Back to Back Channels 1

[Zivil]

Is there a standard procedure for analyzing a beam made of back to back channels with a non-continuous spacer plate between? I'm working backwards, I know the I-beam size for what I need, but want complete out of channels so I can place a tension rod between the channels. I have looked at the shear flow, but since the plates are not continuous I am unsure if I can do this. Also, would I have to look at the bolts for tension capacity? Any help would be appreciated.

 

[BAretired]

Two channels back to back form a doubly symmetric member similar to an I-beam. The properties Sx and Ix are simply the sum of the properties of the individual channels. Spacers must be close enough to prevent localized buckling of individual channels.

Shear flow is not an issue because of symmetry.

BA

 

[JAE]

BA - in terms of shear flow for weak axis bending (or lateral torsional buckling for strong axis bending) you do have some shear flow.

Two unconnected channels back to back have very different buckling characteristics (for strong axis bending) than when they are connected together.

Granted, the shear required to avoid LTB is small, but it does exist.

 

[BAretired]

JAE,

My earlier comments apply to double channels connected together. It was my understanding that the OP intends to connect them with non-continuous spacer plates.

I agree that shear flow exists, just as it does in an I-beam, but shear flow in the flanges of one channel counteracts shear flow in the flanges of the other so that the shear center and the centroid of the combined shape are in the same place.

I hadn't thought about lateral torsional buckling, but if the unbraced length is substantial, that would be a consideration.

BA

 

[weab]

The top flanges are going to want to separate and the bottom flanges are going to want to come together. I would make the spacer plates as deep as possible and vertically separate the 2 rows of bolts likewise.

 

[BAretired]

Zivil,

Perhaps, instead of 1 1/2" spacer plates, you could use 1 1/2 x 1 1/2 HSS spacers.

BA

 

[asixth]

If the channels share the vertical load, is LTB governing the load capacity of the system?

 

[Zivil]

Thank you all for the input, I appreciate it. I am having difficulty figuring the tension capacity needed to size the bolts. Any ideas?

 

[vandede427]

the bolts are in shear, not tension.

 

[Zivil]

I believe the bolts are in both when the spacer between the channels is non-continuous. Is this incorrect? As Weab mentions above, "The top flanges are going to want to separate and the bottom flanges are going to want to come together. I would make the spacer plates as deep as possible and vertically separate the 2 rows of bolts likewise."

 

[BAretired]

Without bolts, each channel would act as a separate beam, loaded eccentrically. The eccentricity, e is the horizontal distance from the point of application of load to the shear center of the channel.

If you provide bolts each side of the tension rod, then:

T = P*e/d

where T is the tension in the upper bolts each side of the tension rod (T/2 each).
P is the applied load in the tension rod.
e is the eccentricity
d is the vertical distance between bolts (perhaps a bit conservative)

In calculating e, the point of application of gravity load may be taken as the center of channel web.

Tension in the bolts is expected to be very small.



BA

 

[BAretired]

In the above, P is the load per channel, so it is only half the load in the tension rod.

BA

 

[Zivil]

Thank you all, I appreciate your help.