Groove welds in Lap Joints 2

[little49]

I realize that fillet welds are most common in lap joints. However, due to certain restrictions I need to use a groove weld in a lap joint to acquire the desired strength. My question is if this is acceptable or if I need to alter my design in another way to accomodate the decreased weld strength. (Just for a little background, the weld is to be used in cross bracing for a curved girder bridge and the existing design is coming up short in fatigue. I am using MDX to analyze the structure and seem to be getting the same results as it is unfortunately. Any info would be greatly appreciated.)

 

[henri2]

After clicking the linkbelow have a look at Figure 7.6

http://www.millerwelds.com/education/TIGhandbook/pdf/TIGBook_Chpt7.pdf

Other groove welds which can be used in lap joints with plate include inter-alia: J-groove and square groove

This sort of joint will not be considered prequalified per AWS D1.1 and qualification testing will be required. I guess you are using D1.5 and I suspect it too may also require qualification testing.

 

[HgTX]

I'm guessing the lap joint is a bracing member to a gusset?

Keep in mind that a groove weld in that context will be a partial joint penetration weld, which is probably a big no-no for your application unless you can weld down the sides as well (as I assume you would have with the fillet welds) to keep the space along the lap from opening under reversal of stress.

Have you tried enlarging your gussets so you can use longer fillet welds?

Hg

Eng-Tips policies: faq731-376

 

[little49]

Thanks for the help. The weld would be treated similar to the fillet in the essence that it would continue down the sides just the same. The problem with enlarging the plates is that I am dealing with relatively tight clearance requirements. You are correct that it is the weld between the member and the gusset. I was hoping to be able to benefit from the increased strength of the groove weld without increasing my gusset too much. (If I increase it, then I have to increase the bolted conection at the girder, thus offsetting the center of gravity of the bolted connection from the path of the member force... and so on)

 

[HgTX]

Another possibility (don't take my structural word for it but it seems no worse than welding around the end of the member) is creating a slot in the member or gusset and welding fillets in the slot. (Note: this is not a slot weld. This is a form of lap.) Check your fatigue categories for the weld configuration. Watch out for slot
shape and edge finish.

Hg

Eng-Tips policies: faq731-376

 

[henri2]

HgTX,

Is the requirement in D1.5 prohibiting the use of groove welds as a solution similar to Section 2.17.1 in D1.1?

 

[HgTX]

I said what I said based on the picture shown for the groove prep for a lap, with nothing running down the side. The PJP groove weld is no worse than a fillet weld in the same situation--same susceptibility to prying apart at the root.

The difference between a PJP groove weld and a fillet weld can get a little subtle--and a little irrelevant, if I may say so.

However, a strict reading of the bridge code puts more restrictions on the PJP groove weld than on the fillet weld. 2.6 says "PJP groove welds shall not be used where the applied tensile stress is normal to the effective throat of the weld."

Section 2.10 has a lot to say about lap joints in general. They keep talking about fillets and never about PJP but the only difference between one and the other is the bevel prep. In a way, a PJP groove weld can be considered a fillet in a groove. Others will now jump all over me about how they're TOTALLY DIFFERENT AND HOW DARE YOU SAY OTHERWISE.

If it were my call, though, I wouldn't have a problem with beveling back the lapping member to get more weld in there, thus technically making it a groove rather than a fillet weld.

Okay, off to ask AWS about why PJPs would be barred where fillets are allowed...

Hg

Eng-Tips policies: faq731-376

 

[Ron]

HG brings up a good point....slots or plug welds will easily supplement the fillet

 

[little49]

Thanks for all of the response. I think I have gathered enough info to finish this thing.

 

[HgTX]

So whadja do?

Hg

Eng-Tips policies: faq731-376

 

[little49]

I basically sized up the connections to the gusset and the connection plate as little as possible to cover the increased stresses and increased the diagonal member thickness, so as to keep my same clearance. Reran everything and it checked out within a few percent of the allowable stress. Given the extents of my conservative analysis, I decided that the few percent were good enough and went with it. It has some beefy bolts (to the connectio plate) and some relatively heavy fillets (to the gusset) but I still have my clearance and I don't have to mess with groove welds and justification of them.

Thanks again for all of your help.