seismic results generated from software 4

[wrxsti]

hello eng-tips gang hope everything is in good order and you've had wonderful holidays
happy new year to all

i am here again to ask for your input on this problem

i modeled up a steel structure in software(etabs) and was analyzing lateral loads on said structure and
found the some results to be as below

as you can see the chevrons are carrying about 40kips

but the beam sharing the joint with the chevron is carrying around 5kips

i was expecting to see something more like this image below with lateral building up until it reaches
the chevron and then some trig ratio to get to the chevron force

second example is a bit simplified i was just testing to see what would happen

anyways would like to hear your opinion on this matter

thanks

 

[JoshPlumSE]

Do you have a diaphragm that could be carrying the load into those braces?

This is especially common when people model "semi-rigid" diaphragms where you use plate elements. Typically, I would still design the drag / collector elements as if the force were dragged through them. But, the diaphragm (from an FEM/ stiffness standpoint) probably carries more of the load.

 

[wcfrobert]

You probably have a rigid diaphragm modeled on the roof level.

A rigid diaphragm enforces every node on a level to "move together" (zero in-plane axial deformation). I'm not sure how ETABS treats rigid diaphragms. Some software generates rigid many internal link members, while others perform kinematic condensation where the number of DOF in the structure actually reduces.

That said, since the ends of the collector beams move the same amount => no axial deformation => no force. This is what I would assume is happening.

However, I believe your intuition about what should happen is correct. You need to consider those axial forces in the beam when designing collectors.

 

[JoshPlumSE]

wcfrobert -

I think you're essentially correct. I would have said the same thing, except that he does have SOME axial force in his members. This should not be possible with a rigid diaphragm. That's why I thought it was likely to be a semi-rigid diaphragm instead.

 

[Agent666]

Agree with the others on the reason you are getting that distribution of loads. As far as the FEM model is concerned the horizontal component of the brace loads basically goes straight into the slab. In reality this is obviously unrealistic if for example you have distributed shear studs along the collector beam and potentially beams in adjacent bays that will transfer the seismic shear into the slab over some appropriate length.

Aspects like this require some judgement rather than full reliance on the model results to give you all the loads and load paths involved. You have potentially correctly assumed that for the load to get into the slab it has to go through the steel members to some degree. You might need to manually assess this load transfer.

 

[wrxsti]

thanks for the help guys

i turned off visibility on a membrane i had put on earlier in the model to get the slab load and forgot to turn off inplane stiffness
and was the culprit.

=(((

either way in reality this structure has a form deck(22ga. non composite) screwed onto support beams (12" CRS) with a concrete topping
no shear studs

with this configuration i don't think the slab would contribute, only the decking

what's your opinion on using diaphragm action with the decking or would you be conservative and ignore this?

 

[JoshPlumSE]

There is nothing wrong with using it during the analysis. But, I still make sure the beams were capable of resisting the larger axial load. And, I would make sure that may deck to beam details where sufficient for transferring the shear into the beam too.

 

[wrxsti]

@JoshPlumSE, @Agent666, @wcfrobert noted

thanks for the responses guys, appreciate it