores together and then form is very common. The typical thin out is from 3-6 mm. Based on this thin out, all heads have to be rejected by the code. I am not going to implement this code because it makes no sense at all. Otherwise, I have to tell all big oil companies the vessels we purchased for them are no good.

I might have wrong interpretation. Someone knows code better may advise. My business for the vessel is as usual and trust vendors and inspectors know better than me. I won't bother that restriction.

 

[r6155]

More information from PD 5500
4.1.5 Localized thinning
Localized thinning, below the nominal thickness, resulting from cutting, forming,
weld preparation or weld dressing shall not result in a thickness less than the
minimum thickness plus specified allowances (see 1.6), except where all the
following conditions are fulfilled.
a) Thickness reduction shall not exceed the smaller of e/20 and 5 mm.
b) Area of thickness reduction shall fit in a circle of diameter equal to the
smaller of e and 60 mm.
c) Any two areas of thickness reduction shall be at least (D.e)½ apart
d) The total area of thickness reduction shall be not greater than 2% of the
total vessel area.
e) Area of thickness reduction shall not be in the knuckle area of a dished end.
f) Details of thickness reductions shall be recorded in the final documentation
(see 1.5.2.2).
where
D is the internal diameter;
e is the nominal thickness, of the component under consideration.

Localized thinning that does not satisfy the above conditions shall be referred to
the designer for consideration.

I prefer ASME UG-79, it is clear to me.
Regards