Chevron brace connection with split gusset plates

[Nitesh Sadashiva]

Hello all,

Anyone did chevron brace connection with split gusset as shown in the attachment? Typically we use single gusset where both the braces will be connected and then we will calculate the forces at section a-a (gusset to beam interface) and section b-b (vertical section of center of gusset). However due to the slope of the brace, we need to split the gusset into two instead of using the single large gusset. In this case how we will calculate the forces and check the connection? Any design example or reference will be appreciated.

 

[human909]

Calculate them as two single gussets. Watch out for additional bending moment in the beam and additional LTB effects from loading the bottom flange.

Alternately just connected the two braces in once gusset. That is what I would do. The consequences of moving the work point down the beam and closer to the flange are general not of notable concern.

 

[Nitesh Sadashiva]

@Human909

Did you meant as shown below?

[URL unfurl="true"]https://res.cloudinary.com/engineering-com/image/upload/v1647948295/tips/Chevron_boyvnt.pdf[/url]

 

[human909]

I don't really understand that second PDF. And I don't haven't the full picture of the forces so I'm unsure how to guide you further. For example are both braces predominately carrying compression forces? Or it is a predominately lateral system where one is in tension and the other is in compression? Or it depends?

When I uses struts of significant size that join at a single node I normally have a single and substantial gusset.

Oh and regarding the additional bending moment I mentioned. I don't really think it is of concern here when you have a deep beam. I is more an issue that arises in columns where the bending capacity is significantly less.


I actually look forward to others chiming in. The goal of ensuring that all you member's axis meet at the same point sometime compromises your connection design.

 

[KootK]

I agree with human909 across the board, including the preference for non-concentric joints in some situations:

1) This can be done and has been done.

2) For the most part, it's two independent connection designs to be checked for all of the usual stuff that the single gusset connection would have been.

3) There are some nuances, in my opinion, that are a function of the tiny bit of vierendeel action that you can now expect across the node (shown below). That implies some local moments, as humun909 mentioned, as well as the need to take 100% of the nodal vertical shear across the beam which which is likely not a big deal unless this is a high seismic thing following capacity design principles. These tendencies also exist with a single gusset with the difference being that, in that instance, the gusset is there to help resist the shear and stiffen the section against local bending moments.

4) With regard to the force distribution, I'd recommend attempting free body diagrams of the constituent parts yourself and then posting them here for review. That method lowers the bar for participation on our part which will tend to garner you more help overall. You may luck out and have someone just draw the whole thing for you though, who knows.

 

[human909]

I agree with human909 across the board.

I almost fell off my chair when I read that!

I actually had started to second guess some of my first comments (about bending moment) being relevant and applicable in this case, so I figured I'd hedge my bets and ask the question. Your post also reflects and better elaborates on my views on the topic.

including the preference for non-concentric joints in some situations

I see some engineers struggle to suitably detail lateral braces due a deep seated desire for them to meet concentrically or the desire that they must attach to the column rather than the adjacent beam.

 

[KootK]

I actually had started to second guess some of my first comments (about bending moment) being relevant and applicable in this case, so I figured I'd hedge my bets and ask the question.

Yeah, you'd pretty much have to yield the beam web in shear before you'd develop any appreciable moments I think. And, even at that, it would still be a flexural stress distribution within the "node". With or without a split gusset, I feel like what's going on within the node is a bit of a crapshoot short of FEM modeling.

On another note, I've been thinking about "unideal" and have realized that I have my own version of it: medium-good.

 

[Deker]

Watch these two videos and you'll be an expert on all things related to bar gussets: Link #1, Link #2.