Interesting Lattice Tower Leg Connection

[Fontain]

Hi

We had a number of towers supplied to us about two years ago with the leg connection shown in the attachment at the point where the lower tapered part joins the non tapered part of the tower. As can be seen all the forces are transferred from the upper part to the lower part through the leg splices, two outer and two inner plates. With the towers supplied then, the contractor forgot to install the outer plates, which resulted in the inner plates buckling under the load. The situation was finally sorted by reinforcing the connection, after a number of uncomfortable discussions with the contractor. This past week we run into the same connection with a different client, and it has sparked a lot of debate. I must say it is different from what I usually see but argued that if the splice plates are thick enough,strength wise, it should be fine. The concern I had was to do with stability, I would be somewhat more relaxed if there was a horizontal somewhere in the joint to hold it all together, because as it is, any twist from the upper portion must be resisted and transmitted entirely by those splice plates without allowing the tower to deform in the horizontal plane at the joint. I believe having a horizontal and perhaps some diaphragm bracing would arrest such deformation thus making the joint more stable. But that's just my thinking, I thought it wise to seek an alternative opinion. So, what are your thoughts in regard to this connection?

Thanks

 

[skeletron]

Here is what I observe with the connection:
1. Braces on the top section are on inside/outside, whereas the lower section is outside/inside. Are these gussets separate plates or wrapped around the leg? Workpoints seem all over the place, but it's a bit of a parallax error.
2. The gap at the splice is a little unsettling. As you said, if the plate is thick enough it should theoretically work. But it seems like an inefficient and unstable transfer mechanism.
3. I agree on the horizontal bracing to stabilize the plate. Or even stiffen the plate at the section.
4. So the gusset is on the outside, the leg in the middle, and the splice plate on the inside?

 

[KootK]

1) I have no practical experience of this industry so factor that into my credibility score.

2) The connection looks very sketchy to me and I can see why it's drawn your attention.

3) I would love to see the calcs on one of these connections. Yeah, there's a path there for all the loads but they appear pretty complex and inefficient as skeleton mentioned. Even just regular truss shear would have to utilize the splice plate bolt group in bending.

4) X3 for the logic of having putting a diaphragm at this level. That intentional, horizontal gap is pretty large. I almost wonder if there wasn't supposed to be horizontal bracing there. Although, I can't really see what that would connect to.

 

[GC_Hopi]

In general agreement about the connection. Why wouldn't they use a angle instead of two splice plates? But the best solution would be to move the leg splice above or below the braces. That is the standard practice.

 

[KootK]

Looking at this closer, I feel as though the disconnected gusset plates are really going to crank some moment into the angle legs. And I don't think that any thickness of plate is going to address that unless somebody's up there doing slip critical bolted connections.

It hard to know with these things. Is it an error? Or a very carefully designed thing developed by someone with an eye for constructability? Probably the former. Usually the standard is the standard for a good reason and anything thought to be a better mouse trap... isn't. Long live innovation.