Truss Composite load transfer

[siddhams]

Dear All
Please Consider the below description for the structural system.
Modeling software ETABS
________________________________________________ Truss
| | | | | | | |
| | | | | | | |
| | | | | | | | Secondary beams
| | | | | | | |
| | | | | | | |
________________________________________________ Truss
| | | | | | | |
| | | | | | | |
| | | | | | | | Secondary beams
| | | | | | | |
| | | | | | | |
________________________________________________ Truss

When modeling the above system with the composite floor, We are getting 11 000 kN axial force in the central truss, and the same truss is experiencing 30 000 kN if the slab is modeled( equivalent forces are applied as UDL in secondary beams). This clearly shows 20000 kN needs to be transferred from the central truss to the composite slab. How to transfer such a huge force from the central truss to the slab? Is there any good book.

Regards
M.S.Boobathi

 

[BridgeSmith]

For highway bridges with composite steel girders, the shear transfer is typically accomplished by studs automatically end-welded to the girder flange. In your case, the compression chord of the truss would require a couple hundred studs extending into the slab. The AASHTO bridge design spec covers the design aspects for shear connectors, studs and other types. I don't know of any books specific to the subject.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[sticksandtriangles]

Why does the axial compression load change if the slab is modelled? Maybe I am not following the description. Did you not apply slab self weight loading to the secondary beams in option 1?
I would argue that if this load is just vertical and due to the self weight of the composite slab, that this delta of 20000kn does not to be transferred via the composite slab, it just works its way into the truss due to gravity.

Side note: one thing to be very careful with ETABs (and all structural software programs in general) is to ensure that your floor/diaphragm is not sucking up more axial load than you would want.

In ETABs, I will bring down the in-plane stiffness of the floor slab to something small and then use the section cutting tool to see how much load is in the frame elements vs the slab elements. If the frame element carries a majority of the axial load, then I call it good for gravity design. This matches typical hand calculations for determining forces in trusses.



S&T

 

[BridgeSmith]

Yeah, the selfweight of the slab is applied to the truss - no composite action for that. Only loads applied after the the slab has hardened are applied to the composite section.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[dik]

...and helps with LL deflections a lot, too.

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

-Dik

 

[dik]

I've not used programs for composite floors and have always used the perimeter members for diaphragm chord forces. Do the programs distribute compression forces to individual interior members?

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

-Dik

 

[sticksandtriangles]

Do the programs distribute compression forces to individual interior members?

For lateral analysis with a semi rigid diaphragm in ETABs (and I assume other programs that do semi rigid diaphragm analysis), the program will distribute compression forces (and tension forces) to individual members based on the stiffness of the whole floor and beam assembly. It can often get quite confusing when looking at results under a lateral load case if you have all the gravity system members included in your model, as each "gravity" beam will have an axial component due to the diaphragm wanting to bend between lateral force resisting systems.

I will often use the section cut tool to look at the total moment in the diaphragm and then do a traditional Axial Force = Moment / depth for steel members to determined the demand on chord members at the perimeter.

S&T