seal weld definition

[Tmoose]

An internal standard prohibits GMAW due to fear of "Short-circuiting metal transfer" GMAW-S / MIG.
GMAW is barely allowed on "seal welds."

I'm wondering if there is a definition somewhere of when a weld is a "seal weld."

Our weld prep has been defined for decades like the attached jpg. Pretty much full penetration, on a 3/8 inch thick pipe (8 to 22 inch) with design code requirements of 50 psi, or possibly even much less.

thanks

Dan T

 

[SnTMan]

Tmoose, a weld is a seal weld when it performs no strength function. I doubt anyone would describe a full penetration circ seam weld as a seal weld.

Regards,

Mike

 

[metengr]

The function of a seal weld is to prevent leakage of fluids in service across an interface. A seal weld would be considered a fillet weld, by design.

 

[metengr]

What is the definition in lieu of function? Well, I usually use my source (ASME Section IX) and under definitions we have

seal weld - any weld designed primarly to provide a specific degree of tightness against leakage.

 

[Duwe6]

What metengr said. The most typical 'seal weld' is a threaded joint where a weld is put around the area where the threads meet the hole. This weld has no stress on it, as the strength is in the threads. The weld is only there to seal the threads and prevent leakage.

And I fully concur with your "internal standard". GMAW-S is too easy for a welder to turn from a 'cold' weld into a hot braze. Nothing wrong with brazing, unless the joint was calculated to provide a weld strength, not the lower strenght of a braze. Happens way too often to allow GMAW-S on items requiring 'real' welds, not 'some weld, some braze'.

 

[Tmoose]

Hi Duwe6,

I may have not described our internal standard clearly. Actually it is NOT real clear, so it may be impossible to describe it clearly.

In more than one of our stds, the entire broad category of SMAW/MIG is forbidden for structural parts, engineered parts, or "pressure" containing parts.
Unfortunately, at the same time, some of the same stds say GMAW-S >>may<< be approved for "seal welds" or possibly low stress welds.
So we occasionally specifically forbid GMAW-S welds, but also forbid GMAW welds for fear they "might" be GMAW-S welds.

AWS structural codes have a bunch of prequalified structural joints that are OK for GMAW, but carry a note that not for GMAW-S. Maybe it their industry bias, but that seems to imply there are ways to avoid accidently create GMAW-S welds in production.

Our rational reportedly is, there is no on-off switch for short circuit mode. If volts or amps or whatever are cranked down a little low, the process changes to short circuit mode, and aside from the sound, there is no convenient way to know it. Google searches suggest the concern about poor fusion using GMAW-S on thick sections is not unfounded.

============================

Did you mean to refer to a P1 TO P1 weld (steel>steel with steel filler) with poor or incomplete fusion as "brazing?"

thanks

Dan T

 

[gtaw]

Many people use the term seal weld to indicate the weld is not designed to transmit a load/stress/or force from one member to another. The problem with that presumption is that a welded structure is monolithic. As such, when the assembly is loaded, the weld will share a portion of the load whether it is by design or because of the monolithic nature of welded structures. With that in mind, the minimum size weld (both PJP groove welds and fillets) should be sized to ensure sufficient heat input and/or preheat is developed to ensure the weld has sufficient ductility to prevent cracking. AWS D1.1 provides tables of minimum weld sizes when welding carbon and high strength low alloy steels. The sizing of the “seal weld” is based on metallurgical considerations, not strength.

The welder should be following a WPS that stipulates the welding parameters to be used. The parameters, i.e., voltage, wire feed speed, electrode extension, and shielding gas will determine the transfer mode.

The welders must be trained to work within the welding parameters listed by the WPS and it is incumbent on the individual developing the WPS to understand how to control the transfer mode via the welding parameters. There are references available to that can serve as useful aids when writing the WPS to ensure the appropriate parameters are provided to produce the desired transfer mode. AWS publishes a Recommended Practice for GMAW and many manufacturers such as Lincoln and ESAB have information available on their websites.

If the WPS is written to ensure spray or globular or pulsed transfer is used, short circuiting transfer should not be an issue if the welder is working within the permitted ranges of voltage, wire feed speed, etc. listed on the WPS. Dropping the welding parameters to the point where short circuiting transfer mode is produced indicates the welder is not following the WPS.

The fabricator/manufacturer should make a surveillance program part of any QC program to check the welding while it is in-process. The surveillance program will help to ensure the welders are following the approved WPS.


Best regards - Al

 

[Tmoose]

thanks all

Dan T

 

[henri2]

AWS A3.0 has a definition for seal weld. Duane Miller's paper on seal welds is pretty informative.

http://www.jflf.org/pdfs/papers/design_file6.pdf

 

[Guest102023]

In all seriousness, there's no such thing as a non-load-carrying weld.

 

[Duwe6]

brimstoner, in an engineering sense you are absolutely correct. In a Code sense, not so much.

For an ASME threaded connection [NPT or similar], the seal weld cannot be counted as contributing any strength to the joint. Officially, the strength member is the threads. Same thing under AISC for structural; on a bolted plus welded connection, you can only take credit for the bolts OR the weld(s). No combinations permitted.