hello ， everyone
I used 2D model to represent a full frame, for example, there is ten bays in the full frame, and the 2D lateral force-resisting frame is only four of the ten bays. How to model the P-Delta effect? which one is the correct in the following two methods?
1、 the 2D lateral force-resisting frame presenting full building holds the weight of only half of the building, and some leaning columns represent half of the rest of the building. The gravity load is applied to the nodes with the mass in the lateral force-resisting frame and leaning columns.
2、 There isn't leaning column in the model, the 2D lateral force-resisting frame presenting full building holds the whole weight of the building , and gravity load is applied to the beams, using Geometric Transformation Command "geomTransf ColTransfType PDelta" to present P-Δ effects.
can you tell me which metod is correct to consider the P-Δ effects, or tell me any another way to model the building considering the P-Δ effects.
Look forword to your response, thank you.
model second order effect with leaning column
Moderators: silvia, selimgunay, Moderators
Re: model second order effect with leaning column
doesn't look like either is correct, though 1 would be closest!
the gravity loads in the frame should only be those the frame will carry (otheriwse the axial forces in these members will be incorrect). leaning columns are provided for those bays to accout for mass and loads of those bays not accounted for.
the gravity loads in the frame should only be those the frame will carry (otheriwse the axial forces in these members will be incorrect). leaning columns are provided for those bays to accout for mass and loads of those bays not accounted for.
Re: model second order effect with leaning column
fmk wrote:
> doesn't look like either is correct, though 1 would be closest!
>
> the gravity loads in the frame should only be those the frame will carry
> (otheriwse the axial forces in these members will be incorrect). leaning
> columns are provided for those bays to accout for mass and loads of those
> bays not accounted for.
Thank you for your reply!
you mean that the gravity load of force-resisting frame is only applied on the force resisting frame, and all the other gravity load of rest of the building is applied on the leaning column?
another question: some models simulate response of structures under earthquake, it usually applies the gravity load on the node, not the beams, does it mean doesn't consider the effect of beams applied gravity during the earthquake?
> doesn't look like either is correct, though 1 would be closest!
>
> the gravity loads in the frame should only be those the frame will carry
> (otheriwse the axial forces in these members will be incorrect). leaning
> columns are provided for those bays to accout for mass and loads of those
> bays not accounted for.
Thank you for your reply!
you mean that the gravity load of force-resisting frame is only applied on the force resisting frame, and all the other gravity load of rest of the building is applied on the leaning column?
another question: some models simulate response of structures under earthquake, it usually applies the gravity load on the node, not the beams, does it mean doesn't consider the effect of beams applied gravity during the earthquake?
Re: model second order effect with leaning column
yes.
it's not one or the other, both are possible. it really depends on the structure and the load path. remember typically you just model the lateral system. if that frame is on the edge of a building, most loads might be coming to the column from the beams perpedindicular to the columns. If the frame was internal, the gravity loads could be coming from the beams in the system and the beams framing in from the perp. direction.
it's not one or the other, both are possible. it really depends on the structure and the load path. remember typically you just model the lateral system. if that frame is on the edge of a building, most loads might be coming to the column from the beams perpedindicular to the columns. If the frame was internal, the gravity loads could be coming from the beams in the system and the beams framing in from the perp. direction.
Re: model second order effect with leaning column
fmk wrote:
> yes.
>
> it's not one or the other, both are possible. it really depends on the
> structure and the load path. remember typically you just model the lateral
> system. if that frame is on the edge of a building, most loads might be
> coming to the column from the beams perpedindicular to the columns. If the
> frame was internal, the gravity loads could be coming from the beams in the
> system and the beams framing in from the perp. direction.
you mean that the load path of the gravity could change during the earthquake different from the static state, applying the gravity load on the beam-column node considers the change of load path in different frame? In the dynamic state, applying the gravity on nodes is more correct than on the beams? but I think the gravity load should be appled on the beams of the building whether in the static state or dynamic state, that feels more actual, isn't it?
> yes.
>
> it's not one or the other, both are possible. it really depends on the
> structure and the load path. remember typically you just model the lateral
> system. if that frame is on the edge of a building, most loads might be
> coming to the column from the beams perpedindicular to the columns. If the
> frame was internal, the gravity loads could be coming from the beams in the
> system and the beams framing in from the perp. direction.
you mean that the load path of the gravity could change during the earthquake different from the static state, applying the gravity load on the beam-column node considers the change of load path in different frame? In the dynamic state, applying the gravity on nodes is more correct than on the beams? but I think the gravity load should be appled on the beams of the building whether in the static state or dynamic state, that feels more actual, isn't it?
Re: model second order effect with leaning column
if elastic no, if nonlinear yes. just ask yourself what happens to gravity loads should a hinge develop at a beam end. i did not say more correct adding loads to the column nodes would be more correct, i said it depends on the structure being analyzed and ONLY YOU KNOW THAT!