Gusset Plate Design 1

[chad73]

I am designing connections of a Special Truss Moment Frame. The top and bottom chords are continuous while the web members are connected by welding to the gusset plates. The joint is treated as rigid. As a result, the web members will carry axial force, shear force and moment. I can't find any book or anything on the net that tells me how to design the gusset plate of this type. Help me please.

 

[ToadJones]

...if this is the case I can see using a fixed joint for the "special" portion of the truss that is designed to dissipate seismic forces.
However, I still think the remaining portions of the truss should have joints that are pinned.

 

[connectegr]

A double angle member has very little moment capacity. Basically the bending capacity of the back to back legs. So the moment capacity of the angles can be developed in the welds to the gusset plate. But, I don't thick the larger moment capacity of channels can be transferred in the same way.

http://www.FerrellEngineering.com

 

[ToadJones]

Connect-
I think you are getting at the point I am trying to make (albeit poorly so). Even if modeling members like double angles as pins, they probably in reality do develop some secondary moment but since their moment capacity is so low it can be ignored in design.

 

[OHIOMatt]

I have only designed one of these in my career (hopefully in 25 years, I will still be able to say that). If memory serves, the the connections of the webs to the chords have to be sized to yield the web in tension + some factor that escapes me now as the intent of this truss is to absorb enery from seismic motion through the buckling of webs is the special segments. I ended up with gusset plates that where rather larger with a large amount weld. These connections ended up being extremely stiff. Since these are designed almost exclusively for the resistance of lateral seismic forces (gravity loads are limited to something like 5% of allowable stress), I can see were th OP is coming from.

Back to the original question. I believe that if you size the gussets as required by AISC 341, the gussets will end up being sufficiently stiff to be considered moment connections.

Chad73 you did mention above that you are using C shapes for all memmbers. AISC 341 requires that the webs in the special section be constructed of flat bars. If you follow this, there it the webs would likely not have much moment capacity.

 

[StrPE]

Chad73, moments in truss web members is very rare and should be avoided, however, there are times when fixed ends are required and you simply have to deal with the resulting moments. It is true gusset bending, especially weak-axis bending, will result in thick gusset plates, so if in-plane moments can not be released in your model, try like hell to release your out-of-plane moments which are producing torsion on your chords. Otherwise it's simply a matter of P/A + Mc/I. Your normal stress will add directly to your Mx and My stresses, and then your shear stress is applied. It's the old unity equation we've all seen since we were kids. The result will be if using LRFD(P/A + Mx/Zx + My/Zy)/.9Fy + (Hshear/A)/.6Fy <= 1.0.
Good luck.

 

[hokie66]

Sage advice from connectegr, as usual. The chords of trusses need to be designed for moments, but the webs should be strictly axial members. Except for rigid frames, often referred to as Vierendeel trusses.

 

[jberg]

Connectegr:

Agree with your assumptions about pinned connections for trusses and everyone else's comments about how to model the chords, but this discussion has raised another question that I have always struggled with. Would you still assume pinned connections if you are designing a HSS truss with all-around welded connections? I don't want to hijack this thread, but I thought it might be an appropriate question given the theme of this thread.

Thanks!

JWB

 

[connectegr]

Jberg
Yes. HSS welded panel point is a pinned connection. Similar to a bracing connection with gusset on top an bottom flange. The beam is still considered a simple beam. But ductility limits on endplate or angle thinks would not apply.

http://www.FerrellEngineering.com

 

[connectegr]

jberg,
My initial answer was brief, because I was typing on my cellphone.

IMHO
Welded HSS trusses with staggered member sizes have pinned panel point connections. As an example of staggered, I mean 10" chords, 8" verticals, and 6" diagonals. Unless the walls are extremely thick, the connection flexibility is found in the ductility of the HSS walls.

If the members are all the same width, so the welds are directly in alignment with the supporting HSS wall...
The connections have less flexibility. But, you cannot obtain the full strength of the walls with the welds. An AWS BTC-P10 (J-Grove PJP) has a maximum effective size of 5/8 x the wall thickness. And there is still ductility in the walls along the fillet welded sides.

This is another case, where the modeling software has no problem analyzing these rigid panel points. But, developing the welds and checking the HSS walls may be impractical. And may require backing up and changing the model or member sizes.

http://www.FerrellEngineering.com

 

[jberg]

Thanks connectegr:

That is exactly what I thought, but recently there has been a discussion at the office, debating whether truss connections in general behave as "pinned" or "fixed". I have always thought, in reality, truss connection behavior was somewhere between pinned and fixed, but everyone assumes pinned for the reasons previously stated. Once again, thanks for the prompt responses.

 

[OHIOMatt]

While I agree with most of the posts above regarding treating the joints as pinned for typical truss design, I must point out that the question was specifically for Special Truss Moment Frames. These are trusses that are designed for the resistance of lateral seismic forces and not for gravity loads.

When you look at the actual requirements for the design of the joints and web members in the special segment of the truss you will see that the gusset plates and connections will be large enough to transfer moment. Since the web members must be flat bars, they have very little bending capacity. Thus, for the truss specifically asked about by the OP, it is more accurate to model the joint as fixed.

Since the intent of the truss is to allow for significant elastic deformation in the special segment, moments created at the joint due to the webs should not be discounted.

 

[LSM76]

As has been stated, in practice truss joints are usually, and should be, assumed pinned both during the design of the main members and during the connection design. I have posted some thoughts regarding the design of trusses here:

http://larrymuir.building.officelive.com/Trusses.aspx

Near the end of the page there is a discussion of the consideration of moments at truss joints with the opinions of several experts listed.