Which cross section to use in analyzing a splice? 3

[LarryAndrews314]

Any assistance with this question is much appreciated. I've been talking with two Ca. Civil P.E.'s about this connection and we still don't have a definite answer. What we have is two 10' bed sections being spliced together. The span is 20' and the splice is at the center of the span. One P.E. used the cross section of the bed and angle combined. The other P.E. used the cross section of the continuous members through the splice which is the splice plate and 4"x4"x1/4" angle. Which is the correct cross section to use?

Attachment #1-calculations page showing bed section and angle cross section properties

Attachment #2-drawing illustrating splice and splice cross sections

 

[chicopee]

A splice where you have a maximum bending moment, a no-no on bridge construction. Splice s/b at points of inflection where bending moment is zero. But since this is going to China, I say go for it.

 

[hokie66]

The capacity through the splice, as shown, would only be the sum of the splice plates and angles. The capacity beyond the splice would likely be the composite capacity of the light gage conveyor bed and angles. Just splicing the verticals of the bed will not achieve equivalent capacity to the unspliced section.

 

[LarryAndrews314]

Yes, it's definitely the worse place for a splice and that notch in the splice plate couldn't be in a worse place. I'm in the rough spot of coming into this firm and being asked to review previously constructed work. This splice really caught my attention and the engineer who first deemed it acceptable used the entire cross section of the bed. Another engineer and I performed our own analysis considering just the splice plate and angle and it failed. Can I consider the bed as contributing anything? Do the four 3/8" bolts attaching the beds together make the top of the bed an effective compression flange? I'm really hoping someone with experience in structural splicing can offer input on this.

The analogy I use to visualize this is: The moment capacity of this splice is the summation of all the connections. There are three connections.
1) The angle
2) The splice plate
3) The Bed Section top flange

Imagining the splice with just one of each connection at a time the the first two can handle a moment but the third one cannot.

Ironically this is a current installation in the States but it's the Chinese who are requesting calculations before it goes up over there. No matter where it's going this is a people mover and the splice that needs to be acceptable.

 

[StructuralEd]

Your drawing notes "issued for assembly'.

 

[BAretired]

If the bed section top flanges are held in tight contact with each other by the bolted angles, they are capable of providing axial compression. That, together with tensile capacity from the angle and splice plate provides considerable moment resistance. The kink in the two top flanges does not help matters but there is still some axial capacity available if the top flanges are held in firm contact.

Even without the kink, it is not correct to use properties of the entire section. The 46" wide 10 gauge plate has an effective width considerably less than 46" according to the Canadian code for cold formed steel. I am not familiar with the U.S. code but I would expect that it has similar provisions.

The analogy I use to visualize this is: The moment capacity of this splice is the summation of all the connections. There are three connections.
1) The angle
2) The splice plate
3) The Bed Section top flange

I disagree with the above comment. Connection 3) is worth nothing in the simple summation of connections but it raises the position of the compression force, hence it increases moment capacity.

BA

 

[LarryAndrews314]

Thank you BA. I'll try finding something in the AISC about the effective width of the top flange. I imagine it's dependent on distance from the webbing. I'll try recalculating the moment of inertia as if it was a composite beam and put it up here for review.

 

[BAretired]

From Wikipedia:

United States Specification: North American Specification for the Design of Cold-Formed Steel Structural Members, document number AISI S100-2007 published by the American Iron and Steel Institute in October 2007. Building Code: IBC and/or NFPA may be enforced, but both reference AISI S100.

BA

 

[dhengr]

It seems to me that at the splice you have the following:
1. Two bottom reinforcing angles, 4x4x.25" which could have a higher Fy, and maybe be a huskier angle.
2. Two web splice plates which could be thicker and longer, and one section shows no top stiffener lip, while the other section and the iso view shows a top stiffener lip, which is it?
3. Two top reinforcing angles which have a bunch of funny notches cut out of the vert. leg, and should be continuous across the splice and be about 10-12' long, but they are not.

I would look at the two sides of this bed section as two light plate girders. The top and bottom angles form a couple to take the moment and are held apart/together by the splice plate and light 10ga. bed web. In its simplest form, a first calc. would be the two angles about 14.5-15" apart, as a couple, taking the full moment. What does that axial force have to be in the angles, and can the angles take that compressive and tensile force? Then, with the lengths and bolt arrangement you have can you get those forces back into the 10ga. bed section in a rational way. Then you can start finessing all of the sections acting together, as a composite unit. While I am sure the angle reinforced horiz. cross bed splice at the top will take some compression, that’s a tough one to call because of the kinked bend right at the splice. This would be a good place for some FEA.

The idea of the four reinforcing angles and the two web splice plates spanning the splice is to develop the loads and stressed back into the formed 10ga. bed section through shear flow and shear lag; and I’m not sure the bolting layout is very well thought out to do this. I would also want to give some thought to really tensioning the bolts to provide something akin to a slip critical splice joint, maybe with some thicker backing plates. Your description, so far, is pretty short on details, dimensions, loads, moments, shears, etc. which would help give the problem more perspective/definition. Is the 10ga. bed section formed in two pieced and spliced at the middle of the section? How is that continuous lengthwise connection made?