Weld Buildup - Legal? 1

[weldtek]

There's no question that ASME Sec VIII Div 1 allows weld buildup under certain conditions. Specifically, it is allowed to 'restore' base metal for strength, and, for other purposes such as transitions in thickness.
My question is, does the Code allow a fabricator to take a piece of plate say 1" thick and use weld metal buildup to increase it to say 1.25" thick?
My AI says no, it use is limited to 'restoration', and, I concur, however, I have a customer who argues otherwise. By the way, the customer has purchased an item which was the correct thickness, but, after a change in design criteria, no longer meets required design thickness.

 

[vesselfab]

have never thought about that

has never come up

while I concur with AI, I do not know where it states that.

But in the end...Your AI will have to "sign off" on it and his opinion is pretty important unless you can show him where code says it is allowable or does not dis-allow it

 

[doct9960]

weldtek,
I also concur with your AI. Either buy a thicker plate or consider the possibility of doing layered construction per Part ULW.

vesselfab,
See UW-42.

 

[metengr]

My question is, does the Code allow a fabricator to take a piece of plate say 1" thick and use weld metal buildup to increase it to say 1.25" thick?

Reply; No. Repair welding under Section VIII, Div 1 and Section I is permitted, and in most cases by the plate material specification, as indicated above. However, IMO attempting to use weld metal to increase the entire plate thickness for design would result in the plate having been modified from the original SA plate material specification. You now have a cast/wrought plate structure that is not a true wrought product in accrodance with ASME SA 20.

Cylindrical components (pipe and tube) can be handled a little differently, where weld overlays will be counted as part of the pressure boundary, based on a recent revision to ASME Code, Section I. In the past, weld overlays were not counted as part of the pressure boundary. This is only limited to pipe and tube under Section I, and not to plate under Section VIII.

 

[weldtek]

Thanks for all your input. The individual pushing us to use weld buildup happens to be on several Code committees, and is taking the stance that, in his words, since the Code is mute on this point, it's ok. Our position is simply, show us a Code paragraph or interpretation that allows it.
Metengr, do you happen to know where in Sec 1 overlay is addressed?
Again thanks to all.

 

[eliebl]

The codes are silent on many things. They may not prohibit something but that does not mean it is good engineering practice.

 

[metengr]

weldtek
The revision to Section I will be in the 2008 addendum of ASME Section I. It is a revision to PG-27, which will allow the use of clad material for strength calculations, whereas, clad material under present Section I rules is not considered.

By the way, if these so called "consultants" are using their code affiliation to provide these types of recommendations, I would have them yanked from their respective committees.

 

[weldtek]

metengr,

Thanks for the reference. As always your input is much appreciated.

 

[jte]

Guess I'll take a somewhat contratrian position (no, I am not currently involved with a weld buildup issue)... Shields Up!

Although I would not advocate weld metal buildup for increasing the plate thickness of an entire shell, the code (i.e. VIII Div. 1) explicitly allows weld metal buildup for strength. For example, UW16(c)(1) explicitly allows the use of weld metal buildup to provide area of reinforcement for openings. See also Interpretation VIII-1-04-57. From a field inspection perspective, I'd rather see weld buildup used for opening reinforcement instead of re-pads. No one can tell me how thick a shell is under a re-pad without entering the vessel. Since shell areas in the vicinity of nozzles are often subject to deterioration, I'd rather be able to check the wall thickness externally via UT. With a weld buildup I can grind a flat spot for UT monitoring on the run. With repads I have to wait until the vessel is out of service and cleared for an internal inspection.

Presuming the vessel in question is a typical carbon steel, one approach may be to use a weld overlay (i.e. corrosion resistant) instead of weld buildup (i.e. same nominal chemistry as base material). This is allowed in VIII-1 in UCL-23(c):

Base Material with Corrosion Resistant Integral or Weld Metal Overlay Cladding With Credit for Cladding Thickness.
When the base material with corrosion resistant integral cladding conforms to one of the specifications listed in UCL-11(a), or consists of an acceptable base material with corrosion resistant weld metal overlay and the joints are completed by depositing corrosion resisting weld metal over the weld in the base material to restore the cladding, the design calculations may be based on a thickness equal to the nominal thickness of the base material plus Sc /Sb times the nominal thickness of the cladding after any allowance provided for corrosion has been deducted...

I've never used this approach, but do not see why I shouldn't keep the option open... It's not too much of an extension to question "if I can use a dissimilar metal for weld overlay and take credit for the strength, why can't I use same nominal chemistry weld buildup?"

jt

 

[vesselfab]

another take on this.

The ""Customer"" who bought an item that is too thin for the design conditions and the same ""customer"" wants to do a weld build up to thicken head/shell.

This also becomes a sticky point code law.

Whoever performs this weld build up and rerates the vessel, now becomes responsible for all the design and manufacturing functions of the vessel. The original fabricator, even though this appears to be a new unit, is absolved of all wrong doing and the person who puts on the new R stamp for altering and rerating the vessel is now responsible.

From a fabricator's standpoint, even if it was decided the code allows it, I would not accept the job.

Shrinkage and deformation of the shell would be a major concern. I can see adding to the openings reinforcement elements if there are any.

I would just not be willing to jump off into a can of worms with that much exposure to risk. Right of wrong, it's still a bad practice in this old man's point of view

 

[metengr]

jt;
Keep your shields down with your quote

"if I can use a dissimilar metal for weld overlay and take credit for the strength, why can't I use same nominal chemistry weld buildup?"

. In reality, why would anyone want to consider this type of weld build-up unless you go to my Point 2 below OR someone made a huge mistake during the design phase of the vessel and ordered plate that was too thin!


As with any Construction Code or Standard, one cannot write rules for every application or intended use. As a metallurgical engineer, I can understand and support the use of taking credit for a barrier alloy applied to a substrate (weld overlay), during design, that serves to protect the substrate from corrosion in service or a layered vessel construction.

However, using welding to build-up the entire thickness of a completed vessel after the fact (see my point 2 below), is not analogous to a weld overlay, as intended in the Code.

Point number 2, in the situation above;

By the way, the customer has purchased an item which was the correct thickness, but, after a change in design criteria, no longer meets required design thickness.

Change in design criteria? This constitutes a big red flag to me as a vessel re-rate for service. This does not mean that the vessel is being designed and fabricated for a re-rate, the vessel has already been built, and now someone wants to come along and add weld metal to thicken the shell so that it can be used under re-rated design conditions?

If the vessel has been built and stamped, this work would fall under the NBIC, as an alteration (RC-3022). Under the current rules of the NBIC, weld build-up is treated as a repair and is limited as to the amount because of concerns related to the effect of residual stress on the substrate material. Also, I believe it was limited to avoid this very type of situation.

I believe attempting to do this type of work on an as-built vessel would not fly with the AI or with the Jurisdiction unless the vessel was subjected to a detailed design analysis (including review of a thermal treatment to reduce residual stress effects from the weld build up), supplemental NDT, and hydrotesting.

 

[vesselfab]

my thoughts exactly, but much more eloquently stated.

 

[jte]

I agree that it is not ideal by any means to weld build-up material. However, mistakes happen. Sometimes things need to be scrapped, sometimes things need to be saved. Based on the OP, I'm looking at this from the perspective of a consultant who gets brought in after the vessel is substantially fabricated but before final assembly, hydrotest, PWHT (if req'd), and stamping. So the fab shop has a bunch of cans rolled and welded with nozzles attached, and... someone discovers that (for example) the hydrostatic head was not accounted for. There will be great pressure to resolve the issue without scrapping parts of the vessel since this will inevitably be the one vessel which is critical to the project schedule etc, etc, and scrapping and starting over will be a 6 month delay. Nothing ideal about such a situation, but it has happened before and will certainly happen again. So, now what?

I think my recommendations, in order of "best" to "worst" would be:
* Apply a corrosion resistant weld overlay. Use what was the corrosion allowance for strength. Hopefully this alone will solve the problem.
* Apply a corrosion resistant weld overlay, use what was the CA for strength, and use the overlay for strength. Thicken the overlay as needed.
* If weld buildup on the order of 25% of the wall thickness over the bottom shell course and head solves the problem, consider that. Implement UW-42 for the weld buildup. PWHT to relieve stresses. Remember, this is a shop, not field, application. I would have a hard time explaining to a plant manager why they can weld build 50% wall thickness in the field for "repairs" but cannot weld build 25% wall thickness in a shop for new construction.

Of course, a materials and/or welding engineer would be involved. In my organization I'm heavily outnumbered by them (I'm stress guy) and would be beaten to a pulp if I didn't involve them...

In any case, this is an interesting discussion. Glad that for me its hypothetical!

jt

 

[JoeTank]

Doesn't the Code require radiography of the weld overlay to the sample extent as the shell if it's used for strength? If so, that's RT of the entire surface of the vessel!!! How round do you think that vessel will be after all that weld metal is trowelled onto the vessel surface.

Joe Tank

 

[jte]

Joe-

See UW-42(a). I think what you are thinking about is UW-42(b) which deals with weld joints and tapered transitions. I do not think the RT applies to the "restoring the thickness" part.

Pondering this a bit further, I'm wondering what exactly is meant by "restoring the thickness". Since VIII-1 is a new construction code, the steel has not corroded since it has never been in service. We are not talking about NBIC repairs here. The only application I could think of in which thickness would have to be restored in new construction would be material which has thinned due to a forming operation such as spinning a head or a mistaken machining operation such as a grinder accidentally contacting the shell plate and creating a small thin spot.

jt

 

[metengr]

jte;
Something is rotten here, and I can't put my finger on it. If this is new construction, you have a design, bill of materials and low and behold a vessel is built per contract documents. Now, you have a design change either on the front end or probably what is closer to the truth, a shipment of plate is received and someone in the shop can't read a micrometer or conduct a UT thickness check and material that is thinner than specified is used. This is a nonconformance and must be dealt with in accordance with the Certificate holder's Quality Program, and the AI must by off on this, plain and simple.

If the vessel design is changed in process, why would this even ocurr? If the vessel has been built after the fact, stamped and hydrotested, this would apply to post construction activities.

 

[jte]

metengr-

I've seen design liquid levels change after a PO has been issued for a vessel. Its not unusual for the long lead items to be well along the fabrication timeline by the time detailed piping design begins. It was common for me to order vessels with nozzle elevations known but no orientations since the detailed piping design hadn't occurred yet. Then the first HAZOP is conducted and someone realizes the high-high liquid level is actually 30'. But the initial design for the vessel indicated a HHLL of 15'. Or the senior process engineer reviews the project and decides that the boot on a horizontal drum needs to be increased. So its not entirely inconceivable that an honest design change comes through after the plate has already been rolled and nozzles welded in etc but it hasn't had all the internals installed nor sent to be PWHT . I could see the value in performing a weld buildup on the bottom head and first shell course or two rather than scrap the entire bottom section of a column, procure new plate and a new head, and rework the bottom section from scratch. I'd imagine a weld-build would result in a week delay vs several weeks for the re-do.

jt

 

[vesselfab]

jte

I build columns for a living. If a client asked me to do a weld build up of a head and a couple of courses of shell, the answer would be a FLAT NO! No discussion, no debate.

In the case you cited, 15 ft of head only about 8-10 lb of pressure would not effect that much, and if there was not enough original material thickness to compensate, he would have a couple of choice to make. if the mawp can be lowered and still be above design pressure...great. If it can't...he can lower design pressure to the maximum I can give him or he can pay for new heads and shell and the additional labor incurred. NO DEBATES.

The same thing would stand true on a boot/sump. NO DEBATES.

On some item, when dealing with user or his engineering company, we stand firm and will not put out a product that is manufactured in this manner.

No matter how much make-up you put on a sow...she's still a pig.

Sometimes the user and the process designers and the original engineer just have to pony-up and face the fact that a mistake was made and they have to pay for it.

Questionable fabrication techniques and band-aids to not make for a good product in the end.

 

[metengr]

jte;
Thanks for your perspective on the fabrication side. As an owner/user, I would not permit weld build-up of a complete pressure retaining component because if I were asked to render an opinion for my boss, I could not accept a weld build-up to as means to provide the necessary thickness for vessel re-rating. You have a cast and wrought sandwiched material that must somehow be identified as such on a data report. What happens later on in service should a possible lack of fusion defect result in fusion zone cracking and I am not around to know what was done as the Owner?

Keep in mind that Section I does now allow surface weld build-up to be treated in the pressure boundary calculation, with more restrictive requirements instead of a simple weld overlay. Before I would go around and consider this option for pressure vessels, I would explore more restrictive requirements should this method be considered during fabrication, which might kill the idea in the first place.

 

[weldtek]

In our case platform were added after the fact and the thickness of the lower shell courses was increased, but, the head is a long lead item and so the customer is trying to modify it so it will make it. Incidentally, we're suggesting removal of the entire straight flange to get around this issue.