Curved Beam to Cantilever Beam 1

[Gile_]

Can anyone please advise on the connection and the choose of member shape of the below structure?
It is a roof with 5m high curtain wall below
I have AISC manual for curved beam. Just wondering what shape and connection would you choose for the cantilever beam?

 

[JLNJ]

Where you have torsion, use a tubular shape. Where you don't have torsion, use wide flange shapes. Be mindful of the connections since they are critical to the system's performance.

It looks to me that controlling deflection will be a challenge since you essentially have a double-cantilever type system.

 

[BAretired]

The simplest (and possibly most economical) solution is to add framing as shown below.

BA

 

[dik]

That's a tiny little tube with 10' cantileversl... What are the outside members? You may be able to eliminate the diagonal beam depending on the outside tube sizes... The detail at the cant, the curved beam, the horiz beam and the diagonal would drive a detailer to drink...What is the typical floor framing?

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[BAretired]

Just wondering what shape and connection would you choose for the cantilever beam?

Use a cap plate on top of the column. Run one wide flange beam through and splice the other, using a full strength moment connection.

BA

 

[Gile_]

Thank you everyone.

@BAretired. Is below the intention of the framing?

It is a roof with purlins running on top as shown (pink one) below

I just did a FEA with 1kN/m gravity load applied to the curved beam and surprisingly it looks like the open section cantilever beam will be twisting as well. (curved beam rotational pinned at ends but flexural pinned )

It gets better if change the cantilever beam to closed section. Thoughts?

 

[centondollar]

"It gets better if change the cantilever beam to closed section. Thoughts?"
Closed sections (RHS, CHS) have a significantly larger stiffness against pure torsion and warping torsion than do I-sections or channel sections.

 

[BAretired]

@BAretired. Is below the intention of the framing?

Yes to both questions.

EDIT: Dropping the beam is the easiest, but the beams could be flush on top if necessary, by making moment connections through the main beam.

BA

 

[BAretired]

It gets better if change the cantilever beam to closed section. Thoughts?

That is to be expected, but your FEA model ignores the stiffness of the beam beyond the cantilever. The curved beam imparts a moment which is not aligned with either the cantilever or the edge beam. The cantilever beam does not need to be designed for torsion because the edge beam will resist that moment more effectively than the torsional stiffness of the cantilever.

BA