Edge Joint welds

[pbjansson]

Hello there!
I'm currently reviewing an edge weld (E70) that joints two 25-mm-thick plates (A36) with a maximum throat of 3 mm. Its size seems to be too small, if compared with the plates' thicknesses, however, since I'm an ordinary structural engineer, I only know the limitations to the fillet welds size from the AISC 360 code. There must be other regulation that covers this kind of connection. Could someone give me some guidance on this topic?

thanks in advance!

 

[dhengr]

Pbjansson:
A sketch would help explain what you’re looking at, we can’t see it from here. But, let’s see if I’m reading you correctly; you have two 1" (25mm, A36 steel) plates, a lap joint, welded with about an 1/8" throat fillet weld btwn. them. That’s about a 3/16" fillet weld, a fillet with about 3/16" (5mm) legs, the legs define the fillet weld size, right. That weld does sound very small to be load carrying weld on those 1" plates. Is it just a seal weld, what are the loads and their directions, etc.? Without digging them out, both AWS and AISC do have tables which give min. and max. fillet weld sizes as a function of the plate thicknesses being joined. The min. size has to do primarily with providing enough heat input during the welding process to get good penetration and so that the weld doesn’t cool so quickly as to crack. The thicker pls. act as too great a heat sink, on a very small weld. The max. size is usually a 1/16 or 1/8" (leg length, weld size) less than the thickness of the pl. edge being welded to. This is to prevent melting the sharp edge (corner) of the plate at the top of the vert. leg of the weld. Look this up in AWS and AISC, and see what their commentaries have to say on the matter. Take a look at Lincoln Electric’s web site, they have some very good books on welding, for example “The Procedure Handbook of Arc Welding.” For us Structural Engineers, the books by Omer W. Blodgett are particularly good; “Design of Welded Structures” and “Design of Weldments.” Don’t be an ordinary structural engineer, strive to keep learning more about what we do.

 

[pbjansson]

hi dhengr,

thank you very much for your answer.

sorry for not having posted a picture along with the question, that would have certainly helped! so, here it is:

http://files.engineering.com/getfile.aspx?folder=8eeadea2-a5b7-4e8f-ba41-21f174fbb7bc&file=weld.png

as you see, it is not a proper fillet weld, that's the reason to my doubts regarding the minimum weld size, since the aisc code is restricted to fillet welds...

Anyway, I'll take a look to Mr. Blodgett's books to watch for some additional guidance.

 

[dhengr]

Pbjansson:
The volume of that weld looks to be about (3mm)(15 or 18mm)(its length), a fairly significant waste of time and welding consumables. I agree, the throat on that weld is 3mm or maybe a little less than that, since some/many codes will allow only about 75-80% of the actual throat as the design or effective throat on that type of joint. But, once you get about 3 to 5mm away from the faying surface btwn. the two plates (the center of that joint), in either direction (up or down), the rest of the weld metal is wasted. For not much more work or effort, and considerably less welding effort, you could cut a bevel on the edges (corners) of one of the two plates for a small single bevel groove weld; or nibble or grind small bevels on both plates for a double bevel groove weld; and fill that groove, with good penetration at the root, a couple passes would fill it, and leave a slight convex contour on the surface. This would provide a much more effective weld for most load conditions. In trying to understand what’s going on here you must talk about the loads on the two plates, their magnitudes and directions. Are you just holding the plates together, with very little shear flow btwn. them at the faying surface? Then that weld is probably o.k. If there is a significant shear flow or if the plates are being pulled apart, putting that weld in tension, then you have a fairly ineffective weld. The weld you show is not a disallowed weld, but it is basically only good to hold the two plates together, a seal weld to keep water out of the faying surface, without much stress in the welds.

 

[racookpe1978]

It's a seal weld alright.

But, not much of a pressure barrier. It might seal against vacuum (at reasonable vaccuum pressures) but not much than that. IF the two plates were fastened to resist ALL weight and moment and movement stresses.

Physically, why so "high" (15-18 mm edge-to-edge ??) while only 3 mm in thickness? IF 3 mm "sealed" the gas/liquid inside the two 1 inch plates against leakage, then grind or machine a 2 mm 45 deg chamfer on each plate, and weld the resulting 90 deg bevel back convex. Don't bother grinding back to flush.

 

[Duwe6]

Most of the strength of your 'face weld' [preferred term] is in the penetration of the weld into the base metal. VERY hard to calc, thus not used for any structural purposes -- see racookpe comments - nor for pressure boundary stress-carrying items.

My review is NFG, non-structural only, will probably seal out rain and vacuum. Is merely decorative and cannot be used in any strength calc's.

Most of have been asked this question before; there are a bunch of misdesigned and/or misinstalled weldments in existing buildings. Reject it resoundingly, or walk away from this job. It is a sneaky attempt to get someone to sign off on bad, scary work, after the erector has driven off.

 

[Tmoose]

As several other said, what is the weld supposed to do?
If there are indeed strength requirements then changing one plate's overall dimensions 1/2 inch or so, or offseting one plate, will provide a real fillet with no additional plate prep.

http://www.tboake.com/SSEF1/CONNECTIONS/WELDED/weld-DSC_7825.jpg