Steel truss bottom chord in compression 5

[LJ_]

Hello,

I have this truss that has the bottom chord in compression. The length between the truss supports is 90ft and the section is HSS. The joints in the truss are every 8 ft. The trusses are spaced 20ft.

When I check for compression in the bottom chord, the length is 90ft so my HSS fails. Can I use gusset plates in the joints so I can provide for out of plane rigidity and thus reduce the length of analysis?. If the length was 8ft my member wouldn’t fail. Or is it necessary to provide bracing by connecting the trusses in the joints.

What do you think?

 

[JoshPlumSE]

The supports are welded to a base plate and then the base plate is anchored to a column. This is why I think it should be pinned pinned and not pinned roller

LOL What do others think? Has he done this analysis correctly?

If you do what I suggested (compare the results for pinned-pinned vs pinned-roller). If so, what do you get for deflection that would allow it to be considered pinned-roller vs reaction the support has to be able to resist for it to be considered pinned-pinned?

 

[LJ_]

What I actually did was to add columns as supports instead of a fully fixed pinned connection. This gives a result in between both conditions. But pinned roller is not okay in this case unless you design/detail for that.

 

[EZBuilding]

I agree with JoshPlum - pinned-roller.

 

[jayrod12]

How is the baseplate anchored to the column? Is that plat welded to the column, or are you using bolts between a column cap plate and the truss base plate? If bolts, I'd expect there to be some slip in that connection as well that wouldn't be captured by the model which would likely push it closer to true pin-pin.

 

[LJ_]

The truss is welded to a plate and the plate is anchored to the column. The anchoring is an embedded plate in the column and the anchors are welded to the bottom surface of the base plate.

 

[r13]

Model the truss frame with truss pinned to the column.

 

[LJ_]

Okay, even if it didn’t have that fixity which I also considered. The problem is the bottom chord compression under wind load not gravity load as a pinned roller support.

 

[KootK]

I revised the AISC Design Guide 24 and the hollow structural section book by Packer and the latter explains the procedure to calculate the K for the bottom chord given top chord and web elements and it also states that this K can be as low as 0.3 ir even less.

Can you direct me to where the Packer book says that? I have that reference and am curious to check that out.

 

[Settingsun]

KootK: Section 2.3.3 "Long laterally unsupported compression chords"

 

[r13]

For the special shaped truss ends, I am thinking it is possible to make a rigid truss-column joint by extending the truss end gusset plates on both faces down to the column. The rigid joints will reduce the unbraced length, and help for lateral drift.

 

[LJ_]

Thank you @Retired13 that’s actually what solved this. A solid rod connected to the hss and the rod to the plate that connects to the column. I don’t get compression in the bottom chord anymore due to wind. I also modeled the column instead of the pinned support.

 

[r13]

Glad that you got it resolved. Thanks for let us know.

 

[JoshPlumSE]

I have the strong suspicion that the results from this latest model will almost exactly match the pinned-roller model that I had suggested.

 

[BAretired]

I agree Josh. Hinge and roller supports are typically used in the analysis of a truss.

BA

 

[jayrod12]

And have been for long before computers were used for analysis. And yet we are still coming out with similar sizing and detailing even with the advent of these new fancy computer programs.

It's amazing

 

[r13]

If you design an pin-roller supports, you need to design a support that is free to move laterally. It is not likely/easily achieved in the beam-column connection though.

 

[BAretired]

Not true. You can safely neglect the stiffness of the columns as they have virtually no effect on the forces on truss members. Furthermore, you cannot rely on "fixed" bases for columns.

BA

 

[r13]

BA,

We did went though that exercise though other posts. You can model a simple frame to verify it.

 

[JoshPlumSE]

If you design an pin-roller supports, you need to design a support that is free to move laterally. It is not likely/easily achieved in the beam-column connection though.

No offense intended, but that's actually quite incorrect in reality. I know why you think it's correct. But, it's just due to the fact that you haven't seen cases like this nearly as often as I have.

Classic example is this case. Why did modeling in the columns solve his problem? Well the lateral stiffness of the column is relatively low. So, as soon as the guy models that, the behavior of the truss reverts to pinned-roller behavior that we expect. I spent 16 years in tech support for RISA. Every time someone called up with a truss modeled as pinned-pinned, they got funny answers that just couldn't be true (typically the bottom chord in compression for most load cases).

Telling them to move towards pinned-pinned to pinned-roller would solve the problem virtually every time. Some people (like you and the OP) would object. So, I'd tell them to do one of two things to prove them wrong:
1) Compare the lateral reaction (R) in the pinned-pinned model to the lateral deflection (delta_r) in the pinned-roller model. Typically they would have very large lateral reactions and extremely small deflections. So, much so that they immediately realized the rigid lateral restraint was not realistic.
2) If they still didn't accept my argument, I would tell them to model in whatever supports the truss on both ends. This would solve the problem, because the stiffness of their lateral support would have to be much greater than R/delta_r for the model to behave as pinned-pinned rather than pinned-roller.

The reality is that all situations are between pinned-roller and pinned-pinned. But, the real world result is virtually always much, much closer to the pinned-roller case.

 

[BAretired]

BA,

We did went though that exercise though other posts. You can model a simple frame to verify it.

I understand what you are saying, but we make certain simplifications when designing a truss. One such simplification is we assume all members are pinned at every node when in fact, the chords are usually continuous. The other simplification we make is that columns offer negligible lateral restraint to the truss, so it is usual to neglect column stiffness.

There may be circumstances where these assumptions are questionable and when that is the case, continuity should be properly considered, but for the vast majority of cases, the above simplifications are warranted.

BA