I think this joint is an interesting concept. I think the usefullness of the design may be that in general, without UT or RT, a full pen weld on tubular structures would be more likely to have defects than a fillet weld. Therefore, it could be used to increase assurance of a high quality joint without the extra cost of NDT. What do you think?

Regarding your specific questions:
1) The 1/4" filler plate will have greater strength in the direction of rolling than in the through thickness, but the through thickness strength may or may not be above the nominal strength required for the plate specification. The through thickness is not required to be tested, and I have never seen plates this thin that have been tested for that property. How would you check this anyway? The only thought I would have, is since hardness is directly related to tensile strength, if you cut a cube out of the material, would it have a different hardness in the 3 different planes?

2) I think this question could only be answered for sure with extensive engineering analysis. However, intuition says that if the volume of weld deposited is smaller, residual stresses could be smaller. The cross section of two 1/4" fillet welds is .0625 sq in, and the cross section of a groove weld using detail B-U2a-GF(AWS D1.1 pre-approved full pen joint) assuming 1/4" root opening, the cross section is .09375 sq in. Given this, one could estimat that the two fillet welds should not have greater residual stress than the groove weld. However, the residual stress will be in the through thickness direction of the 1/4" filler plate, and poor qulity filler plate could have lammelar tearing, but I do not know if it is common in such thin plate.

3) No I haven't.

Mr. Austin, after looking at your picture the question that comes to mind is why?? The only thing in my experience that comes close is an end plate with bolt holes. I once fabricated some specialty overhead cranes to sit on top of two very large Cincinnati press brakes and every connection had to have a weld plate installed on both pieces and the final connection was bolted. The idea was to be able to disassemble the entire aparatus when they outgrew their current building. I have eight employees in my shop, one of them has been welding longer than I have been alive, none of us have ever used or seen this type of connection. As for lamellar tearing, the cranes have been in service for many years now with no sign of any problems. Also, is lamellar tearing less of a concern with thinner pieces? I hope I have not wasted your time with this long winded observation.

Respectfully,
Mike Sherman
Shermans Welding

Perhaps the joint is always in compression so failure at a lamination would not be a factor. If it is wholly or partly in tension at least the plate is large compared to the area of the end of the tubing so the lamination would have to cover a fairly large portion of the plate (I would still worry).

In the case of bolted end plates well tightened bolts will keep the plates in compression (at least near the bolts).

Regards
Bill

Thanks to all who responded.

As I am not in the structural steel business, I am having a bit of difficulty understanding what I see in the picture. As such, I may be misinterpreting the whole situation. If this is the case, please help me right.

At any rate, I believe that possibly there could be two reasons for placing this "filler plate" in the joint.

1) It is very difficult to obtain a full penetration butt weld in this configuration. To resolve the problem, fillet welds are used that would give the same strength as is the case with a full penetration butt weld.
2) It mat be a way to turn the joint into a standard D1.1 joint prep so that additional procedure qualification tests are not required.

If the structural tubing had a great wall thickness, I would also have a problem with residual stresses in the through thickness direction in the plate. I do however believe that this tubing is rather thin. As such, it would probably just distort with very little resultant stress in the through thickness direction in the filler plate. In addition, because the area of the plate is so much greater than is the case with the tube, the stresses in the through thickness direction of the filler plate, from external loading, should be substantially lower than the stresses in the tube.

Regards
Niekie Jooste