Design of Cranked Steel Beam 3

[VictorAK]

Hello,
Does anyone know where I can find the design of a cranked steel beam along with some examples?
Thank you.

 

[paddingtongreen]

If there are no fixities at the supports, it is simple statics to find moments and shears. Stabilities are determined by connecting members (I assume there will be members connecting at the top and bottom of the crank).

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[csd72]

Generally,

the bending moment does not change but the deflection will increase.

Be careful of overall stability at the crank as well as the strees transfer at the actual crank.

 

[ToadJones]

Are we referring to a beam that cut/ mitered and spliced to make a ridge?

 

[VictorAK]

ToadJones,
Yes - the beam is 31'-6" long with a 135 degree bend 8' from one end and a factored point load of 5,800 lbs. located 5' from the opposite end of the crank.

 

[slickdeals]

@csd72,
Can you explain why the deflection would increase? Won't a cranked beam peaking upward have less deflections? Would it not behave as an truss of some sorts?

We are Virginia Tech
Go HOKIES

 

[ToadJones]

I have actually done several of these within the last year or two. I guess I never gave them any special design consideration.
My intuition was that the deflections would be lower but I checked them using FEA.

I'd be interested in what ConnectEngr has to say about the splice location.

 

[ToadJones]

I guess another thing to watch is the amount of compression or tension you get in the member.

 

[nutte]

Blodgett discusses this type of connection, including local reinforcing required at the crank.

 

[ToadJones]

nutte-
You know where in Blodgett? (I'm being lazy)

 

[slickdeals]

Blodgett is a vast resource, it always suprises me how much good information is contained in a single book (and each one is so practical)

We are Virginia Tech
Go HOKIES

 

[nutte]

Starting on page 5.11-7 is the example I was thinking of, with the forces at the crank. This is actually for a rigid frame knee, but the concept would apply to the cranked beam.

 

[ToadJones]

Thanks nutte
Slick- Other than Moby Dick, Blodgett's Design of Welded Structures is the greatest book of all time.

 

[csd72]

You would only get significant axial load in this if there was restraint at the walls preventing spread.

If it is like most beams of this type then it is effectively on roller supports and therefore cannot have axial loads. You would therefore have the same bending moment as this is based on the distance between supports whereas deflection is more dictated by the actual length. You can verify this using first principles and force vectors.

As per the detail at the ridge, there has been several different controvertial discussions on this. My personal opinion if that you need a gusset plate between the flanges to equalise the out of plane force vectors in the flanges. I have been through this several times and will not be convinced otherwise.

If this is of the roller type then you also need to limit the horizontal spread of the rafter under vertical loads.

 

[ToadJones]

csd-
can you describe further what you mean about "a gusset plate between the flanges"?

 

[csd72]

A stiffener plate/ web plate/fin plate whic runs between the two flanges at the point where they change direction (i.e. the ridge)

 

[paddingtongreen]

Toad, I think csd means to use stiffeners as continuations of the cut flanges. I don't like that detail, I think it is over-restrained. I prefer to weld a piece of tee under the joint to deepen it. I agree with him that deflections are greater if the ends are not restrained horizontally.

I started off thinking that a "cranked beam" was actually a cranked beam, horizontal to vertical to horizontal again, a very different animal.

I don't have Kleinlogel anymore, but I found that I still have a copy of Crosby Lockwood's "The Steel Designer's Manual", this case is in there.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[csd72]

paddingtongreen,

No I mean a vertical stiffener at the point where you weld the two sections together. The problem with the two solutions you have suggested is that they do not restrain the top flange against buckling at the tips where it all of a sudden changes direction.

 

[ToadJones]

Well, I guess I was wondering why the stiffener would be necessary always.
I have done a few beams like this that were lightly loaded and due to the availability of materials the beams were way over capacity. In this case I did not use a stiffener where csd suggests, but I do like the idea.

 

[csd72]

Toadjones,

Yes I think that why this is not a common failure point is because most cranked beams are dictated by deflection and therefore over capacity for strength.

I always look at these things like a truss with the flanges/stiffeners able to take compression or tension and the webs only able to take diagonal tension. If you sketch that apex force diagram you will see what my issue is with it.